As firms and agencies urge their employees to work from home during the global pandemic, their employees’ confidential phone calls run the risk of being heard by Amazon.com Inc. and Google.
Mishcon de Reya LLP, the U.K. law firm that famously advised Princess Diana on her divorce and also does corporate law, issued advice to staff to mute or shut off listening devices like Amazon’s Alexa or Google’s voice assistant when they talk about client matters at home, according to a partner at the firm. It suggested not to have any of the devices near their work space at all.
Mishcon’s warning covers any kind of visual or voice enabled device, like Amazon and Google’s speakers. But video products such as Ring, which is also owned by Amazon, and even baby monitors and closed-circuit TVs, are also a concern, said Mishcon de Reya partner Joe Hancock, who also heads the firm’s cybersecurity efforts.
“Perhaps we’re being slightly paranoid but we need to have a lot of trust in these organizations and these devices,” Hancock said. “We’d rather not take those risks.”
The firm worries about the devices being compromised, less so with name-brand products like Alexa, but more so for a cheap knock-off devices, he added.
Like Wall Street, law firms are facing challenges trying to impose secure work-from-home arrangements for certain job functions. Critical documents, including those that might be privileged, need to be secured. Meanwhile in banking, some traders are being asked to work at alternative locations that banks keep on standby for disaster recovery instead of makeshift work-from-home stations to maintain confidentiality.
Smart speakers, already notorious for activating in error, making unintended purchases or sending snippets of audio to Amazon or Google, have become a new source of risk for businesses. As of last year, the U.S. installed base of smart speaker devices was 76 million units and growing, according to a Consumer Intelligence Research Partners report.
Amazon and Google say their devices are designed to record and store audio only after they detect a word to wake them up. The companies say such instances are rare, but recent testing by Northeastern University and Imperial College London found that the devices can activate inadvertently between 1.5 and 19 times a day.
Tech companies have been under fire for compromising users privacy by having teams of human auditors listen to conversations without consent to improve their AI algorithms. Google has since said that users have to opt-in to let the tech giant keep any voice recordings made by the device. Amazon now lets its users set up automatic deletion of recordings, and opt out of manual review.
The law firm’s warning first surfaced on an Instagram account “justthequant,” where people share their intel and struggles of working from home.
This story was originally published on Bloomberg.com. Read it here.
This blog is no longer active and is maintained for archival purposes. It served as a resource and platform for sharing insights into forensic multimedia and digital forensics. Whilst the content remains accessible for historical reference, please note that methods, tools, and perspectives may have evolved since publication. For my current thoughts, writings, and projects, visit AutSide.Substack.com. Thank you for visiting and exploring this archive.
Featured Post
Welcome to the Forensic Multimedia Analysis blog (formerly the Forensic Photoshop blog). With the latest developments in the analysis of m...
Showing posts sorted by date for query automatic. Sort by relevance Show all posts
Showing posts sorted by date for query automatic. Sort by relevance Show all posts
Wednesday, March 25, 2020
Tuesday, October 8, 2019
Mind the Communication Gap
I received a panicked email and phone call last week that can easily be summarized by the graphic above. Yes, there are still vendors selling to the police services that don't understand or accommodate the agencies needs or schedules.
As most in government service know, often the ability to spend money on tools and training happens within a short window of time. Such a window had opened for this person and they reached out to a company in response to a bit of marketing on new redaction functionality in a particular piece of software.
In the email that was shared with me, the requestor made it clear that they wanted to redact footage from body worn cameras. A trial code was shared for the company's tool. The requestor eagerly downloaded and installed the software, then assembled his supervisors and stakeholders to evaluate the new software. That's when the problems began.
Redacting the visual information was relatively straight forward, but cumbersome. However, he couldn't figure out how to redact the audio portion of the file. He reached out to the company's rep. Unfortunately, it was late on Friday in California. He received no reply from the company's rep, who'd likely gone home for the weekend. Thus, having previously communicated with me on technology issues, he reached out hoping that I'd still be around.
I let him know that to the best of my knowledge, none of Amped SRL's products redact audio. Yes, they'd promised audio redaction in FIVE at the 2016 LEVA conference. But, it's never materialized. As it's not been developed, it's quite likely that there is no audio redaction functionality in Replay.
He received a note from the sales rep on Monday. It informed him that the redaction functionality was built around a CCTV use case, and as such, did not concern itself with audio. The order taker that responded to the inquiry could have saved the requestor a ton of time and stress by simply reading the request - redaction of BWC footage - adding a helpful disclaimer in the reply that the tech doesn't redact audio.
Redaction remains a huge issue in California. Agencies are looking for a quick fix, but no easy solutions have presented themselves. The marketing from Amped seemed to provide a glimmer of hope to the requestor. But those hopes were dashed when the advice from the order taker, three days late, was that there is no single redaction solution - which isn't entirely true.
As I informed the requestor on Friday evening, the Adobe Creative Suite has all the necessary tools to perform a standards-compliant redaction for California's new laws. The automatic tracking in PremierePro is the best currently available. Plus, at $52.99/month, the cost savings is substantial vs. FIVE or Replay.
Given that the requestor had until 1700 PDT to spend the allocated funds, not receiving a reply until the following Monday was not an option. Thankfully, I was able to answer his questions and get him on the right path. Next stop for him is my redaction class, featuring the Adobe tools (link).
Have a great day friends.
As most in government service know, often the ability to spend money on tools and training happens within a short window of time. Such a window had opened for this person and they reached out to a company in response to a bit of marketing on new redaction functionality in a particular piece of software.
In the email that was shared with me, the requestor made it clear that they wanted to redact footage from body worn cameras. A trial code was shared for the company's tool. The requestor eagerly downloaded and installed the software, then assembled his supervisors and stakeholders to evaluate the new software. That's when the problems began.
Redacting the visual information was relatively straight forward, but cumbersome. However, he couldn't figure out how to redact the audio portion of the file. He reached out to the company's rep. Unfortunately, it was late on Friday in California. He received no reply from the company's rep, who'd likely gone home for the weekend. Thus, having previously communicated with me on technology issues, he reached out hoping that I'd still be around.
I let him know that to the best of my knowledge, none of Amped SRL's products redact audio. Yes, they'd promised audio redaction in FIVE at the 2016 LEVA conference. But, it's never materialized. As it's not been developed, it's quite likely that there is no audio redaction functionality in Replay.
He received a note from the sales rep on Monday. It informed him that the redaction functionality was built around a CCTV use case, and as such, did not concern itself with audio. The order taker that responded to the inquiry could have saved the requestor a ton of time and stress by simply reading the request - redaction of BWC footage - adding a helpful disclaimer in the reply that the tech doesn't redact audio.
Redaction remains a huge issue in California. Agencies are looking for a quick fix, but no easy solutions have presented themselves. The marketing from Amped seemed to provide a glimmer of hope to the requestor. But those hopes were dashed when the advice from the order taker, three days late, was that there is no single redaction solution - which isn't entirely true.
As I informed the requestor on Friday evening, the Adobe Creative Suite has all the necessary tools to perform a standards-compliant redaction for California's new laws. The automatic tracking in PremierePro is the best currently available. Plus, at $52.99/month, the cost savings is substantial vs. FIVE or Replay.
Given that the requestor had until 1700 PDT to spend the allocated funds, not receiving a reply until the following Monday was not an option. Thankfully, I was able to answer his questions and get him on the right path. Next stop for him is my redaction class, featuring the Adobe tools (link).
Have a great day friends.
Wednesday, August 21, 2019
Experimental Design
Before one begins any sort of research, one usually surveys the literature on the topic to see if any research has been completed and what, if anything, was concluded. Sure, the researcher has a general idea about what they want to study, but a literature review helps to inform and refine the eventual design of the study. According to Shields and Rangarajan (2013), there's a difference between the process of reviewing the literature and a finished work or product known as a literature review. The process of reviewing the literature is often ongoing and informs many aspects of the empirical research project. See what I just did, I discovered some research on literature reviews, and inserted the summary into my paragraph. Usually, there's an accompanying citation. Here it is: Shields, P., Rangarjan, N. (2013). A Playbook for Research Methods: Integrating Conceptual Frameworks and Project Management. Stillwater, Oklahoma: New Forums Press. ISBN 1-58107-247-3.
I received some feedback about the few posts I've written regarding "headlight spread pattern analysis." One was very intriguing - "... assume the premise is true, that there is uniqueness that can be discovered through experimentation. Where would you begin? What would the experimental design look like?" Hmmm....
Given the term, "headlight spread pattern analysis," there are four distinct elements - "headlamps," "the diffusion of light," "pattern matching," and a methodology for "analysis." Each of these would need to handled separately before adding the next element - the recording of this diffusion of light within a scene.
Let's just do a bit of research on the types of headlamps available to the general commercial market, leaving the other three elements for later.
Our first discovery is that we must separate the "lamp" from the "bulb." The bulb provides the "light" and the "lamp" is a system for the projection of that light.
For the lamp's housing, there are two general types: "reflector" and "projector." Of the light sources ("bulb"), there are several available types: Tungsten, Tungsten-halogen, High-intensity discharge (HID), LED, Laser. Of the "filament" type lamps, there are over 35 types available in for sale in the US, and covered by the World Forum for Harmonization of Vehicle Regulations (ECE Regulations), which develops and maintains international-consensus regulations on light sources acceptable for use in lamps on vehicles and trailers type-approved for use in countries that recognise the UN Regulations.
Given the eventual experimental design, it's important to note that the US and Canada "self-certify" compliance with the ECE Regulations. No prior verification is required by a governmental agency or authorised testing entity before the equipment can be imported, sold, installed, or used.
For a bulb's operation, there are variables to consider. There's voltage (usually 12V) and wattage (between 20w - 75w) - collectively known as "Nominal Power." Then there's "luminous flux." In photometry, luminous flux or luminous power is the measure of the perceived power of light. It differs from radiant flux, the measure of the total power of electromagnetic radiation (including infrared, ultraviolet, and visible light), in that luminous flux is adjusted to reflect the varying sensitivity of the human eye to different wavelengths of light.
Lots of big words there. But two stand out - luminous flux (the measure of the perceived power of light) and radiant flux (the measure of the total power of electromagnetic radiation). For our experiment, we'll need to differentiate between these as someone / some people are going to compare patterns (perception is subjective). We'll also need an objective measure of the total power of our samples. Luminous flux is used in the standard as the point of headlamps is to improve the drivers perception of the scene in front of them as they drive.
Luminous flux is measured in lumens. On our list of bulbs, the luminous flux values are reported as being between 800lm and 1750lm with a tolerance of between +/-10% and +/15%. This makes the range between 680lm and 2012.5lm. It's important to remember that the performance of a bulb over it's life span is not binary (e.g. 1550lm constantly until is stops working). Performance of lamps degrade over time.
Back to the lamp as a system. There are the general types of fixed lamps - they're bolted on to the front of the vehicle by at least four fasteners. These types need to be "aimed" at the time of installation, which can shift over time as the fasteners loosen. There are also "automatic" lamps, which feature some form of "beam aim control." These "beam aim control" types include, headlamp leveling systems, directional headlamps, advanced front-lighting system (AFS), automatic beam switching, Intelligent Light System, adaptive highbeam, and glare-free high beam and pixel light.
Now in the cases that I've reviewed, it seems that "headlight spread pattern analysis" was employed when a proper vehicle make / model determination failed due to a lack of available detail - usually due to a low nominal resolution.
Given what I've just shared above, about the potential variables in our study, an important revelation emerges. If there is insufficient nominal resolution to conduct a vehicle make / model determination, which considers class characteristics like presence / quantity of features like doors and windows before considering the presence / quantity / type of features within those items, then how could there be a determination as to type of lamp system and bulb that would be necessary for any comparison of headlight spread pattern? What if there's a general match of the shape of the pattern, but the quality of light is wrong? Or, what if the recorder's recording process and compression scheme corrupt the shape of the light dispersal or change the quality of the light? How then is a "scientific" comparison possible?
Short answer - it's not. This is one of the ways in which "forensics" (rhetoric) is used to mask a lack of science in "forensic science."
But, let's take a look at that question in a different way. Given all of the variables listed above, what would a normal distribution of "headlight spread patterns" "look like" (observed without recoding) for each of the possible combinations of system, bulb, and mounted position? What would they look like after being recorded on a DVR? This adds more variables to the equation.
For the recording system, there's the camera / lens combination, there's the transmission method, and there's the recorder's frame rate and compression scheme to contend with. Sure, you have the evidence item. But you don't know if the system was operating "normally" during that recording, or what "normal" even is until you produce a performance model of the system's operation in the recording of everything listed above. You'll need the "ground truth" of the recorder's capabilities in order to perform a proper experiment.
Remember, the recording may be "known" - meaning you retrieved it from the system and controlled it's custody such that the integrity of the file is not in question. But, what is unknown is the make / model of the vehicle. THIS CAN'T BE PRESUPPOSED. IT MUST BE DETERMINED.
In the cases that I reviewed, each comparison was performed against a presupposed make / model of vehicle - the "suspect's vehicle." If convenience samples were employed for a "comparison," then it was a few handy cars of the same make / model / year as the accused's vehicle. THIS IS NOT A VEHICLE DETERMINATION. This is no different than a single-person line-up, or a "show-up." This method has no relationship with science.
Back to the literature review and how it may inform a future experimental design.
What I've discovered is that the quantity of variables is quite large. Actually, the quantity of system types, then the variables within those systems, is quite large. This is before considering how these will be recorded by a given recorder. This information would be required to validate case work, aka a CASE STUDY. A case study is only applicable to that one case. If one wanted to validate the technique, then an appropriate amount of recorders would need to be included (proper samples of complete system types).
Given all of this, the cost of a single case study would be beyond the budget of most investigative agencies. It's certainly beyond my budget. The cost of testing the questions, "headlight spread pattern analysis has no validity" (H null) and "headlight spread pattern analysis has validity" (H1) would be massive.
Nevertheless, given all of the above, to conclude "match" - it is "the accused's vehicle" - one must rule out all other potential vehicles. Given that estimates put the number of cars and trucks in the United States at between 250-260 million vehicles for a country with 318 million people, then "match" says "to the exclusion of between 250-260 million vehicles" - which doesn't include the random Canadian or Mexican who drove their car / truck across the border to go shopping at Target. Because of this, "analysts" usually equivocate and use terms like "consistent with" or "can't include / exclude." Which, again, is rhetoric - not science.
I received some feedback about the few posts I've written regarding "headlight spread pattern analysis." One was very intriguing - "... assume the premise is true, that there is uniqueness that can be discovered through experimentation. Where would you begin? What would the experimental design look like?" Hmmm....
Given the term, "headlight spread pattern analysis," there are four distinct elements - "headlamps," "the diffusion of light," "pattern matching," and a methodology for "analysis." Each of these would need to handled separately before adding the next element - the recording of this diffusion of light within a scene.
Let's just do a bit of research on the types of headlamps available to the general commercial market, leaving the other three elements for later.
Our first discovery is that we must separate the "lamp" from the "bulb." The bulb provides the "light" and the "lamp" is a system for the projection of that light.
For the lamp's housing, there are two general types: "reflector" and "projector." Of the light sources ("bulb"), there are several available types: Tungsten, Tungsten-halogen, High-intensity discharge (HID), LED, Laser. Of the "filament" type lamps, there are over 35 types available in for sale in the US, and covered by the World Forum for Harmonization of Vehicle Regulations (ECE Regulations), which develops and maintains international-consensus regulations on light sources acceptable for use in lamps on vehicles and trailers type-approved for use in countries that recognise the UN Regulations.
Given the eventual experimental design, it's important to note that the US and Canada "self-certify" compliance with the ECE Regulations. No prior verification is required by a governmental agency or authorised testing entity before the equipment can be imported, sold, installed, or used.
For a bulb's operation, there are variables to consider. There's voltage (usually 12V) and wattage (between 20w - 75w) - collectively known as "Nominal Power." Then there's "luminous flux." In photometry, luminous flux or luminous power is the measure of the perceived power of light. It differs from radiant flux, the measure of the total power of electromagnetic radiation (including infrared, ultraviolet, and visible light), in that luminous flux is adjusted to reflect the varying sensitivity of the human eye to different wavelengths of light.
Lots of big words there. But two stand out - luminous flux (the measure of the perceived power of light) and radiant flux (the measure of the total power of electromagnetic radiation). For our experiment, we'll need to differentiate between these as someone / some people are going to compare patterns (perception is subjective). We'll also need an objective measure of the total power of our samples. Luminous flux is used in the standard as the point of headlamps is to improve the drivers perception of the scene in front of them as they drive.
Luminous flux is measured in lumens. On our list of bulbs, the luminous flux values are reported as being between 800lm and 1750lm with a tolerance of between +/-10% and +/15%. This makes the range between 680lm and 2012.5lm. It's important to remember that the performance of a bulb over it's life span is not binary (e.g. 1550lm constantly until is stops working). Performance of lamps degrade over time.
Back to the lamp as a system. There are the general types of fixed lamps - they're bolted on to the front of the vehicle by at least four fasteners. These types need to be "aimed" at the time of installation, which can shift over time as the fasteners loosen. There are also "automatic" lamps, which feature some form of "beam aim control." These "beam aim control" types include, headlamp leveling systems, directional headlamps, advanced front-lighting system (AFS), automatic beam switching, Intelligent Light System, adaptive highbeam, and glare-free high beam and pixel light.
Now in the cases that I've reviewed, it seems that "headlight spread pattern analysis" was employed when a proper vehicle make / model determination failed due to a lack of available detail - usually due to a low nominal resolution.
Given what I've just shared above, about the potential variables in our study, an important revelation emerges. If there is insufficient nominal resolution to conduct a vehicle make / model determination, which considers class characteristics like presence / quantity of features like doors and windows before considering the presence / quantity / type of features within those items, then how could there be a determination as to type of lamp system and bulb that would be necessary for any comparison of headlight spread pattern? What if there's a general match of the shape of the pattern, but the quality of light is wrong? Or, what if the recorder's recording process and compression scheme corrupt the shape of the light dispersal or change the quality of the light? How then is a "scientific" comparison possible?
Short answer - it's not. This is one of the ways in which "forensics" (rhetoric) is used to mask a lack of science in "forensic science."
But, let's take a look at that question in a different way. Given all of the variables listed above, what would a normal distribution of "headlight spread patterns" "look like" (observed without recoding) for each of the possible combinations of system, bulb, and mounted position? What would they look like after being recorded on a DVR? This adds more variables to the equation.
For the recording system, there's the camera / lens combination, there's the transmission method, and there's the recorder's frame rate and compression scheme to contend with. Sure, you have the evidence item. But you don't know if the system was operating "normally" during that recording, or what "normal" even is until you produce a performance model of the system's operation in the recording of everything listed above. You'll need the "ground truth" of the recorder's capabilities in order to perform a proper experiment.
Remember, the recording may be "known" - meaning you retrieved it from the system and controlled it's custody such that the integrity of the file is not in question. But, what is unknown is the make / model of the vehicle. THIS CAN'T BE PRESUPPOSED. IT MUST BE DETERMINED.
In the cases that I reviewed, each comparison was performed against a presupposed make / model of vehicle - the "suspect's vehicle." If convenience samples were employed for a "comparison," then it was a few handy cars of the same make / model / year as the accused's vehicle. THIS IS NOT A VEHICLE DETERMINATION. This is no different than a single-person line-up, or a "show-up." This method has no relationship with science.
Back to the literature review and how it may inform a future experimental design.
What I've discovered is that the quantity of variables is quite large. Actually, the quantity of system types, then the variables within those systems, is quite large. This is before considering how these will be recorded by a given recorder. This information would be required to validate case work, aka a CASE STUDY. A case study is only applicable to that one case. If one wanted to validate the technique, then an appropriate amount of recorders would need to be included (proper samples of complete system types).
Given all of this, the cost of a single case study would be beyond the budget of most investigative agencies. It's certainly beyond my budget. The cost of testing the questions, "headlight spread pattern analysis has no validity" (H null) and "headlight spread pattern analysis has validity" (H1) would be massive.
Nevertheless, given all of the above, to conclude "match" - it is "the accused's vehicle" - one must rule out all other potential vehicles. Given that estimates put the number of cars and trucks in the United States at between 250-260 million vehicles for a country with 318 million people, then "match" says "to the exclusion of between 250-260 million vehicles" - which doesn't include the random Canadian or Mexican who drove their car / truck across the border to go shopping at Target. Because of this, "analysts" usually equivocate and use terms like "consistent with" or "can't include / exclude." Which, again, is rhetoric - not science.
Thursday, May 25, 2017
FFT for video - yes, video
Eliminating repeating pattern noise from images has been a pain for forensic analysts for quite some time. There are tons of freeware apps and plug-ins ranging in price from inexpensive to cost prohibitive Some of their limitations include: they only work on images, their reliability is spotty, you must document your work without quite knowing how the tool is doing what it's doing.
Imagine having to account for the frequency spikes in each frame of video. This has been the barrier to using FFT tools to fix noise in video. Until now ...
Enter Amped FIVE (Axon Five*). FFT has been a part of the tool set for quite some time. Last year, we added Automatic Selection functionality. Auto Selection Mode, automatically identifies frequencies to remove without user intervention. What's the big deal? VIDEO!!!
Check out the video below to see FFT with Auto Selection Mode in action.
For the most part, it's a one-click fix.
As folks buy cheap CCTV systems meant for an indoor installation and place the cameras and cables outside (in the weather), the components degrade, corrode, and create all sorts of problems for the vide signal. With our tools, including FFT with Auto Selection, you can restore the video accurately. Then, when finished, you'll find that our reporting functionality has kept track of your activities. Just generate the report. The ease of use of our reporting tool is second to none.
*The Axon Forensic Suite tools are powered by Amped Software technologies.
Imagine having to account for the frequency spikes in each frame of video. This has been the barrier to using FFT tools to fix noise in video. Until now ...
Enter Amped FIVE (Axon Five*). FFT has been a part of the tool set for quite some time. Last year, we added Automatic Selection functionality. Auto Selection Mode, automatically identifies frequencies to remove without user intervention. What's the big deal? VIDEO!!!
Check out the video below to see FFT with Auto Selection Mode in action.
For the most part, it's a one-click fix.
As folks buy cheap CCTV systems meant for an indoor installation and place the cameras and cables outside (in the weather), the components degrade, corrode, and create all sorts of problems for the vide signal. With our tools, including FFT with Auto Selection, you can restore the video accurately. Then, when finished, you'll find that our reporting functionality has kept track of your activities. Just generate the report. The ease of use of our reporting tool is second to none.
*The Axon Forensic Suite tools are powered by Amped Software technologies.
Tuesday, August 5, 2014
First Responder debuts at NaTIA
One of the many vendors making a big splash at this year's NaTIA conference is MediaSolv. MediaSolv has teamed up with Amped Software to introduce it's First Responder product.
Here's how it works:
1. Retrieve the original video from the DVR (Flash/thumb drive, DVD, CD, etc.)
2. Because the format is proprietary, it cannot be played with Windows Media Player and Windows does not recognize the file.
3. Open First Responder.
4. When you import the video (drag & drop, or Video Loader filter), if First Responder does not recognize the format, you then have the option to convert the video to a standard format.
5. Convert the video, then you can play it in the First Responder Player.
6. If you aren't a MediaSolv Commander user, you can then export the video using the Video Writer Filter to a format standardized by your agency for play in Window Media Player or any other standard player.
7. If you are a Commander user, export to MediaSolv Commander along with other evidence items.
(Time start to finish ~10 minutes.)
First Responder is designed around Amped FIVE's innovative fast workflow and real-time filter concept to dramatically reduce the time required to process data and improves the success rate of various cases. First Responder will run on standard desktop or notebook computers and does not rely on third party commercial photo or video editing software, plug-ins, scripts, or special hardware. This makes the total cost of ownership much more manageable and is just one platform to learn, maintain, and deploy on hardware you already own.
Here's how it works:
1. Retrieve the original video from the DVR (Flash/thumb drive, DVD, CD, etc.)
2. Because the format is proprietary, it cannot be played with Windows Media Player and Windows does not recognize the file.
3. Open First Responder.
4. When you import the video (drag & drop, or Video Loader filter), if First Responder does not recognize the format, you then have the option to convert the video to a standard format.
5. Convert the video, then you can play it in the First Responder Player.
6. If you aren't a MediaSolv Commander user, you can then export the video using the Video Writer Filter to a format standardized by your agency for play in Window Media Player or any other standard player.
7. If you are a Commander user, export to MediaSolv Commander along with other evidence items.
(Time start to finish ~10 minutes.)
First Responder is designed around Amped FIVE's innovative fast workflow and real-time filter concept to dramatically reduce the time required to process data and improves the success rate of various cases. First Responder will run on standard desktop or notebook computers and does not rely on third party commercial photo or video editing software, plug-ins, scripts, or special hardware. This makes the total cost of ownership much more manageable and is just one platform to learn, maintain, and deploy on hardware you already own.
- Designed from top to bottom as a purpose built self-contained tool for video evidence ingest, viewing, fast editing, and conversion
- Automatic generation of technical report
- Support for images, videos and live streams
- Integrated lossless DVR capture tool
- Tools for converting proprietary files (transcode to proxy) for viewing on standard media players
- Native support for Milestone XProtect® surveillance live feeds and archived files
- Easy to use tools for fast/simple editing and evidence processing
- Unique concept of filters: Drop, add, delete, modify, move, copy, paste, any filter in any position. Modify any parameter of any operation in any order; the results can be applied and seen immediately, even while playing a video
So, if you're in San Diego for the NaTIA conference, stop by the MediaSolv booth and check out First Responder. If you can't make it, send them a note requesting more information.
Wednesday, July 16, 2014
Marketing vs. news you can use
This month, I received a renewal notice for Evidence Technology Magazine. It went straight into the trash. Why? It's not really about cutting edge technology from the standpoint of learning a technique that can add value to your workday - at least from a DME Forensics standpoint. It's a marketing device for manufacturers to get their tools shown in a positive light. I should know, I've written one of those articles (about Ocean Systems' ClearID).
This month's edition features an article titled "Advanced Video Forensics." I'm not sure what's so "advanced" about a tools that are fairly old. Consider the last update to the Omnivore was in May of last year, dTective's last update was more than 2 years ago, and the recent update to ClearID offers "No new features" (it just updates the installer to work with the new Creative Cloud versions of Photoshop. Sure, the Field Kit is "new." But, the tech driving the field kit is old - it's a laptop, an Omnivore, and a scan converter.
An interesting quote from the article, "It is estimated that video evidence is involved in approximately 80 percent of crimes. That staggering abundance of video brings some other complications—namely, the wide variety of video formats, each with its own proprietary characteristics and requirements. To be used, the files must be converted into a standard format that can be read and cataloged, then exported in a compressed format that will fit on a DVD for a courtroom. In the “bad old days,” that could translate into hours of work to parse formats, including some that required technical wizardry just to split different methods of encoding by different manufacturers ...", features a solution that is the absolute slowest solution to this problem - the Avid based dTective plugins.
Another is equally frustrating, "Union County’s four field kits can export instant video copies in file formats that can be played by anyone without needing proprietary equipment. These represent huge advantages for real-world use. The agency still retains the downloadable, native video files so they retain the original evidence, should it be needed." Is it faster to do a "real time" screen capture of the proprietary file with the Omnivore, or to simply work with the file's contents in FIVE or other ffmpeg solution? The real expense of the Field Kit is in the Bridge - the scan converter. It plugs nicely into the Omnivore, but it's still just a scan converter. If all you're doing is taking and working with proprietary digital files, FIVE works on low cost laptops. That means that your license of FIVE and a $1,000 laptop is still less expensive than the Field Kit - and FIVE gives you a ton more functionality than the Omnivore.
Which is a nice transition into this statement, "Union County’s new equipment also features an advanced video-editing platform and software plug-ins that allow technicians to visually focus and clarify an image. For example, they can filter and highlight a specific suspect or victim, magnify or enlarge objects such as an individual or a vehicle, and examine image areas down to individual pixels. There is even a module to remove “noise” such as darkness, rain, and snow. And, with the original video separated, the investigative tools leave the primary evidence untouched.
Union County chose its new system because of those advantages, as well as a highly comprehensive format. “It’s a real turnkey solution,” McCabe said. “It’s really comprehensive.”
Remember, clarification is not analysis. They spent a ton of money on an Avid NLE with some plug-ins that haven't been updated in years. FIVE gives them everything listed, plus gives them the option to do actual analysis - photogrammetry, content analysis, etc., with the report being an automatic function. FIVE is updated several times per year to address the rapid changes in technology and the law.
If this was a piece of journalism, you might expect a bit of counterpoint. There's none here. It's a marketing piece, pure and simple - and well done at that.
Given that many of the vendors in the FVA space have their own PR departments and send out e-mail spam on new products, updates, and etc., I'm going to save a small tree and skip the renewal. I'll get my marketing first-hand.
This month's edition features an article titled "Advanced Video Forensics." I'm not sure what's so "advanced" about a tools that are fairly old. Consider the last update to the Omnivore was in May of last year, dTective's last update was more than 2 years ago, and the recent update to ClearID offers "No new features" (it just updates the installer to work with the new Creative Cloud versions of Photoshop. Sure, the Field Kit is "new." But, the tech driving the field kit is old - it's a laptop, an Omnivore, and a scan converter.
An interesting quote from the article, "It is estimated that video evidence is involved in approximately 80 percent of crimes. That staggering abundance of video brings some other complications—namely, the wide variety of video formats, each with its own proprietary characteristics and requirements. To be used, the files must be converted into a standard format that can be read and cataloged, then exported in a compressed format that will fit on a DVD for a courtroom. In the “bad old days,” that could translate into hours of work to parse formats, including some that required technical wizardry just to split different methods of encoding by different manufacturers ...", features a solution that is the absolute slowest solution to this problem - the Avid based dTective plugins.
Another is equally frustrating, "Union County’s four field kits can export instant video copies in file formats that can be played by anyone without needing proprietary equipment. These represent huge advantages for real-world use. The agency still retains the downloadable, native video files so they retain the original evidence, should it be needed." Is it faster to do a "real time" screen capture of the proprietary file with the Omnivore, or to simply work with the file's contents in FIVE or other ffmpeg solution? The real expense of the Field Kit is in the Bridge - the scan converter. It plugs nicely into the Omnivore, but it's still just a scan converter. If all you're doing is taking and working with proprietary digital files, FIVE works on low cost laptops. That means that your license of FIVE and a $1,000 laptop is still less expensive than the Field Kit - and FIVE gives you a ton more functionality than the Omnivore.
Which is a nice transition into this statement, "Union County’s new equipment also features an advanced video-editing platform and software plug-ins that allow technicians to visually focus and clarify an image. For example, they can filter and highlight a specific suspect or victim, magnify or enlarge objects such as an individual or a vehicle, and examine image areas down to individual pixels. There is even a module to remove “noise” such as darkness, rain, and snow. And, with the original video separated, the investigative tools leave the primary evidence untouched.
Union County chose its new system because of those advantages, as well as a highly comprehensive format. “It’s a real turnkey solution,” McCabe said. “It’s really comprehensive.”
Remember, clarification is not analysis. They spent a ton of money on an Avid NLE with some plug-ins that haven't been updated in years. FIVE gives them everything listed, plus gives them the option to do actual analysis - photogrammetry, content analysis, etc., with the report being an automatic function. FIVE is updated several times per year to address the rapid changes in technology and the law.
If this was a piece of journalism, you might expect a bit of counterpoint. There's none here. It's a marketing piece, pure and simple - and well done at that.
Given that many of the vendors in the FVA space have their own PR departments and send out e-mail spam on new products, updates, and etc., I'm going to save a small tree and skip the renewal. I'll get my marketing first-hand.
Tuesday, June 10, 2014
Manual data carving - DVRs vs. Phones
A reader sent a note about how to explain the difference between computer forensics' ability to find deleted files on hard drives, but the relative inability of the team to recover files from a retrieved DVR hard drive. It seems that someone had retrieved a hard drive from one of those systems that will format the drive when you plug it back in, so they needed to retrieve the files without accessing the DVR's hardware (never mind the hardware decoding issues).
Popular mobile phone and computer forensic programs offer the ability to manually carve files of known types from the raw data. When folks delete text messages, images, and videos, forensic experts can often retrieve the files from the raw data dump of the device. This is largely due to the fact that common file types are coded in a certain way.
Because of the standards that are in place, we know that if we can find the JPEG's header (FF D8) and footer (FF D9) in the raw data, we can use our tools to extract / carve the image and save it out to a separate file. In this way, rarely is anything really deleted.
Also because of the standards, there are tools made specifically for carving multimedia files or for recovering multimedia files from hard drives or removable storage media - some are free, some are cheap, some are quite expensive.
The problem with applying this paradigm to DVRs is that coding, the header / footer for the proprietary file type is not generally published and is certainly not standard. If you're able to manually find and carve data from a Q-SEE DVR, the information gathered will not be of much use if you're trying to carve a raw dump from a Pelco DVR. Because of this high variability, the industry standard computer / mobile forensic tools aren't much help in automatic mode. It also means that it will likely take a considerable amount of time to decipher the encoding and begin the retrieval. In private practice, folks might not want to pay for that many hours of work. In public service, command staff might not have the patience required when waiting for results that might take a week or two to materialize.
But, if you have the time and the money to get into this type of work, there are a few training options out there. Your first stop should be with Jimmy and Jason at DME Forensics. They're offering classes on byte level analysis of DVRs. You also have the option to purchase their core product, DVR Examiner. I'd recommend that you do both if you're looking to get into this line of work.
Popular mobile phone and computer forensic programs offer the ability to manually carve files of known types from the raw data. When folks delete text messages, images, and videos, forensic experts can often retrieve the files from the raw data dump of the device. This is largely due to the fact that common file types are coded in a certain way.
Because of the standards that are in place, we know that if we can find the JPEG's header (FF D8) and footer (FF D9) in the raw data, we can use our tools to extract / carve the image and save it out to a separate file. In this way, rarely is anything really deleted.
Also because of the standards, there are tools made specifically for carving multimedia files or for recovering multimedia files from hard drives or removable storage media - some are free, some are cheap, some are quite expensive.
The problem with applying this paradigm to DVRs is that coding, the header / footer for the proprietary file type is not generally published and is certainly not standard. If you're able to manually find and carve data from a Q-SEE DVR, the information gathered will not be of much use if you're trying to carve a raw dump from a Pelco DVR. Because of this high variability, the industry standard computer / mobile forensic tools aren't much help in automatic mode. It also means that it will likely take a considerable amount of time to decipher the encoding and begin the retrieval. In private practice, folks might not want to pay for that many hours of work. In public service, command staff might not have the patience required when waiting for results that might take a week or two to materialize.
But, if you have the time and the money to get into this type of work, there are a few training options out there. Your first stop should be with Jimmy and Jason at DME Forensics. They're offering classes on byte level analysis of DVRs. You also have the option to purchase their core product, DVR Examiner. I'd recommend that you do both if you're looking to get into this line of work.
Friday, April 4, 2014
Quantitative, Qualitative, and Trial Support
We've discussed Quantitative vs. Qualitative before. But, a few readers are upset about the way it's been presented. So, I wanted to address their issues buy going over a few popular solutions from the standpoint of - what does the tool actually do, and is that "analysis?"
From Ocean Systems (italics are mine):
Each Complete dTective system provides the the following capabilities:
Process Digital (DVR) or Analog Video Evidence
DVR security video - decode digital video (DVR) files from proprietary security systems into uncompressed video [change of format is a trial support function]
Demultiplex video [qualitative improvement]
Enhance dark video and poor quality video security and surveillance video [qualitative improvement]
Print stills from Video to photo paper or digital files for fast distribution [trial support]
Archive to CD or DVD [trial support]
Highlight or obscure an important areas of forensic video tapes [qualitative / trial support]
Magnify video to display the result either full screen, or to a user defined portion on screen. [qualitative / trial support]
De-interlace field recorded forensic video to avoid blurry stills printed from video [qualitative improvement]
ClearID Forensic video and image enhancement tools [qualitative improvement - Photoshop plug-in]
Contrast and brightness adjustment [qualitative improvement]
Video stabilization [qualitative improvement]
Insert case notes linked to forensic video evidence [trial support]
Automatic project archiving [trial support]
Variable slow motion speed adjustment to compensate for time lapsed video [qualitative improvement]
Picture in Picture [trial support]
Titling and labeling [trial support]
Combine and stack tools [UI flexibility]
User definable layout and keyboard/interface short cuts [UI flexibility]
Not one of the above are quantitative in nature - counting, measuring, comparing, or authenticating.
Cognitech's Tri-Suite 2011 contains VideoInvestigator [qualitative improvements], VideoActive [qualitative improvements], and AutoMeasure [quantitative analysis - photogrammetry].
There may be an additional cost to using Cognitech software in that you'll need their Video Acquisition card. If your PC can't support it, then you'll need a new PC.
Amped Software's product selection includes FIVE [quantitative and qualitative tools in one package] and Authenticate [quantitative analysis - authentication]. Trial support functions are shifted to their friends at GP SIFT. The nice thing about this approach is that the quantitative / qualitative tools are mostly in one place, FIVE. If you don't do trial support, you won't need to spend extra on GP SIFT. Likewise if your not doing authentication, you won't need Authenticate.
So, there it is. You decide what's best for you.
From Ocean Systems (italics are mine):
Each Complete dTective system provides the the following capabilities:
Process Digital (DVR) or Analog Video Evidence
DVR security video - decode digital video (DVR) files from proprietary security systems into uncompressed video [change of format is a trial support function]
Demultiplex video [qualitative improvement]
Enhance dark video and poor quality video security and surveillance video [qualitative improvement]
Print stills from Video to photo paper or digital files for fast distribution [trial support]
Archive to CD or DVD [trial support]
Highlight or obscure an important areas of forensic video tapes [qualitative / trial support]
Magnify video to display the result either full screen, or to a user defined portion on screen. [qualitative / trial support]
De-interlace field recorded forensic video to avoid blurry stills printed from video [qualitative improvement]
ClearID Forensic video and image enhancement tools [qualitative improvement - Photoshop plug-in]
Contrast and brightness adjustment [qualitative improvement]
Video stabilization [qualitative improvement]
Insert case notes linked to forensic video evidence [trial support]
Automatic project archiving [trial support]
Variable slow motion speed adjustment to compensate for time lapsed video [qualitative improvement]
Picture in Picture [trial support]
Titling and labeling [trial support]
Combine and stack tools [UI flexibility]
User definable layout and keyboard/interface short cuts [UI flexibility]
Not one of the above are quantitative in nature - counting, measuring, comparing, or authenticating.
Cognitech's Tri-Suite 2011 contains VideoInvestigator [qualitative improvements], VideoActive [qualitative improvements], and AutoMeasure [quantitative analysis - photogrammetry].
There may be an additional cost to using Cognitech software in that you'll need their Video Acquisition card. If your PC can't support it, then you'll need a new PC.
Amped Software's product selection includes FIVE [quantitative and qualitative tools in one package] and Authenticate [quantitative analysis - authentication]. Trial support functions are shifted to their friends at GP SIFT. The nice thing about this approach is that the quantitative / qualitative tools are mostly in one place, FIVE. If you don't do trial support, you won't need to spend extra on GP SIFT. Likewise if your not doing authentication, you won't need Authenticate.
So, there it is. You decide what's best for you.
Friday, March 7, 2014
Send images from Amped FIVE to Word and PowerPoint
This just in from the folks at Amped Software:
"We just released an update to FIVE with a lot of improvements, bug corrections and customer requested features.
Despite FIVE having very advanced and automated reporting capabilities, it’s undeniable that most of us use very often Word, PowerPoint and other programs to show our results to courts, customers and colleagues. They could be simple investigative reports or some tutorials for explaining how to work with images and video.
I do it daily and I was sick of saving pictures and importing them in other various applications, so we added a very convenient way to interact with third party programs.
In the latest version of Amped FIVE (release 5989), if you right click on the Save Snapshot button in the Player Panel of FIVE you will have access to new settings for the Save Snapshot button.
These are the available options:
The last selection is kept in memory so the next time you simply click the button Save Snapshot it will be reapplied.
Imagine how useful it is the combination of these features: you can show and explain step by step in a document or presentation, all the operations done with FIVE."
It's amazing to have a company respond to suggestions by implementing changes in such a speedy fashion. For those agencies with pre-made report templates in MS Word, or who use MS PowerPoint extensively, this is a huge time saver.
"We just released an update to FIVE with a lot of improvements, bug corrections and customer requested features.
Despite FIVE having very advanced and automated reporting capabilities, it’s undeniable that most of us use very often Word, PowerPoint and other programs to show our results to courts, customers and colleagues. They could be simple investigative reports or some tutorials for explaining how to work with images and video.
I do it daily and I was sick of saving pictures and importing them in other various applications, so we added a very convenient way to interact with third party programs.
In the latest version of Amped FIVE (release 5989), if you right click on the Save Snapshot button in the Player Panel of FIVE you will have access to new settings for the Save Snapshot button.
These are the available options:
- Save Image (Automatic Filename): the displayed image is saved in the default path, along with the current project file needed to recreate the image from scratch.
- Save Image (Selected By The User): like above, but the user can select the filename and folder of the file.
- Send Image To Clipboard: the displayed image is copied to the system clipboard to be pasted in other applications.
- Send Image To Word: the displayed image is copied in the currently opened Word document.
- Send Image to PowerPoint: the displayed image is copied in the currently opened PowerPoint document.
- Current Image: saves the currently displayed image alone.
- Current Window: saves the full Amped FIVE interface.
- Full Screen: saves the whole screen.
The last selection is kept in memory so the next time you simply click the button Save Snapshot it will be reapplied.
Imagine how useful it is the combination of these features: you can show and explain step by step in a document or presentation, all the operations done with FIVE."
It's amazing to have a company respond to suggestions by implementing changes in such a speedy fashion. For those agencies with pre-made report templates in MS Word, or who use MS PowerPoint extensively, this is a huge time saver.
Thursday, December 26, 2013
Academic sources for authentication of digital images
For folks needing references for their authentication work, here's a short list (in no particular order):
- E. Kee, M.K. Johnson and H. Farid, "Digital image authentication from JPEG headers", IEEE Transactions on Information Forensics and Security, vol. 6, pp. 1066-1075, 2011.
- J. Kornblum, "Using JPEG quantization tables to identify imagery processed by software", Digital Investigation, vol. 5, pp. S21–S25, 2008.
- CCITT Recommendation T.81, ISO/IEC 10918-1:1994, "Information technology - Digital compression and coding of continuous-tone still images: Requirements and guidelines ", 1992.
- Z. Lin, J. He, X. Tang, Chi K. Tang "Fast, automatic and fine-grained tampered JPEG image detection via DCT coefficient analysis", Journal Pattern Recognition, Vol. 42, pp. 2492-2501, 2009.
- Alin C. Popescu and H. Farid, "Statistical tools for digital forensics", Lecture Notes in Computer Science, vol. 3200, pp 128-147, 2005.
- H. Farid, "Exposing digital forgeries from JPEG ghosts", IEEE Transactions on Information Forensics and Security, vol. 4, pp. 154-160, 2009.
- M.C. Stamm and K.J.R. Liu, "Forensic Detection of Image Tampering Using Intrinsic Statistical Fingerprints in Histograms", IEEE Transactions on Information Forensics and Security, vol. 5, pp. 492-506, 2010.
- J. Lukas, J. Fridrich and M. Goljan, "Digital Camera Identification from Sensor Noise ", IEEE Transactions on Information Security and Forensics, pp. 205-214, 2006.
- Mo Chen, J. Fridrich and M. Goljan , "Digital Imaging Sensor Identification (Further Study)", Proceedings. of SPIE Electronic Imaging, Security, Steganography and Watermarking of Multimedia Contents, pp. 0P-0Q, 2007.
- W. Wang, J. Dong and T. Tan, "Tampered Region Localization of Digital Color Images Based on JPEG Compression Noise", Proceedings of the 9th International Conference on Digital watermarking, pp. 120-133, 2010.
- R. Gonzalez and R. Woods, "Digital Image Processing (3rd ed.)", Prentice Hall, pp. 165–168, 2008.
- M. Kirchner and T. Gloe, "On Resampling Detection in Re-compressed Images". IEEE Workshop on Information Forensics and Security, pp. 21-25, 2009.
Tuesday, October 1, 2013
Does Photoshop CC Shake Reduction pass the test?
One of the new features in Photoshop CC is Shake Reduction. Thankfully, I'm now only getting one e-mail per blast from Adobe (thank you Adobe Marketing for straightening that issue out) per news release.
According to Adobe, you can "quickly restore sharpness to images blurred by camera motion, whether the blur was caused by a slow shutter speed or long focal length. With Camera Shake Reduction, you can make your shots steady — even if your hands aren’t." As always, I like to test new features with older files that I know have specific issues. To me, when I hear "Shake Reduction," I think PSF. So, lets see how well it performs.
First, I open my file and convert the image to a Smart Object (Layer>Smart Object>Convert to Smart Object). I do this as I may want to go back later and edit the settings, and this is the only way to do that. Simply choosing Filter>Sharpen>Shake Reduction, without first converting the file to a Smart Object, won't allow me to edit the settings later. But we already know about, and love, Smart Objects.
The Shake Reduction dialog box opens, and for my file, automatic produces a modern art masterpiece. It does not automatically detect and correct the PSF issues in my test file. Thus, I must perform a manual correction.
In the top left corner of the dialog box, you'll see four icons. We'll want the Blur Direction Tool (the second from the top). We use this to trace the point that has become a line (PSF). Here, it's a bit off its mark initially. But there are further controls available.
With these controls, I can visually dial in the PSF and correct the problem. I don't want smoothing as details are important. Artifact Suppression dials down the ripple effect of this process.
Once I'm satisfied with the results, I simply click OK and continue to work my workflow. As the PSF process corrects focus, it's done at the Focus step of my workflow. After the focus issues are addressed, I move on to global color and light issues.
That's the good news. Here's the bad.
If you're relying on Photoshop to track your edits and record your settings, you'll be disappointed. It doesn't record the settings for this filter.
There it is, from opening to saving. I have the Preferences for my History Log set to Detailed. Just to make sure that I wasn't doing anything wrong, I quit Photoshop, relaunched, and performed the test again. Again, the same results in the History Log. So, one more check - do Shake Reduction without converting the image to a Smart Object. Same results (data ...)
In terms of reproducibility, you'll need to record the settings yourself; Photoshop seemingly won't. Screen shots of the settings area might help you here.
So in the end, as a Photographer, I think it's a cool addition to the venerable program that I've loved since discovering version 3. As a forensic scientist, it's one more little annoyance - why can't the data be included in the History Log? Plus, as it's an Adobe product, I'll have to guess at the algorithm they used. But, if all you can afford is a CC subscription, at least you'll know now how to correct some PSF issues in Photoshop. (It did not do well when the pixel bubbled out in a circle, only when it blurred into a line. That's more of a Smart Sharpen (blind deconvolution) issue in Photoshop. But with that test, Photoshop fails miserably due to lack of precise control and refresh issues.)
So, I'll stick with FIVE. FIVE gives me the control I need. It's quicker to the punch in performing PSF corrections. It can handle, precisely, all my PSF needs. It reports all of the settings, gives me the plain English explanation of what I've done, as well as the reference for the algorithm in the report. And, as Photoshop is still a verb in California, I don't get all the baggage that comes with trying to explain how the image was "Photoshopped" correctly.
According to Adobe, you can "quickly restore sharpness to images blurred by camera motion, whether the blur was caused by a slow shutter speed or long focal length. With Camera Shake Reduction, you can make your shots steady — even if your hands aren’t." As always, I like to test new features with older files that I know have specific issues. To me, when I hear "Shake Reduction," I think PSF. So, lets see how well it performs.
First, I open my file and convert the image to a Smart Object (Layer>Smart Object>Convert to Smart Object). I do this as I may want to go back later and edit the settings, and this is the only way to do that. Simply choosing Filter>Sharpen>Shake Reduction, without first converting the file to a Smart Object, won't allow me to edit the settings later. But we already know about, and love, Smart Objects.
The Shake Reduction dialog box opens, and for my file, automatic produces a modern art masterpiece. It does not automatically detect and correct the PSF issues in my test file. Thus, I must perform a manual correction.
In the top left corner of the dialog box, you'll see four icons. We'll want the Blur Direction Tool (the second from the top). We use this to trace the point that has become a line (PSF). Here, it's a bit off its mark initially. But there are further controls available.
With these controls, I can visually dial in the PSF and correct the problem. I don't want smoothing as details are important. Artifact Suppression dials down the ripple effect of this process.
Once I'm satisfied with the results, I simply click OK and continue to work my workflow. As the PSF process corrects focus, it's done at the Focus step of my workflow. After the focus issues are addressed, I move on to global color and light issues.
That's the good news. Here's the bad.
If you're relying on Photoshop to track your edits and record your settings, you'll be disappointed. It doesn't record the settings for this filter.
There it is, from opening to saving. I have the Preferences for my History Log set to Detailed. Just to make sure that I wasn't doing anything wrong, I quit Photoshop, relaunched, and performed the test again. Again, the same results in the History Log. So, one more check - do Shake Reduction without converting the image to a Smart Object. Same results (data ...)
In terms of reproducibility, you'll need to record the settings yourself; Photoshop seemingly won't. Screen shots of the settings area might help you here.
So in the end, as a Photographer, I think it's a cool addition to the venerable program that I've loved since discovering version 3. As a forensic scientist, it's one more little annoyance - why can't the data be included in the History Log? Plus, as it's an Adobe product, I'll have to guess at the algorithm they used. But, if all you can afford is a CC subscription, at least you'll know now how to correct some PSF issues in Photoshop. (It did not do well when the pixel bubbled out in a circle, only when it blurred into a line. That's more of a Smart Sharpen (blind deconvolution) issue in Photoshop. But with that test, Photoshop fails miserably due to lack of precise control and refresh issues.)
So, I'll stick with FIVE. FIVE gives me the control I need. It's quicker to the punch in performing PSF corrections. It can handle, precisely, all my PSF needs. It reports all of the settings, gives me the plain English explanation of what I've done, as well as the reference for the algorithm in the report. And, as Photoshop is still a verb in California, I don't get all the baggage that comes with trying to explain how the image was "Photoshopped" correctly.
Tuesday, April 23, 2013
Blurring faces in Amped Five
Here's some good news from Amped Software about blurring faces in video: "Usually the main task put in front of the forensic video analyst is to process a video or image in order to make the details like a car, face, or other fine details more visible. Since cross-disciplined tasks are becoming much more common, we see that the request can also go in the opposite direction, and the expert is asked to hide sensitive details from the video. In order to preserve the privacy of people not involved in the case, it is often needed to mask faces or license plates from videos which will be shown in courts or given to the media.
Since having this type of capability can lend some additional utility to our customers, we included tools in Amped Five to achieve this result with the filter “Hide Selection”. Since just pixilation does not work for all applications, we have given the user the flexibility to selectively darken, blur or pixelate an area of the image. The area can be static, or can follow a subject of interest along the whole video or just parts of it; as this tool works with our other tools to provide the results needed.
The tracking of the area, person, or object can be done both automatically and manually. While the automatic approach is very effective on simple object such as license plate, often for people and faces it is necessary to use the manual approach which allows a better customization of tracking and ensures accurate results that won’t compromise the privacy or sensitivity of the obscured area.
Even in our manual mode, the process is very quick and effective, as the user just needs to set the area of interest only on some frames where the motion or the subjects has a higher rate of change. The software will automatically calculate intermediate positions on the other frames.
Click here for a quick video that show how easy is to accomplish the desired task of hiding the face of a suspicious subject from a surveillance video."
This is yet another reason to love Amped FIVE. How many more do you need?! :)
Since having this type of capability can lend some additional utility to our customers, we included tools in Amped Five to achieve this result with the filter “Hide Selection”. Since just pixilation does not work for all applications, we have given the user the flexibility to selectively darken, blur or pixelate an area of the image. The area can be static, or can follow a subject of interest along the whole video or just parts of it; as this tool works with our other tools to provide the results needed.
The tracking of the area, person, or object can be done both automatically and manually. While the automatic approach is very effective on simple object such as license plate, often for people and faces it is necessary to use the manual approach which allows a better customization of tracking and ensures accurate results that won’t compromise the privacy or sensitivity of the obscured area.
Even in our manual mode, the process is very quick and effective, as the user just needs to set the area of interest only on some frames where the motion or the subjects has a higher rate of change. The software will automatically calculate intermediate positions on the other frames.
Click here for a quick video that show how easy is to accomplish the desired task of hiding the face of a suspicious subject from a surveillance video."
This is yet another reason to love Amped FIVE. How many more do you need?! :)
Thursday, April 18, 2013
Photoshop Sneak Peek: See What's to Come
Adobe's John Nack shared this video about what's coming in a future version of Photoshop.
Based on the short video, it looks like they're attempting to do an automatic PSF correction to motion blurred images - also called deconvolution.
You know me by now, I'm not a huge fan of automatic corrections. From a repeatability standpoint, they're problematic. They're also problematic from a documentation standpoint. Thankfully, I've had the ability to do this for a while now - in AmpedFIVE. Here's a comparable video example - from 4 years ago - demonstrating the same functionality in FIVE. The nice thing about FIVE, is that you can do this on images or video - make precise, reliable, and repeatable adjustments - right now ... you don't need to wait on CS7.
Enjoy.
Based on the short video, it looks like they're attempting to do an automatic PSF correction to motion blurred images - also called deconvolution.
You know me by now, I'm not a huge fan of automatic corrections. From a repeatability standpoint, they're problematic. They're also problematic from a documentation standpoint. Thankfully, I've had the ability to do this for a while now - in AmpedFIVE. Here's a comparable video example - from 4 years ago - demonstrating the same functionality in FIVE. The nice thing about FIVE, is that you can do this on images or video - make precise, reliable, and repeatable adjustments - right now ... you don't need to wait on CS7.
Enjoy.
Thursday, November 15, 2012
History log function now in Amped FIVE
Yesterday, Amped Software announced a new update to FIVE. Along with some bug fixes, FIVE now has an automatic logging feature. You'll find this by clicking on View>Open Log Folder.
Since the new default is to log every action in FIVE, I just wanted to remind you of a post I wrote a few years ago about the discovery issue related to history logs.
... Smith and Bace make clear the fact that the reports we generate will provide fodder for cross examination. This includes the History Log. If your log is not handled correctly, it will likely include a running list of every image ever processed by you on your computer. Do you want to turn this list over for discovery? Imagine the questions it would generate. If your Log only shows work on a few images, but hundreds were turned in for discovery, you should expect a long series of questions ...
As the FIVE history logs record not only the actions, including play and pause, but also the local time, file locations, and etc ... you might want to run the use of this feature past your chain of command and your risk management section. In the mean time, you can turn this feature off in the Program Options window, if you want to.
Enjoy.
Since the new default is to log every action in FIVE, I just wanted to remind you of a post I wrote a few years ago about the discovery issue related to history logs.
... Smith and Bace make clear the fact that the reports we generate will provide fodder for cross examination. This includes the History Log. If your log is not handled correctly, it will likely include a running list of every image ever processed by you on your computer. Do you want to turn this list over for discovery? Imagine the questions it would generate. If your Log only shows work on a few images, but hundreds were turned in for discovery, you should expect a long series of questions ...
As the FIVE history logs record not only the actions, including play and pause, but also the local time, file locations, and etc ... you might want to run the use of this feature past your chain of command and your risk management section. In the mean time, you can turn this feature off in the Program Options window, if you want to.
Enjoy.
Wednesday, October 17, 2012
Meet Amped Software at the LEVA Conference
This just in from Amped Software:
We are happy to announce that we’ll be present at the Law Enforcement & Emergency Services Video Association (LEVA) 2012 Annual Training Conference on October 22-26, 2012 at the BAHIA Hotel in San Diego, California.
We will be demonstrating and showcasing the new 2012 version of our popular Amped Five Professional forensic video enhancement software. VideoScanner, our new product for forensic video e-discovery, will be shown in action at our booth. This is a new tool that can save countless hours for investigators looking for video evidence on a suspect’s computer.
Training conferences like LEVA are very important for us: with respect to other kind of events, here we have the possibility to be in touch with our actual end users and meet with world class forensic video analysts. Here we can present our technology to experts who can fully understand its potential and give concrete feedback for its further improvement. For this reason we are presenting here not only our popular software for forensic video analysis, Amped Five, but also our latest projects.
In particular with VideoScanner we created a very simple tool which can save hours looking and inspecting video files in digital forensic cases. VideoScanner allows the investigator to quickly extract few representative frames to all the videos found on a device for a efficient inspection. As a bonus it add several forensic facilities, such as the possibility to calculate the hash code on all the analyzed files and the automatic creation of the analysis report.
We are using this conference to show a preview of Authenticate, our new project which for digital photo authentication. Authenticate will incorporate an innovative workflow concept into a ground-breaking software for image authentication within a feature-packed, but easy to use, user experience. We see this as a critical tool for prosecutors and investigators in states such as California in light of the recent Beckley court ruling on requirements for digital evidence authentication. Evidence is being challenged in courts on the basis of authenticity and Authenticate will provide several tools for evaluating the originality of an image. We are offering several types of tools in Authenticate and have combined several different techniques, from simple metadata and quantization table analysis to actual pixel value inspection and cutting edge techniques used to specifically match an image to a camera or device.
At LEVA, Amped Software will be in booth 122 in the main exhibit hall. We have donated a VideoScanner tool for the LEVA Door Prize Drawing. You have to be present to win, so we'll see you there!
We are happy to announce that we’ll be present at the Law Enforcement & Emergency Services Video Association (LEVA) 2012 Annual Training Conference on October 22-26, 2012 at the BAHIA Hotel in San Diego, California.
We will be demonstrating and showcasing the new 2012 version of our popular Amped Five Professional forensic video enhancement software. VideoScanner, our new product for forensic video e-discovery, will be shown in action at our booth. This is a new tool that can save countless hours for investigators looking for video evidence on a suspect’s computer.
Training conferences like LEVA are very important for us: with respect to other kind of events, here we have the possibility to be in touch with our actual end users and meet with world class forensic video analysts. Here we can present our technology to experts who can fully understand its potential and give concrete feedback for its further improvement. For this reason we are presenting here not only our popular software for forensic video analysis, Amped Five, but also our latest projects.
In particular with VideoScanner we created a very simple tool which can save hours looking and inspecting video files in digital forensic cases. VideoScanner allows the investigator to quickly extract few representative frames to all the videos found on a device for a efficient inspection. As a bonus it add several forensic facilities, such as the possibility to calculate the hash code on all the analyzed files and the automatic creation of the analysis report.
We are using this conference to show a preview of Authenticate, our new project which for digital photo authentication. Authenticate will incorporate an innovative workflow concept into a ground-breaking software for image authentication within a feature-packed, but easy to use, user experience. We see this as a critical tool for prosecutors and investigators in states such as California in light of the recent Beckley court ruling on requirements for digital evidence authentication. Evidence is being challenged in courts on the basis of authenticity and Authenticate will provide several tools for evaluating the originality of an image. We are offering several types of tools in Authenticate and have combined several different techniques, from simple metadata and quantization table analysis to actual pixel value inspection and cutting edge techniques used to specifically match an image to a camera or device.
At LEVA, Amped Software will be in booth 122 in the main exhibit hall. We have donated a VideoScanner tool for the LEVA Door Prize Drawing. You have to be present to win, so we'll see you there!
Tuesday, June 12, 2012
The Legal Implications of BYOD
With so many of us using our personal phones and iPads for work (Bring Your Own Device), I thought that this article from the InfoLawGroup would be worth a read.
"... The first rule of policy development and compliance is to create policies that can be followed and will be followed when implemented. Policies should not be aspirational and should reflect the “reality on the ground” as closely as possible. If certain BYOD activities are already taking place, it may be necessary to develop policies that reflect those activities or terminate or limit certain activities on a going forward basis ..."
"... Overall, the policy and control set around BYOD should be made flexible enough to deal with unusual situations. This is especially important in the BYOD context because personal device technology and risks are constantly evolving at a very rapid pace ..."
"... Device configuration requirements may include enabling wiping/bricking capabilities of a device, disabling automatic back-up or cloud storage of data stored on a device, prohibiting the use of a personal device as a mobile hotspot, requiring or prohibiting certain application installation, enabling auto-patching for operating systems and applications, and prohibiting jail breaking or modding of devices. The policy will often also require employees to install mobile device management or other software on personal devices. These system, software and configuration requirements often vary depending on the particular type of personal device and/or operating system, and may need to be updated when new devices are being used or new security vulnerabilities discovered ..."
"... The first rule of policy development and compliance is to create policies that can be followed and will be followed when implemented. Policies should not be aspirational and should reflect the “reality on the ground” as closely as possible. If certain BYOD activities are already taking place, it may be necessary to develop policies that reflect those activities or terminate or limit certain activities on a going forward basis ..."
"... Overall, the policy and control set around BYOD should be made flexible enough to deal with unusual situations. This is especially important in the BYOD context because personal device technology and risks are constantly evolving at a very rapid pace ..."
"... Device configuration requirements may include enabling wiping/bricking capabilities of a device, disabling automatic back-up or cloud storage of data stored on a device, prohibiting the use of a personal device as a mobile hotspot, requiring or prohibiting certain application installation, enabling auto-patching for operating systems and applications, and prohibiting jail breaking or modding of devices. The policy will often also require employees to install mobile device management or other software on personal devices. These system, software and configuration requirements often vary depending on the particular type of personal device and/or operating system, and may need to be updated when new devices are being used or new security vulnerabilities discovered ..."
Monday, June 4, 2012
more to like about AmpedFive
The more I work with AmpedFive, the more I like it. One of the cool features, especially for those in a Frye state, is the reporting feature.
AmpedFive's automatic report generation feature includes the sequence of all operations applied to all files, the used parameters and a detailed description of their usage and implementation. It also has links to the relevant science behind the operations. How cool is that?
AmpedFive Professional is the perfect addition to your tool kit. It's purpose built for the forensic analysis of video and images. Check out a demo today. I think that you'll like it.
Enjoy.
AmpedFive's automatic report generation feature includes the sequence of all operations applied to all files, the used parameters and a detailed description of their usage and implementation. It also has links to the relevant science behind the operations. How cool is that?
AmpedFive Professional is the perfect addition to your tool kit. It's purpose built for the forensic analysis of video and images. Check out a demo today. I think that you'll like it.
Enjoy.
Tuesday, May 8, 2012
nitpicks with Bicubic Automatic
There's an interesting read on the various issues with the new Bicubic Automatic setting in Photoshop CS6 over in the Adobe Forums.
Enjoy.
Enjoy.
Monday, May 7, 2012
Bicubic Automatic
Cool. I've got the new version and I'm getting it set up and I see this:
I'm not a fan of anything "automatic." So, a little digging is in order. Here's what I found out.
This from Photographyuncapped.com: "By just having ‘Bicubic Automatic’ checked in the CS6 preferences, all Photoshop functions will default to Adobe’s best choice for the method. No muss. No fuss. Photoshop will decide the best way forward for you. If you want to override this setting on a case-by-case basis, you still can choose other methods in the Image>Image Size dialog under Resampling Method. And, of course, if you want to change this on a continuing basis, the Preferences can be changed at will."
I think I'll stay in control. I'm appreciative of Adobe's efforts to simplify my life ... but ... I'd rather choose the interpolation method.
Enjoy.
Thursday, March 29, 2012
Get a sneak peek at Adobe Audition CS6
One of the more difficult tasks in managing evidentiary multimedia is taking the audio from a body-worn recorder and merging it with various sources of covertly recorded video. Often times, the body-worn recorder produces the best audio, but the body-worn video misses the hand-off or the face shot. Thus, an investigator will want the better audio inserted into another video feed from the operation (pole cam, hand-held, etc.).
For this reason, I'm looking forward to the release of Audition CS6. According to Adobe, "Automatic Speech Alignment: Designed to align overdub dialogue quickly with existing production audio, the new automatic speech alignment feature does much more than merely align initial transients. When used on overdubbed dialogue, Adobe Audition analyzes the frequency spectrum involved as well as the waveforms, so each word of a dialogue clip is matched to its counterpart in the production audio—the real power is a new algorithm to provide alignment of noisy location audio to reduce mis-alignment. If you have ever worked with ADR or overdubbing, you know that this is a time consuming task and our goal was to provide a powerful tool to help make it a little easier. There’s no replacement for the skill and craft involved in professional ADR, but if you have an overdub performance which matches the location audio’s wording and phrasing, but the timing is off for lip sync, the new automatic speech alignment can help. Where you might normally perform a “stretch” on the audio to fit the timing, or be forced to do lots of editing/compositing to line up each word or phrase, automatic speech alignment can help you get good results quickly."
Yes, I know. This feature is designed for a production environment. Well, Photoshop is for photographers, and we love it in law enforcement. I think that the more LE folks get used to Audition, the more they'll love it. I can't wait to try this with body wire/body-worn matching for field operations.
Enjoy.
For this reason, I'm looking forward to the release of Audition CS6. According to Adobe, "Automatic Speech Alignment: Designed to align overdub dialogue quickly with existing production audio, the new automatic speech alignment feature does much more than merely align initial transients. When used on overdubbed dialogue, Adobe Audition analyzes the frequency spectrum involved as well as the waveforms, so each word of a dialogue clip is matched to its counterpart in the production audio—the real power is a new algorithm to provide alignment of noisy location audio to reduce mis-alignment. If you have ever worked with ADR or overdubbing, you know that this is a time consuming task and our goal was to provide a powerful tool to help make it a little easier. There’s no replacement for the skill and craft involved in professional ADR, but if you have an overdub performance which matches the location audio’s wording and phrasing, but the timing is off for lip sync, the new automatic speech alignment can help. Where you might normally perform a “stretch” on the audio to fit the timing, or be forced to do lots of editing/compositing to line up each word or phrase, automatic speech alignment can help you get good results quickly."
Yes, I know. This feature is designed for a production environment. Well, Photoshop is for photographers, and we love it in law enforcement. I think that the more LE folks get used to Audition, the more they'll love it. I can't wait to try this with body wire/body-worn matching for field operations.
Enjoy.
Subscribe to:
Posts (Atom)











