Featured Post

Welcome to the Forensic Multimedia Analysis blog (formerly the Forensic Photoshop blog). With the latest developments in the analysis of m...

Monday, September 30, 2013

A word about purpose built software solutions

This was posted over in a LinkedIn group, so I thought I'd share it with everyone. This comment is from the creator of Amped FIVE, Martino Jerian.

"I know it's very difficult to have a one-stop solution for everything, especially if you are proficient and know very well other tools.

When I started working on FIVE my vision was precisely that: a one-stop solution (exactly as written in our about page). I know that's a very ambitious target, but it is what is driving our development in the right direction. It is something you can never fully reach, but I think we are asymptotically going there. The thing I hate most is to use different software to solve one single task: converting, exporting, importing is a boring process, tedious and error-prone (unless you work completely via scripting unix-style, but takes a lot of time).

When working on cases I rarely use anything else than FIVE (mainly for format conversion ffmpeg and VirtualDub). If I have a new need, rather than look for third-party software, I create a new tool for FIVE, so that, sooner or later, our customer can benefit from it as well. Granted, that is easy for me, being the developer, but very often we try to do that for our customers too, it the request is feasible. Jim already told you that in a previous post. For what regards George Reis, I implemented several of his requests, but he's so demanding I have to start a new company just to keep up with his features request ;)"

Friday, September 27, 2013

Sinar rePro for Latent Print photography?

A few years ago, I was experimenting with a Hasselblad camera for surveillance photography. The results were stunning. I also examined the H4D for studio use. But that was then. Now comes the Sinar rePro and it gives me a whole new set of ideas - shooting medium format for in-lab latent print photography?

Back in'09, I was testing ACR for Latent Print Analysts and made the comment that if you had $80k to spend, try the Foster+Freeman DCS 4. Re-examining the Foster+Freeman offerings some years later shows that DCS 4 is still on offer. Thus, if I had the same $80k to spend, what would I buy?

I'd take a serious look at the Sinar rePro. "The legendary precision of Sinar view cameras for analog and digital studio photography is already well known and highly valued by a great number of users around the world. Based on the same standards, the Sinar rePro is specifically designed to meet the requirements of reproduction photography. On the Sinar rePro only the essential operating elements have been implemented, the resulting simplicity of operation leads to an optimal reliability and prevents handling errors. The Sinar rePro fits seamlessly in the universal Sinar System and it provides an easy workflow with maximum accuracy." Capturing for a latent print workflow is almost identical to a reproduction workflow in terms of precision / accuracy. This adds a touch more control and, with the right digital back, a huge increase in pixels (the eXact allows for a 192MP 16bit RGB output). SWGFAST has their recommendations in terms of minimums, but why settle for minimums if you've got $80k to spend?

Plus, "The basis for the accurate image sharpness is the perfect flatness of the object to be captured and the parallel alignment of the camera. A further novelty provides valuable assistance: Sinar parallel is an excellent device for precise adjustment. The battery powered Sinar parallel is placed on the capturing table and its laser beam reflects from a mirror on the camera. Precise alignment of the camera is achieved when the laser beam exactly hits the middle of the crosshairs of Sinar parallel." Precision alignment is a key to achieving great results.

So, if money was no object, and I wanted a very high megapixel result combined with benefits of a medium format reproduction camera set-up, then I'd really like to see the results achieved of putting the Sinar rePro together with a decent light source. Hmmm ... Christmas is coming.

Thursday, September 26, 2013

The perils of freeware

Over on LinkedIn, I recently posted a link to a post from Amped's blog in a group dedicated to audio and video forensics. One of the group's members, after accusing me of being an Amped employee (I'm not), went on to note that he's been doing everything that FIVE does for years, using freeware tools like Virtualdub and Pmod. So, I thought, can I work on .pqz files with Virtualdub? Here's what I found when I went to the main Virtualdub web site:

"It lacks the editing power of a general-purpose editor such as Adobe Premiere, but is streamlined for fast linear operations over video. It has batch-processing capabilities for processing large numbers of files and can be extended with third-party video filters. VirtualDub is mainly geared toward processing AVI files, although it can read (not write) MPEG-1 and also handle sets of BMP images." Thus, I'll need to either find or write filters to extend the capability of Virtualdub to work with proprietary file types like .pqz. So I clicked on the third-party video filters link, which displayed the following: "I've decided that I don't have the time or energy to hunt down and maintain a list of all the current video filters available for VirtualDub, so I've appointed Donald Graft as honorary filter website maintainer. You can find his own filters, and a comprehensive list of other third-party filters, at: http://neuron2.net/"(don't go there unless you want to be attacked).

There's lots of links to choose from in no particular order. But, I discovered a big problem on the A/V Links page (other tools and filters), the site tried to load a java script exploit to compromise my browser and gain access to my computer. Thankfully, I have a decent firewall and AV solution, so it blocked the attack.

Therein lies the problem for me. I'm not the greatest when it comes to scripting on my own, nor do I have the time to attempt the many different things I do with a scripting approach, so I'll have to find these scripts on the web. Like searching for codecs, searching for scripts from unknown sources is dangerous. If you're not familiar with scripting - you're launching a program or extending the functionality of a program (script) on your machine. Heck, just getting to the page with the list of scripts was problematic for me - the page automatically launched an attack on my computer.

Thus, I don't like the hassle - so I choose commercial software. As I need to be able to reproduce other experts' work, I tend to have a copy of just about everything from the horribly expensive (MATLAB) to freeware like ImageJ.

So just because its free, doesn't mean it's without cost. Make sure you're well protected before going off into freeware script land.

Wednesday, September 25, 2013

Surveillance Photography vs. Surveillance Videography

Evidence Technology Magazine is out and has a ton of good information, as usual. There was one article that caught my eye: Surveillance Videography Using DSLR Cameras. The author does an outstanding job at explaining the various surveillance scenarios and the best approaches/gear for each situation.

Unfortunately, the author left out one important detail. If you're a privateer doing surveillance in a public space using an off the shelf DSLR, you may unknowingly be committing a felony in many jurisdictions.

In California, Penal Code Section 632 is quite clear: "632. (a) Every person who, intentionally and without the consent of all parties to a confidential communication, by means of any electronic amplifying or recording device, eavesdrops upon or records the confidential communication, whether the communication is carried on among the parties in the presence of one another or by means of a telegraph, telephone, or other device, except a radio, shall be punished by a fine not exceeding two thousand five hundred dollars ($2,500), or imprisonment in the county jail not exceeding one year, or in the state prison, or by both that fine and imprisonment." "632 (d) Except as proof in an action or prosecution for violation of this section, no evidence obtained as a result of eavesdropping upon or recording a confidential communication in violation of this section shall be admissible in any judicial, administrative, legislative, or other proceeding."

Penal Code Section 633 is what gives the law enforcement official in California their permission to conduct audio surveillance, and an exemption to PC 632.

Other jurisdictions have similar language, as I've noted previously.

Remember, if you aren't a law enforcement official conducting an investigation and subject to your agency's policies, and you drop an open mic into a public space without consent of everyone within reach of the mic ... then you're asking for big trouble. In some areas, just using your iPhone to record you and your friends at the local pub can get you into trouble.

But, if you're a law enforcement surveillance specialist, the article is great from a gear buying standpoint.


Calculate camera position from a picture or video

This just in from the folks at Amped Software: "In the latest version of Five, we have added several improvements, like the possibility to select the layers of calibration lines to show on the picture, and the possibility to draw free hand lines (not constrained by the perspective) to help with the estimation of ground floor points.

In the just released update, we added another useful feature to the Measure 3d filter, which is the ability to automatically calculate the camera position with respect to the ground plane.

Thanks to the calibration of the perspective, once you set the reference measure, you’ll know automatically the height of the center of the camera optics from the ground.

You can simple read the value from the Camera Position parameter in the Reference tab."

It just keeps getting better.

Tuesday, September 24, 2013

Work with only I-frames in FIVE

Some adjustments work best if you can isolate the I-frames in a video. Sometimes, you simply want a visual cue as to what type of frame is being displayed. Now, a recently added feature in FIVE helps solve that problem.

In the Player Panel, frame type (I,P,B) is displayed after the frame number, if the info is available. And, better still, you can select only the I-frames with which to work.

From the Select Filter menu, click on IFrames Selector. Wait a few seconds as FIVE scours the file and returns the I-frames for you.

So, not only can you select only the I-frames, the filter's menu will display the frame numbers of the selected frames. Better still, the selected frame numbers are entered in the report automatically. How cool is that?!

Monday, September 23, 2013

Converting containers

The more people see Amped FIVE, the more people share with me what they think is a huge selling point - easy conversion of proprietary file containers. For some analysts, this simple task represents 70% their daily work.

With just a few clicks, proprietary containers can be converted and the files opened for use in FIVE. If all you need to do is convert the file for the investigator, then just write the file out to your chosen format ... and you're done. FIVE offers an excellent level of control that's easy to understand.

Tuesday, September 17, 2013

Forensic Video's Detroit Moment

I doubt that anyone disputes Detroit's significance in the history of the automobile. In the first half of the 20th century, Detroit continued to grow on the back of the powerful auto industry. The US car industry could do no wrong ... until the 1960's.

In the 60's and 70's, the world changed. The world wanted a different kind of car. Companies from Japan and Germany reacted to the change and thrived. US companies scoffed at the change, and are now largely gone. In 1950, you would have been laughed out of the room if you suggested that VW or Toyota would outsell GM worldwide. Today, it's a reality. Yesterday's auto industry wasn't nimble enough to react to world trends. I think about the 65' Chevy that I rode around in as a kid, Dad proudly at the wheel. Today, at 6'7", I can't fit in comfortably in a Chevy. Amazingly, my Honda Fit has more room and better fuel economy than its American equivalent.

But, this is an FVA blog. So, what does this have to do with video and images. A ton, actually. I see the forensic video industry as the US in 1970. The world is changing. But, the big players aren't nimble, they aren't reacting to the changes in the law and the world. Adobe and Avid products demand bigger GPUs and subscription pricing in the face of shrinking agency budgets. Both target the larger photographer and videographer market, leaving the law enforcement side to struggle to afford  the new products and to fit them into their workflow - big expensive square pegs for shrinking round holes.

Then, like German and Japanese auto makers of the 60's, along comes small companies like Amped Software, GP Sift Technologies, DME Forensics, and so forth, offering purpose built solutions directly to the LE market.

I've been asking Adobe to include FFT in Photoshop for over a decade - it's still not there. FFT was in the first release of Amped FIVE. We've been wondering at how to work with native files effectively. FIVE and GP Sift have addressed it in a number of helpful ways. Amped and GP SIFT are Milestone partners, why aren't Avid and Adobe? With Milestone providing the technological backbone to almost 100% of LE's camera systems (Axis & Sony based pole cams, CCTV, etc), why wouldn't the big boys want to partner with them?

So, when the US car industry stopped making cars that supported me and my needs, I shopped around and found Honda. I'm wondering if the big boys of the FVA industry see what's going on. I'm wondering if they and their partners can react in time. Judging on what happened at LEVA, I'm not holding out hope.

Monday, September 16, 2013

Say goodbye to your NLE

The job of a video analyst has evolved. I mentioned in previous posts that the game has changed. Here's another bombshell. In the future, you won't need an NLE. And, guess what ... the future is now. Say good bye to your NLE.

Many of the folks that first started working with video and images in law enforcement came from a production background. They were well versed in broadcast video technology. At the time, it made sense to structure a workflow around products from companies like Avid, Adobe, and Apple. After all, what else was there? Over time, new features were added to your favourite NLE that could be used to assist with case management and courtroom presentation. But in the end, NLEs were expensive and offered a steep learning curve for folks in law enforcement without a background in media/production.

For me, I like to have it all. But, that being said, I work for clients and court systems that don't have everything that I have. I have to be mindful of discovery / repeatability issues when I begin casework. If I use Avid software, and the opposing expert doesn't, how will they be able to view my bin and verify my work. Adobe products have free trial versions,  but the same issues will hold true. The additional wrinkle for me ... Photoshop is a verb. Traditional production oriented tools like Avid's MC and the Adobe CC line are generally not used in my workflow anymore. So, how might I go about creating my demonstratives for court and sharing them with my clients.

Meet GP SIFT. If you haven't heard of them, you might start by reading this white paper or watching this video. SIFT Technologies offer solutions to a number of my current problems, from live capture of interview room and pole cam feeds (Milestone based), to evidence management, to their FlexView product - digital evidence viewer and presentation creation.

The Key Features of their FlexView product:
• Open format platform handles virtually any digital evidence
• Does not require an expensive server
• Does not alter the original digital evidence unless directed by user
• Creates and converts video and picture evidence to multiple industry standard formats
• Video trimmer to easily distribute clips of video evidence
• No Single Point of Failure with real-time automated backup

When I first heard of this company and their product, I knew what a stir they/it would create. It runs on just about any Windows PC. It's reasonably priced. It offers tons of annotation/telestration features. You can trim audio/video files whilst protecting the integrity of the original (works on a proxy). Can be easily shared and protected.

Whilst they were a vendor at the recent LEVA conference, I'm not sure if folks knew what they were seeing. I did, immediately. SIFT offers an alternative to recurring payments and expensive service contracts. It offers a purpose built software for our unique needs. It offers a ton of flexibility. And, as a Milestone partner, it offers a way to work with all the native Milestone video that I get (facility and pole cams).

So, if your agency has Axis based pole cams, if your stations have a Milestone based CCTV system, if your corporate clients are standardized on Milestone (about 70% of the large scale facility market), or if you're looking for a cost effective alternative to the production oriented NLE, give SIFT a look. I know that you'll be impressed.

Friday, September 13, 2013

We now resume our regularly scheduled broadcast

I went to the doctor today to get the stitches out of my wrist. I'll spend the weekend getting my hand used to fine motor skills again. The blog posting will begin on Monday. Boy, do I have a lot to share with you. A lot has happened since I last posted.

Tuesday, September 3, 2013

LEVA class preview

Ok. Surgery was a week ago. Still typing one handed. Pain sucks, but it's nothing that the water of life can't handle. Should make for an interesting class as the bandages don't come off until 3 days after the class is over.

Folks are wondering what's on tap for my Image Analysis class at LEVA in Asheville. For those of you that can't wait, I've got four words for you that completely describe the class ...


If you were in the 2008 class that Chris and I presented in Orlando, you'll understand immediately.

See you in Asheville next week.