Featured Post

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

Friday, November 18, 2016

Those who know don't tell and those who tell don't know.

The old Zen proverb, "Those who know don't tell and those who tell don't know," comes to mind when reviewing the many marketing websites for those trying to find work as video/image analysts for the many court districts around the US.

In most states, the evidence code defines the process whereby a person's testimony may be admitted and how that testimony will be treated. In California, for example, the EC notes,

  • "720. (a) A person is qualified to testify as an expert if he has special knowledge, skill, experience, training, or education sufficient to qualify him as an expert on the subject to which his testimony relates. Against the objection of a party, such special knowledge, skill, experience, training, or education must be shown before the witness may testify as an expert."
  • "721. (a) Subject to subdivision (b), a witness testifying as an expert may be cross-examined to the same extent as any other witness and, in addition, may be fully cross-examined as to (1) his or her qualifications, (2) the subject to which his or her expert testimony relates, and (3) the matter upon which his or her opinion is based and the reasons for his or her opinion."

There are many folks out there advertising themselves as "experts." For these folks, they've given themselves the title of "expert" for the purpose of marketing and helping their brand rise above the others in their field.

To be sure, only the Judge in a trial grants a person expert status. This status, as it relates to the testimony given, goes away once the person is finished with that trial. A person may give expert testimony in one trial, but not in another. If a person refers to themselves as an "expert," red flags should immediately go up.

Thus begins the process I call, Find the Charlatan. A charlatan is a person falsely claiming to have a special knowledge or skill; a fraud.

The common objections, at least in California, center around a witness' qualifications. "The trial court is not required to formally certify or qualify a witness as an expert. Evid. Code §§402, 801. Thus, it is unnecessary to “proffer” or “tender” the witness as an expert. Most judges, if asked to certify or qualify, will state merely that the witness is acceptable or may testify. This avoids giving the witness an imprimatur of undue credibility." (source) If, in checking on the witness' credentials, you find that a witness has improperly listed their testimony, that's another red flag. Remember, only the Judge may declare the witness an expert, and thus the testimony given as "expert testimony."

So how do these folks continue to get away with this false/misleading claim? No one checks or calls them on their activities.

  • Ask for a CV. Make sure that their record of testimony properly segregates testimony into general / expert. If it doesn't ask specifics as to the testimony given and any objections. 
  • Get transcripts / reports from prior testimony. Does the opinion offered in your case track with what they've said in the past? Does their listed training and experience track with previously submitted information? If not, why not? Sometimes, folks will try to say that their Fine Arts degree has an "emphasis" in some scientific discipline. This is very rarely the case, and they're likely embellishing their credentials.
  • Make sure that their training is actually training, not sitting in on a lecture. If they've offered training, make sure that they've trained people - not just presented topics as part of a lecture. Training peers or new employees is quite different from presenting information to a group of people who don't actually work in the discipline - a Bar association or a leadership conference for executives. If there was no syllabus or test involved, then it was likely an information session.
  • If they rely upon their college degree for the authority behind their opinion - get their transcripts and check their claims. If they refuse to let you have their transcripts, then they're likely hiding something. 

To put all of this into perspective, I've spent the week at LEVA's 2016 Training Conference. Some of the sessions are "training," some are "presentations." I presented information around redacting video footage from body worn cameras and in-car video, the technological and policy implications. Folks in my class weren't "trained" to redact footage. That's done at my Basic course in Henderson, or on a customer's site. Likewise for my presentation on image authentication. It's information, not training. If folks want training in image authentication, they can take the training course on that topic.

If you have a doubt about the person you wish to hire, there's likely a good reason. Google their name. Find out about them. Find out what others are saying about them. Ask questions of their peer groups. Find out if they've been discredited or had their testimony excluded in the past. Ask. Ask. Ask.


Monday, November 7, 2016

Is Photoshop a verb?

I'm momentarily out of my self-imposed exile to post this brief note on my old friend, Photoshop.

When processing images and video for the court, many agencies will try to do more with less. They will make the mistake of thinking that standard commercial photo editors will be fine. The software is often very inexpensive or even free. But, would you trust your case to just any piece of software?

In courtrooms and the media, there’s a common term for modifying images and video in a nefarious or bad way: it’s called “Photoshopping.” Fashion models are Photoshopped. Advertisements contain fantastic Photoshopped compositions. So the question is: now that Photoshop is a verb and not a forensic tool, do you want your evidentiary images and video to be Photoshopped?

Is it legal?

There are two major legal “standards” to deal with in the US: Frye and Daubert. There are some variations to these, like California’s modified Frye (Kelly-Frye). Then there are the various evidence codes. Each state has its own evidence code. The US Federal Government has theirs as well.

The Frye Standard
Where novel scientific evidence is at issue, the Frye inquiry allows the judiciary to defer to scientific expertise precisely as to whether or not it has gained “general acceptance” in the relevant field. The trial court’s gatekeeper role in this respect is conservative, thus helping to keep “pseudoscience” out of the courtroom.
In Frye states, Amped Software products meet the “general acceptance” threshold. Amped Software technology is used by all major law enforcement agencies in the US (local, state, and federal), most of Canada’s LE agencies, and around the world, in more the 60 countries as of today.

The Daubert Standard

In Frye states, it’s usually enough for a practitioner to inform the court that the tool in question has “general acceptance.” In Daubert states, and with the US government in Federal Courts, the Daubert standard goes from “tell me” (Frye) to “show me” (Daubert). If there’s a question about a particular tool or technique, either side can request a “Daubert Hearing.” Within the hearing, the tool/technique is demonstrated and questions as to process, science, etc. are asked/answered. If the judge is satisfied that the tool/technique passes the test, then the evidence is admitted.

Daubert places a heavy emphasis on science and the scientific method.
Has the scientific theory or technique been empirically tested? According to K. Popper (1989) in The Growth of Scientific Knowledge, “the criterion on the scientific status of a theory is its falsifiability, refutability, and testability”.

In the case of Amped Software products, the answer is yes. The source of the algorithms is scientific / academic texts and papers. They’re all from peer-reviewed and publicly available sources. The scientific status of the theory / technique is covered in each referenced source. In the case of Amped FIVE, the source of the algorithm, publication date, etc., for all filters used are all included in the report generated by the user.
All trial courts make a preliminary determination of admissibility. This job involves a preliminary assessment of whether the evidence is relevant, competent, and material. In short, can the evidence be properly applied to the facts in this case? This is the traditional “gatekeeping” function of courts.
This has more to do with the competence of the user than the tool itself. Has the practitioner properly used his/her tool? Is the practitioner qualified / trained in the use of the tool? Amped Software has trained hundreds of practitioners all over the world. Thus, this part of the gatekeeping is satisfied with the training.
Has the scientific theory or technique been subjected to peer review and publication? This ensures that flaws in the methodology would have been detected and that the technique is finding its way into use via the literature.
Again, all of the algorithms in Amped Software products come from peer-reviewed publications.
Can the technique and its results be explained with sufficient clarity and simplicity so that the court and the jury can understand its plain meaning? This is just the Marx standard, which is assumed to be incorporated in Daubert as it was with Frye.
This is the real beauty of the reports generated by Amped Software products. Amped FIVE’s report, for example, gives the user the plain English explanation, the more detailed scientific/academic explanation, as well as the filter settings, and the reference source for each filter that is applied within the workflow. No other product on the market does this.

Federal Rules of Evidence (Rule 702)
General acceptance is not a necessary precondition to the admissibility of scientific evidence under the Federal Rules of Evidence (FRE), but the FRE – especially rule 702 – do assign the trial judge the task of ensuring that an expert’s testimony both rests on a reliable foundation and is relevant to the task at hand. Pertinent evidence based on scientifically valid principles will satisfy those demands..”

“If scientific, technical, or other specialized knowledge will assist the trier of fact to understand the evidence or to determine a fact in issue, a witness qualified as an expert by knowledge, skill, experience, training, or education, may testify thereto in the form of an opinion or otherwise, if (1) the testimony is based upon sufficient facts or data, (2) the testimony is the product of reliable principles and methods, and (3) the witness has applied the principles and methods reliably to the facts of the case.”
Not to beat the point to death, but Amped Software products more than meet this standard.

The thing that our customers love so much, is that the work that they do can be presented with such a rock-solid report, that gives them the confidence to present what they’ve done to the judge/jury. They don’t have to be a rocket scientist (though a few of our customers actually are). They just have to follow their training and confidently read from the report.

Q: I see that you’ve applied a Levels filter in step 4. Does that filter’s algorithm come from a peer-reviewed and published source.

A: Yes, indeed. It’s from Dr. Anil Jain’s 1989 textbook, Fundamentals of Digital Image Processing. You can find the algorithm’s discussion and description on page ...

It’s important to know that over 50% of the filters in Photoshop come from proprietary algorithms (internal, patented, secret). That number continues to grow as Adobe supports its main customers--professional photographers.

For those using freeware tools, where’s the references and support? It’s non-existent.

So here’s something to consider when working in law enforcement forensics where a mistake could either convict the innocent or free the guilty: don’t you want a rock-solid toolset that any court in the land would accept? Of course, you would.

For more information, or to request a quote, click here.

Note: the italicized text comes from Mark Stevens, LtCol USMC (Ret), Assistant Professor of Criminology, California State University Fresno, and his course on using multimedia evidence in trial.

Photoshop Copyright © 2016 Adobe Systems Incorporated