Linear vs. Circular

[Charles Parker]

In Tampa a few weeks ago I attended a presentation of which the main subject was bias and how to prevent it. Part of the speaker’s presentation was a little presentation on Linear vs. Circular assignment of friction ridge events between phase 1 (analysis) and phase 2 (comparison) of the examination. The speaker discussed their agency process of Linear Analysis. I had read previously some of the comments on this topic and thought it would be an interesting discussion. I had several questions for the speaker on the topic but the presentation ended on time and another presentation was starting. Besides I was scheduled to attend a different presentation and was not able to ask any further questions on the speaker’s agency procedures. I am hoping that listing here what I can remember of the presentation perhaps readers of this forum can answer some of my questions.

LINEAR—The speaker stated that the objective was to observe and note all the FR Events in the unknown print during phase 1 (analysis) before moving to phase 2 (comparison). The speaker would have two photographs taken of the latent print and during phase 1 (analysis) mark all the features present. During phase 2 (comparison) if additional FR Events or Features were observed these were marked on the 2nd photograph with some type of description as to why they were not observed during the first assessment. The SOP stated that only one photograph was required but if additional FR Events were noted a second photograph would be necessary to indicate the additional FR Events as these could not be noted on the 1st photograph.

Linear Q1: Is this procedure done on all latent prints in an examination or just those that are complex?

Linear Q2: If I have a large palm print (ie 90+ points) do I mark all the ridge events or just a sufficient number that I am going to use for the comparison?

Linear Q3: What kind of responses are normal if ridge events are missed upon the first assessment but are noticed in phase 2 (comparison).

CIRCULAR—I am pretty sure that the speaker’s use of the term ‘Circular’ was taken from the OIG Report of the Mayfield error which is quoted here from the ‘Executive Summary’.

“The OIG found that a significant cause of the missed identification was that the LPU Examiners’ interpretation of some features in LPP 17 was adjusted or influenced by reasoning ‘Backward’ from features that were visible in the known prints of Mayfield. This bias is sometimes referred to as ‘circular reasoning,’ and it’s an important pitfall to be avoided. Having found as many as 10 points of unusual similarity, the FBI Examiners began to ‘find’ additional features in LFP 17 that were not really there, but rather were suggested to the examiners by features in the Mayfield prints. As a result of this process, murky or ambiguous

details in LFP 17 were erroneously identified as points of similarity with Mayfield’s prints.”

The description is that the latent print is assessed (evaluated) for all the features that are present including the ridge flow then three or perhaps four features are selected and then an attempt is made to find that grouping in the same relative area as the inked print. Once that initial grouping is found then the comparison is expanded in what some writers have described as a back and forth approach.

COMMENT: I have always considered this comparison technique as systematic and not circular. I was also taught that you do not take clear detail from one image and place them in ambiguous areas of a different image. It does not matter if it is latent print analysis, track identification, or any other type of pattern comparison. Doing so is “BAD PRACTICE”.

COMMENT: Generally I will break latent prints down into Simple, Moderate and Complex. I have used and will continue to use the description of Linear as a technique involving complex latent prints. But I also use the systematic approach for those that are very simple or moderate. Just as a mechanic will have different tools in their toolbox so should a LPE have different techniques available for the job at hand? Correct and accurate problem solving for latent print examination requires different tools and techniques. I have a pair of dividers as well as picks, but I certainly do not use the dividers in each and every case.

COMMENT: There are other techniques that are sometimes used such as ‘Point-to-Point’ and ‘Running the Ridges’, so why not have two other techniques available for the problem solving of latent prints depending upon the job at hand. Using a negative connotation such as ‘circular reasoning’ to describe a technique that has been used successfully for a number of years because it was not applied properly is not understanding the process of comparison fully.

COMMENT: Although ‘Circular Reasoning’ is considered a fallacy in reasoning, here is what one source says about ‘Linear Thinking’: This results when you do not have enough information and reduce the question to terms that are too simplistic. This fallacy is often expressed by the terms ‘either’ and ‘or,” providing only two sides or alternatives when more is possible”. (http://writing2.richmond.edu/writing/wweb/reason.html).

COMMENT: Using clear FRE in one image and applying those to ambiguous areas of a different image to justify or bolster ones examination is ‘BAD PRACTICE’, nothing more or nothing less.

As usual your comments or rebuttal both pro and con are welcome in this discussion. Also I must say I enjoyed this speakers presentation in Tampa and wished I had he opportunity to ask several questions.

[Gerald Clough]

I don't see this issue as anything complicated, and it shouldn't be made difficult by varying the conditions. Ideally, all the events are located and described during analysis of the evidentiary impression. If you can't characterize an event sufficiently to say in comparison that they have the same character, nothing in the record print can change that without it becoming improperly circular. The question, then, becomes what to do when you happen to be able to characterize a previously unnoticed event in the evidence after you've passed onto the record print or to comparison. It's absurd to be deprived of a feature that can genuinely be characterized well, simply because you missed it the first time. (By that logic, at least one of the "Problem Print" featured errors coult not be corrected, because the examiner clearly missed an outlying feature.) That would not only work an injustice to the potential value of the evidence, but it would also result in the inability to recognize that there was no match. But I think it does require the examiner to self-impose a more critical view of the missed or revised feature. If you did examine it the first time around and decided that it was insufficiently clear to use, I think any imagined further clarity is highly suspicious, because the insight came from the record print. A supposedly corresponding feature in the record print became, in effect, additional data assigned to the evidentiary impression, and that' corrupts the decision.

The Mayfield examination is a clear example of this. Many features of the latent are clearly of a specific character when intimately examined. Some were at least such that you can differentiate them to the point of settling on their general character in terms of ridge flow. But the analysis was clearly forced into congruence with the record print at a later stage. But in other cases, the influence of the record print on the latent analysis is not so clear. A strictly enforced rule of never modifying the analysis of the latent is not appropriate. But there is a hazard in the nature of human perception that makes me leery of mixing comparison and analysis by starting with a few features, locating the same relationship in the record and going back for first analysis of features in the latent with the record print in mind. I think it comes down to, as you say, appropriate professional use of your tools. Perhaps the best mind set (the ultimately vital tool) would be as if you knew the results were going to be reviewed by a very critical opposing expert. That's really what it comes down to, a question of whether each useful feature can be clearly characterized or is ambiguous and just what you can say in comparison about a feature that can be said to be consistent but not perfectly clear. We often talk about science. Well, in this issue, can you defend your proof against a critical expert? The issue of circularity is about the potential for fooling yourself, not the potential for fooling others, and you're kidding yourself if you think you can't be fooled. The practical reality, though, is that because we're rarely reviewed by that kind of critic, it's left to us to be critical of our own decisions.

You know, that's all we ever are able to say, that a feature, in this instance, made an impression that had this or that character. Our names for various types of features are merely names describing various sorts of impressed features. We can only rarely talk about the skin itself, only about the record it made and how it compares to how the same or some other skin made an impression and whether they impress similarly.

Now, what to do if one realizes during comparison that there are features in the record that might match the latent if a different or additional attempt to obtain a better image of the latent was obtained? An easy out would be to say that the new images should be added and the case passed off to another examiner for a fresh start. But that is, I think, tantamount to declaring the the latent analysis must absolutely be locked in first and never revised, and that, as stated above, is inappropriate in that it prohibits correction of errors that cut both ways. Critical thinking and mind set again. You can't legislate it.

It's silly to S.O.P. something that's taking place in the examiner's mind, as if articulating in the S.O.P. shows it's done that way. S.O.P.'s, like all managerial decrees, must be subject to being checked, and you can't check something like this. It also kind of points up that Verification, in the sense it's normally used, is not a panacea. A verification may repeat the same analysis affected in the same way by the record, where a tough critique by an examiner who is looking at the first completed examination might well raise questions about some of the characterizations of latent feature, which is exactly what every opposing expert does.

[RL Tavernaro]

Gerald,
I think what you have described as essentially a 'best practice' is very reminiscent of what I described in another thread ("Case at Supreme Court" the first part of July) as part of my personal experience & development as a latent print examiner.

... Perhaps most important though, confrontation was a major component of workflow in latent print comparisons. In 1980, long before verification was in vogue for most agencies, it was greatly encouraged, albeit practiced in a different form in my lab. The different form involved challenge & confrontation. When an identification was contemplated, it was passed to another examiner. The second examiner would then do an independent comparison with a critical eye & expectation that it would not be an identification (a forced negative bias?). Even if the second examiner concluded that the two prints did come from a common source, they would withhold that conclusion, seeking areas that might require interpretation and/or explanation. The initial examiner would then be confronted & challenged with potential problems in the comparison. The initial examiner was then forced to defend the identification to the satisfaction of a skeptical and knowledgeable critic. Any required interpretations and/or explanations had to be sufficient to withstand rigorous scientific scrutiny (or as we would ask: would it hold water (or was it leaky)?). ....

I agree that benefit can be derived from a back & forth approach, as opposed to strictly linear. Back & forth does not necessarily mean circular as in circular reasoning. An argument could be made that fixing an image in the mind of features present in the known exemplar (as the result of an initial analysis only) could result in a forced fitting to the unknown.

Features found in a given print must be analyzed and evaluated in respect to their independent relationship to all other features in that print. In a comparison with another print, that evaluation extends to the independent relationship to all features in both prints. In other words, the confirmation of an identification (individualization) will not come from a single specific sequence of features, but the fact that whatever sequence of getting from one feature to another is followed (counting & tracing ridges to navigate around the print), the results will always match. Ultimately, that must hold true whether starting with the exemplar, or the evidence print.
Regards, RLT

[16_Points]

Let's not get caught up in the "Linear vs. Circular" issue too much. The real theme of this style of documentation is Transparency. Feel free to conduct your examinations in any fashion you want. Just make sure it's documented in a fashion that another examiner can follow your steps.

Why? Well, I'm glad you asked. Complex latent print examinations are . . . complex? Yes, of course, and sometimes we must interpret or re-interpret the friction arrangements. (Egad, you mean an examiner can change his/her mind during the comparison?) And that's just fine . . . as long as you document the re-interpretation. As a peer reviewer, I want to know how much re-interpretation was needed in order to reach your conclusion. It's a good indicator of how "difficult" the comparison was and the relative risk of error. And of all the friction ridge detail used, how much did I "see" during the analysis and how many minutiae did I mark after looking at the known. No one says you can't add additional details. Just be clear that they were marked after seeing the known. That's transparency.

If you have a large clear print with a 100 minutiae, great you know it's AFIS quality and will make an easy ID. But as Charles suggests, you may not use all 100 minutiae to effect the ID. That's fine too. Remember, it's transparency we're after. Just mark the minutiae you compared (16 points maybe?) as that is an accurate description of the ridge detail you used to make the ID. If you only compared 8 points in a latent with 100 minutiae then the defense should be able to ask you why you only needed those 8. And if you don't document it in your notes, how can you tell a jury what parts of the latent (or how many minutiae) you used to conduct your examination?

Annotating the latent is a powerful documentary tool, especially if you have multiple comparison quality copies. Just poke a hole wherever you find "Level Two details" (just call them points) or whatever discriminating ridge detail you're using. It's not that hard, and it doesn't take much time (in most cases if you have a process to provide multiple copies of the latent).

And keep in mind court cases such as NH v Langill and Illinois v Stafford. The courts are asking us "how" we effected the identification in this case (i.e. the one you are testifying to not IDs in general).

[kevin]

As a peer reviewer, I want to know how much re-interpretation was needed in order to reach your conclusion. It's a good indicator of how "difficult" the comparison was and the relative risk of error. And of all the friction ridge detail used, how much did I "see" during the analysis and how many minutiae did I mark after looking at the known. No one says you can't add additional details. Just be clear that they were marked after seeing the known. That's transparency.

Quick question-by peer review are you speaking towards the verification stage of the examination? or peer review in a more general sense?? Do other examiners out there go over the initial comparison with the other examiner either before or during their verification? I apologize if I'm misreading....

I'm with you on re-evaluating an impression and re-interpreting points of similarity. Often times I mark an 'event' on a latent during AFIS search in which i know there is something there, but I am not 100% certain as to whether it is a bifurcation or ridge ending (and even that may change from the rolled and flat impressions).

[16_Points]

I'm speaking of peer review in a very broad sense. Verification is essentially another examiner reaching the same conclusion given the same latent and known print. Peer review (or technical review) is the critical analysis of another analyst's work product. Basically, as a peer reviewer, I want to know how well the initial examiner did his/her job (not just whether or not you got the right answer). It's like showing all your work when doing a math problem in school. You get a few points for the right answer, but the teacher really wants to know if you applied the math formulas correctly.

Latent examiners have a double whammy. Scientific peer review and court review. Meaning the courts should be able to determine whether or not the examiner applied the methodology correctly without ipse dixit testimony from a verifier (meaning detailed notes, a narrative, annotated latents, a charted enlargement, . . . something).

[Charles Parker]

As I said in my original post I break latent prints down from Simple to Complex and I use a variety of problem solving tools for the job at hand. I should also state that probably only about 5% are by my definition and even Mr. Ashbaugh’s formula is a complex latent print. I have used and will continue to use the linear approach (I refer to it as Super Systematic SSY) on most of those that are complex. I also use the circular approach (I refer to it as Systematic SY) on those other cases.

Gerald States Paragraph 1

“I don’t see this issue as anything complicated, and it shouldn’t be made difficult by varying the conditions.”

I agree that if we were to sit down and dissect these two approaches to the nth degree they would have more in common than differences. Even linear must at some point do a back and forth comparison. What concerns me is misunderstanding. Having been cross-examined in four different cases on the OIG report my concern is in the court room where my approach to the comparison is circular as described by the OIG and the defense makes a good effort in describing this is not the proper approach. I am not going to be able to describe to a jury the differences that we have talked about on this post without probably losing half of them and I certainly cannot turn to the jury and tell them “The OIG got it wrong” (which I personally believe is the correct answer) without seeming arrogant. My concern is the misunderstanding by new examiners, those LPE who only see black and white, the critics of this discipline, the legal profession, and a number of other types of persuasion.

Gerald States Paragraph 1

“But I think it does require the examiner to self-impose a more critical view of the missed or revised feature……………A supposedly corresponding feature in the record print became, in effect, additional data assigned to the evidentiary impression, and that’ corrupts the decision.”

This is the tough part that comes from training and discussion with other LPE and new examiners to get this mind set of critical view. The second part I am on the fence on. For example several years ago there was a latent print in a high profile case. During phase 1 I observed something in the lower quadrant that could be something in substrate or could be a disruption of the friction ridge detail. Only during comparison did I make the observation that it was a disruption (scar) that was very thin and indistinct even in the exemplar. It occupied the same relative position but only after seeing it in the exemplar did I come to the decision of what it actually was. Was my decision corrupted because I saw it in the exemplar? I think not as it only reinforced what I observed. But as you say this could be a tar pit and one must be very careful.

Gerald States Paragraph 2

“But there is a hazard in the nature of human perception that makes me leery of mixing comparison and analysis by starting with a few features, locating the same relationship in the record and going back for first analysis of features in the latent with the record print in mind”.

This very application is what I believe to be the core of human comparison whether it is an art expert examining for a forgery or you going down the aisle comparing prices of products. Yes it maybe that human perception is a hazard but I do not think is as fragile as some might state. It is my understanding that most human minds can only retain 3 or 4 ridge characteristics in sequence at one time. I am sure there are some who can remember more but if the average examiner can only retain 3 or 4 ridge events in sequence there will have to be some type of back and forth even from the linear approach if all FRE are recorded on paper. But you end this paragraph with stating that we must be critical of others and we need to be critical of our own decisions. That is one of the best tools we have.

Gerald Says Paragraph 4

“It’s silly to S.O.P. something that’s taking place in the examiner’s mind…….where a tough critique by an examiner who is looking at the first completed examination might well raise questions about some of the characterizations of latent feature, which is exactly what every opposing expert does”.

I agree and the key phrase is ‘tough critique’ which I will go into further in my response to 16 Points.

RLT Says

“The second examiner would then do an independent comparison with a critical eye…….”

A point I wish to explore further in my response to 16 Points.

RLT Says

“Ultimately, that must hold true whether starting with the exemplar, or the evidence print”.

In Ashbaugh’s book he describes the same thing and I was taught to start with the bad and move to the better and in some cases the exemplar is far worse than the latent print. But is this same principle taught and is it applied by the newer examiners. Is it a principle that we need to drop and only and always concentrate on the latent print first and always?

Bob as always you bring an insight to the discussion and there was not much there to disagree with on my part.

16 Points, welcome to the discussion. When I saw this presentation in Tampa the speaker was concentrating on Bias and in respect to this linear vs. circular was saying that if you do linear you were showing Bias the door where circular you were inviting Bias to sit down and each lunch with you. I disagree with the speaker on that one point. It was the misapplication of the process and not the process itself that was part of the problem. Now this discussion has moved from the realm of Bias to Documentation. Which is not a bad thing and is probably due the very act of documentation in the linear approach.
Now I am in favor of ‘Transparency’ which seems to be the banner of the 21st century. However I also believe that ‘Transparency’ is really a subjective observation. Group A thinks XY is transparent and Group B says it is not enough. Where the line is to be drawn in sand has yet to be accomplished.

The other banner that is waved a lot is the “documented in a fashion that another examiner can follow your steps”. Correct me if I am wrong but I think we are dealing with a couple different types of review here. Gerald used the term ‘Critical’ review and RLT used the term ‘Independent’ review. Before accreditation it was always stated as an independent review for verification. In verification I do not want to see what the other person has done. I want to do an independent review that is critical and if the conclusion is the same then it might go on to some other type of review. If the verification conclusion is different then both sides need to see how each other reached those conclusions. Once the independent review is done then as accreditation states the case must go though an administrative and/or technical review. The technical review is the one where a determination is made if they used proper protocol and methodology.

A long way around but this brings me back to linear vs circular. If a LPE believes that the OIG report is correct and that any approach using the back and forth is circular and therefore not proper methodology then by default if I use the back and forth application then it is circular and my methodology is flawed and I fail the technical review.

But it is good to see that you state “Feel free to conduct your examination in any fashion you want. Just make sure it’s documented…..” By making such a statement you must also believe the OIG got it wrong?

In reading your post I have another question----do you document the simple examinations to the same level as you would a complex examination? In other words, different tools, and different levels of documentation depending upon different levels of examination!

You make the following statement which I agree with you 110% “…..how can you tell a jury what parts of the latent (or how many minutiae) you used to conduct your examination?” I would add how many friction ridge events agreed between the two images---keeping in mind that there are no minimum number required.

Second to the last paragraph. “It’s not that hard, and it doesn’t take much time (in most cases if you have a process to provide multiple copies of the latent).” How hard it is depends upon ones agency SOP’s and what they require. I have a number of documentation forms from different agencies and some are simple and some are quite complex themselves. It also depends if you are required to document the substrate and matrix on each and other features that just does not come into play with the simple type of examinations. Time will add together depending on the number of cases you have. My agency does not have a simple process to provide multiple copies of the latent. With 800 cases a month, do we document them coming in the door or when an examiner is assigned? Simple statements are not simple fixes when there are over 2,000 agencies doing latent print examinations in this country. Do we do the same level of documentation for incidental examinations (victims, officers, legitimate access, etc.). The transparency banner sometimes smothers the ‘timely return’ banner.

The court question is a long one and deserves a different thread. Thanks for the input.

[16_Points]

When I saw this presentation in Tampa the speaker was concentrating on Bias and in respect to this linear vs. circular was saying that if you do linear you were showing Bias the door where circular you were inviting Bias to sit down and each lunch with you.

Again, I think we're missing the point of the Linear vs. Circular vis-a-vis Bias relationship. In fact, it's not really accurate to call it linear. The objective of this approach, as I understood it, is to mark/annotate all the "level two details" (I call them points) in the latent image prior to looking at the known. The key here is deciding what information in the latent print is fair game before moving to the comparison.

Is that a minutiae, pore, edge shape I see in the area of reduced quality (in the latent)? Suppose you didn't see the "ridge detail" during analysis, but you notice it after looking at the known. This is where the linear approach comes in. You have a latent. You mark 7 minutiae and 2 incipient ridges and 1 edge shape during analysis. You search, compare, and match the latent to a known standard. Your charted enlargement shows 11 minutiae and no 3rd level detail. You picked up 4 minutiae during comparison that you didn't "see" during analysis and the 3rd level detail wasn't reproduced in the exemplar. Do you call the ID? Do ask for more standards hoping to reproduce the 3rd level detail?

Going back to the main idea, documenting the analysis shows you where bias may have sneaked in. For example, where did the four extra minutiae come from? Were they in an area of reduced quality of the latent? The published studies on bias agree that subjects are most susceptible to bias when the data is ambiguous. So, making the process "linear" doesn't eliminate or block out bias. It just provides a level of transparency to expose potential bias effects.

In verification I do not want to see what the other person has done. I want to do an independent review that is critical and if the conclusion is the same then it might go on to some other type of review. If the verification conclusion is different then both sides need to see how each other reached those conclusions. Once the independent review is done then as accreditation states the case must go though an administrative and/or technical review. The technical review is the one where a determination is made if they used proper protocol and methodology.

I apologize. I thought I clarified this in my second post. Verification is another examiner reaching the same conclusion from the data set. Peer/Technical review is a critical look at "how" the examiner conducted his analysis. Yes, you want the documentation to be available for technical review. Yes, the verifier doesn't need the documentation.

A long way around but this brings me back to linear vs circular. If a LPE believes that the OIG report is correct and that any approach using the back and forth is circular and therefore not proper methodology then by default if I use the back and forth application then it is circular and my methodology is flawed and I fail the technical review.

Everyone seems to get hung up on this issue. Going back and forth is OK. But if you can only see "ridge detail" in the latent after looking at the known, then it's a problem. Which leads us to, How do I know you're not using the known to interpret the latent if you don't document first?

But it is good to see that you state “Feel free to conduct your examination in any fashion you want. Just make sure it’s documented…..” By making such a statement you must also believe the OIG got it wrong?

Yes and no. I'll have to explain another time.

In reading your post I have another question----do you document the simple examinations to the same level as you would a complex examination? In other words, different tools, and different levels of documentation depending upon different levels of examination!

The key is that bias mainly affects ambiguous data. You don't need as much documentation for clean, clear prints (high clarity).

Second to the last paragraph. “It’s not that hard, and it doesn’t take much time (in most cases if you have a process to provide multiple copies of the latent).” How hard it is depends upon ones agency SOP’s and what they require. I have a number of documentation forms from different agencies and some are simple and some are quite complex themselves. It also depends if you are required to document the substrate and matrix on each and other features that just does not come into play with the simple type of examinations. Time will add together depending on the number of cases you have. My agency does not have a simple process to provide multiple copies of the latent. With 800 cases a month, do we document them coming in the door or when an examiner is assigned? Simple statements are not simple fixes when there are over 2,000 agencies doing latent print examinations in this country. Do we do the same level of documentation for incidental examinations (victims, officers, legitimate access, etc.). The transparency banner sometimes smothers the ‘timely return’ banner.

Excellent point. It is only a matter of time before one or more of the various agencies or committees (SWGFAST, NIJ, NIST, NIFS?) standardizes the documentation process. The Feds via NIJ/NIST will loosen up grant funds to entice agencies into compliance. etc. etc. It's only a matter of time. Further, as I mentioned previously, the courts are beginning to ask these questions regarding documentation and bias. So we'll be squeezed at both ends. I only see it as a matter of time (certainly many years).

[Charles Parker]

16 Points

Maybe I am missing the point. So let me try again with some direct questions but first I must describe the process.

THE PROCESS: I have a case with 25 latent prints. I now take two photographs of each latent print—50 photographs total. I sit one photograph aside. With one photograph I first mark or indicate my pattern flow, then mark all the 2LD I observe. Now I mark all the 3LD and any other intrinsic features such as creases, scars, etc. Next I mark and describe any distortion that may be present. When that is done I make documentation notes on the substrate and the matrix, then developing techniques that were applied (and in which order), then the method of preservation either lift, casting or photography. I also make sure I note the time I started and the time I finished (could be proof that I am not just doing the fluff thing). Once I have done the first photograph on all 25 latent prints now I turn to my exemplar. Document is it live scan or ink or inkless or some other type of recording. Are all the identifiers in place and if they are not, then document that they are not and what steps were taken to obtain the correct identifiers. Make sure I note these corrected identifiers on the exemplar and if I have to scratch something out to make sure I initial the cross through mark. Also initial any hesitation or re-marks that I might have made. Now I add all case identifiers to the exemplar, making sure the information is correct. If not what steps did I take to find the corrected information? Now I am ready for the comparison. During the comparison which I look at several features in the latent print and try to find them in the exemplar. Keeping in mind that I am now comparing an enlargement with original size unless I also photographed each finger and palm to the same size as the enlarged latent print. So I find these points and I still have to go back and forth because I cannot possible remember all the details I marked in the latent print so I go back and forth. All of a sudden I see a small dot or incipient ridge. Go to the latent print and lo and behold there it is and I just thought it was background noise. So now I pull up my second photo and mark the incipient ridge and have to explain why I did not catch it the first time around. Finally I have completed and I make note or an accounting of all the features that I found in agreement between the two. Now I am on to the next one and the next one, and the next one.

Did I say that the case was an auto burglary and the AFIS search yielded the victims prints?

Now that might sound whimsical to you and it probably is but with vague comments about transparency and documentation I assure you this is what some are thinking. People are asking for clarification and examples and nobody and I repeat nobody is providing examples. The only answers they are getting are “Well it does not matter what you do as long as it is transparent and you document it”.

So here are my questions.

1. Do you do that level of documentation with each and all latent prints in each and every case?
2. If you do not use that level for each latent of each case are you able to choose the level of documentation for the type of case being examined?
3. If you allow LPE to choose the level and type of documentation are you not losing your transparency shield because some of them will go the fluff route.
4. Do you document the number of friction ridge events in agreement between the two images. A lot do not because of the “No Scientific Basis For A Set Number Of Points”. But if you do not----then you are not transparent.
5. When you see FRE during the back and forth, what verbiage do you use to document it? “Well I just did not see it the first time around”, “Well it was hiding in kind of a smudge area”, “Well I just had a bad day this morning—we all have those”.

One of my reasons for this posting is that I believe that LPE should be able to choose the level of documentation needed for the case at hand. If they choose wrong then they are held accountable. Legislating and SOPing the mental process is in my opinion the wrong way to approach it.

Another reason for this posting is that I do not believe the OIG is correct in calling that circular and is one of the reasons for the error. The misapplication of that technique was the problem and not the technique itself. Going back and forth is not bad. Only applying friction ridge detail to ambiguous areas is bad.

Finally to my point that I think everyone is missing is that the OIG says it is wrong, the FBI review of the incident says it is wrong, the FBI SOP does not use it, other agencies have stopped (on paper) using it. So now if I continue to use it on selected cases then I am not applying the methodology correctly and I fail the technical process. I might have the right answer but the way I got there is wrong and is not acceptable. I find that situation ludicrous.

But 16 Points---thanks for responding. I hope to hear from you some more.

[16_Points]

16 Points

Maybe I am missing the point. So let me try again with some direct questions but first I must describe the process.

THE PROCESS: I have a case with 25 latent prints. I now take two photographs of each latent print—50 photographs total. I sit one photograph aside. With one photograph I first mark or indicate my pattern flow, then mark all the 2LD I observe. Now I mark all the 3LD and any other intrinsic features such as creases, scars, etc. Next I mark and describe any distortion that may be present. When that is done I make documentation notes on the substrate and the matrix, then developing techniques that were applied (and in which order), then the method of preservation either lift, casting or photography. I also make sure I note the time I started and the time I finished (could be proof that I am not just doing the fluff thing). Once I have done the first photograph on all 25 latent prints now I turn to my exemplar. Document is it live scan or ink or inkless or some other type of recording. Are all the identifiers in place and if they are not, then document that they are not and what steps were taken to obtain the correct identifiers. Make sure I note these corrected identifiers on the exemplar and if I have to scratch something out to make sure I initial the cross through mark. Also initial any hesitation or re-marks that I might have made. Now I add all case identifiers to the exemplar, making sure the information is correct. If not what steps did I take to find the corrected information? Now I am ready for the comparison. During the comparison which I look at several features in the latent print and try to find them in the exemplar. Keeping in mind that I am now comparing an enlargement with original size unless I also photographed each finger and palm to the same size as the enlarged latent print.

Reductio ad absurdum. You can make up any string of procedural steps to make it sound like a chore. Maybe some agencies make you do all that. But would that be the only convoluted and illogical SOP at your agency? And I'm not here to write your SOPs or technical manual. This is an opportunity for all the smart and creative people we have in the latent print community to come up with solutions that fit their agencies and their jurisdictions.

So I find these points and I still have to go back and forth because I cannot possible remember all the details I marked in the latent print so I go back and forth. All of a sudden I see a small dot or incipient ridge. Go to the latent print and lo and behold there it is and I just thought it was background noise. So now I pull up my second photo and mark the incipient ridge and have to explain why I did not catch it the first time around. Finally I have completed and I make note or an accounting of all the features that I found in agreement between the two. Now I am on to the next one and the next one, and the next one.

You see, you can call them points without bursting into flames or being struck by lightning. And basically, yes. You claim that you didn't mark the incipient because "I just thought it was background noise". Your words not mine. And that tells me that you saw it but didn't think it was real ridge detail (otherwise you would have marked it). So if you're changing your analysis from "background noise" to ridge detail contributing to an ID, that's pretty significant. And, yes, you should have an explanation for it.

Why? Because this is an excellent example of circular reasoning. I marked the "background noise" in the latent because in the known print I saw an incipient ridge. You basically made my point for me.

Did I say that the case was an auto burglary and the AFIS search yielded the victims prints?

Essentially irrelevant at this stage of the discussion.

Now that might sound whimsical to you and it probably is but with vague comments about transparency and documentation I assure you this is what some are thinking. People are asking for clarification and examples and nobody and I repeat nobody is providing examples. The only answers they are getting are “Well it does not matter what you do as long as it is transparent and you document it”.

The FBI gave you an outline of their procedure in the aforementioned IAI presentation. Glenn Langenburg, as I understand it, teaches a method of annotation in his Ridgeology workshop. So there are examples out there. But think of it like ASCLD-LAB. ASCLD doesn't write your SOPs for you. They just tell you what they should cover.

So here are my questions.

1. Do you do that level of documentation with each and all latent prints in each and every case?

Asked and answered. The level of documentation depends on the clarity of the print. That is how ambiguous is the ride detail? More ambiguity=more documentation.

2. If you do not use that level for each latent of each case are you able to choose the level of documentation for the type of case being examined?

No, the type of case is irrelevant. The documentation should be proportional to the difficulty of the latent print comparison. However, if you want to limit examinations of minor property crimes, I'm OK with that. For example, only examining AFIS quality prints.

3. If you allow LPE to choose the level and type of documentation are you not losing your transparency shield because some of them will go the fluff route.

Hey, you can fluff any comparison and conspire with your co-workers/supervisors to commit fraud. Hopefully there is some integrity left amongst LPEs. Then the technical reviewer and/or supervisor should be assessing the amount and quality of documentation (they should be doing that anyway).

4. Do you document the number of friction ridge events in agreement between the two images. A lot do not because of the “No Scientific Basis For A Set Number Of Points”. But if you do not----then you are not transparent.

Wow, people really get carried away with this whole "No scientific basis for a set number of points." The Standardization Committee stated, there was no basis for a point standard "at that time". They also recommended a federally funded study to determine the distribution of minutiae in the population so that we would have some statistically significant value associated with the "rarity of features". Go back and read the report of the Standardization Committee if you don't believe me.

Anyhow, so because there is no point standard, does that mean you didn't use any minutiae to make your comparison? And third level detail is not always present in a latent of value for identification. These IDs tend to be based largely (though perhaps not exclusively) on . . . points. So by indicating what minutiae were used you are actually . . . documenting the basis of your identification? Huh, who would have thunk it? (Some examiners look at "ridges in sequence", myself included. You still use minutiae, otherwise all the ridges would have an identical sequence. You can still count the number of ridges in sequence and/or mark the ridges in sequences used to make the ID).

5. When you see FRE during the back and forth, what verbiage do you use to document it? “Well I just did not see it the first time around”, “Well it was hiding in kind of a smudge area”, “Well I just had a bad day this morning—we all have those”.

Sure. But to my knowledge, most agencies that annotate latent lifts use a color coded system to differentiate between initial analysis and minutiae marked during comparison.

One of my reasons for this posting is that I believe that LPE should be able to choose the level of documentation needed for the case at hand. If they choose wrong then they are held accountable. Legislating and SOPing the mental process is in my opinion the wrong way to approach it.

Another reason for this posting is that I do not believe the OIG is correct in calling that circular and is one of the reasons for the error. The misapplication of that technique was the problem and not the technique itself. Going back and forth is not bad. Only applying friction ridge detail to ambiguous areas is bad.

So I'm confused here. Yes, I do get confused . . . quite easily in fact. If you don't document the analysis and the "back and forth" as you call it, how do you know the method was applied incorrectly? When is it OK to look at a latent impression, say "Look there's some background noise", then look at the known and say "Oh no, that's 3rd level detail . . ." And when is that not OK? Or it's just the subjective opinion of the examiner? How does a tech reviewer know that you didn't mark your 3rd level detail until after seeing the known? Or are you saying it doesn't matter?

Finally to my point that I think everyone is missing is that the OIG says it is wrong, the FBI review of the incident says it is wrong, the FBI SOP does not use it, other agencies have stopped (on paper) using it. So now if I continue to use it on selected cases then I am not applying the methodology correctly and I fail the technical process. I might have the right answer but the way I got there is wrong and is not acceptable. I find that situation ludicrous.

But 16 Points---thanks for responding. I hope to hear from you some more.

Again, this process really applies to areas of ambiguity. If you have ridge detail with high clarity, then going "back and forth" doesn't really matter because you can't reinterpret the latent based on the known. It really is just that simple. Good luck to all.

[Gerald Clough]

Gerald used the term ‘Critical’ review and RLT used the term ‘Independent’ review.

The "critical review" that I had in mind was a concept that I thought would be helpful to an examiner who is involved in a complex examination and is trying to carefully maintain their own perspective. I think any qualified person reviewing the results of an examination will readily understand what led the original examiner to make various decisions, assigning characteristics to a feature, judging two sets of points to be congruent, etc. A critical view considers any possible alternative interpretations, to the end of determining if the likelihood of an alternative being correct is sufficient to either remove some structure from the cumulative data or if it represents a sufficiently clear discrepancy. It's the sort of review one might do if given the results of an examination and asked to consider that it might be incorrect and why it might be incorrect. Again, if we're playing science, this raising of questions is commonly done in the practice of demanding the most rigorous demonstrations of particular interpretations of data. While, to date, this kind of review is not common with latent print evidence, it is certainly common done with many other forms of evidence. The more openly criticized are the conclusions in a field, the more common is the presentation of alternative interpretations, and because of the nature of criminal litigation, the level of critique and the degree of certainty that the other analyst is wrong rises to that of the most hotly debated scientific arguments.

In considering how someone else might interpret what one is observing and the judgments one is making from those observations, the examiner can confront those alternatives and either form a strong opinion and the arguments for it or abandon that portion of the data as insufficiently reliable. It still comes down to the ability to maintain the purely analytical mindset. I had an interesting experience once in a crime analysis course. The instructor announced on the first day that only half the class would be successful, for he had found that about 50 percent of his students, all of whom were criminal investigators, could not make the shift from investigative mind to analytical mind. He was right. At the end, about half the class could let the data speak for itself and answer the question, which was a problem of how to use particulars of a series of sexual assault attacks to prescribe the characteristics of a decoy victim who would attract the actor to attempt the next crime. The other half could not adjust to letting the data show the trends that needed to be identified. The half who could make the shift selected exactly the decoy that was used in the actual investigation and that attracted the actual actor. The rigid investigative minds selected a decoy who fit comfortably in the set of actual victims but failed utterly to predict what the actor wanted next. While not precisely analogous to fingerprint examination, the requirement to maintain the proper mind by an act of will was instructive. Most instructive, perhaps, was that the individual students could not accurately predict their success.

[Charles Parker]

16 Points, I glad you responded. After reading my post I was concerned that perhaps I might have been a little harsh, but trust me that was not my intent. I have a ghost reader that sometimes throws ice water on my writing, but she was off. I am glad you responded.

16 Points Said

Reductio ad absurdum. You can make up any string of procedural steps to make it sound like a chore. Maybe some agencies make you do all that. But would that be the only convoluted and illogical SOP at your agency? And I'm not here to write your SOPs or technical manual. This is an opportunity for all the smart and creative people we have in the latent print community to come up with solutions that fit their agencies and their jurisdictions.

Perhaps, but the first part came from the speaker in Tampa, the middle part came from what SWGFAST wants to implement and the last part comes from ASCLD-LAB. Not something my agency dreamed up. From some others perspective I may not be smart or creative, but I do want to do what is best for my agency, this jurisdiction, and the public we serve.

16 Points Said

You see, you can call them points without bursting into flames or being struck by lightning. And basically, yes. You claim that you didn't mark the incipient because "I just thought it was background noise". Your words not mine. And that tells me that you saw it but didn't think it was real ridge detail (otherwise you would have marked it). So if you're changing your analysis from "background noise" to ridge detail contributing to an ID, that's pretty significant. And, yes, you should have an explanation for it.

I do not have a problem with the word points, minutiae, or ridge characteristics, but my favourite is what John Nielson proposed several years ago and that was “Events”. I am not sure why you think I was afraid of the word “Points”.

I do not see it as changing my “Analysis”, but see it more as an observation that provided me with additional information. If it was not in the same relationship with the other “Points” or “Events” then it would remain just background clutter. It is not the event in and of itself but it’s the relationship with other “Points” that both images share.

16 Points Said

Why? Because this is an excellent example of circular reasoning. I marked the "background noise" in the latent because in the known print I saw an incipient ridge. You basically made my point for me.

From everything that I have read so far, additional information gathered or obtained by observations is not circular reasoning. Generally circular occurs when no new information is added. New information was added here by the observations made. Remember relationship is a major key but not the only one.

16 Points Said

Essentially irrelevant at this stage of the discussion.

Maybe for you, but for the others in this office conducting the level of documentation that is being offered are well---to use your words “Reductio ad absurdum” and in lay terms not reasonable. Your office may be able to do that and if you can for each and every latent that is sufficient and/or fails sufficiency---great, go for it. My office cannot. All I ask is that I am not made to jump in the black box with all the others on each and every examination.

16 Points Said

The FBI gave you an outline of their procedure in the aforementioned IAI presentation. Glenn Langenburg, as I understand it, teaches a method of annotation in his Ridgeology workshop. So there are examples out there. But think of it like ASCLD-LAB. ASCLD doesn't write your SOPs for you. They just tell you what they should cover.

There are a lot of examples out there and I have been collecting them for several years. Not to take anything away from Mr. Langeburg as his annotations are good, but I believe the best so far (at least to my liking) is M. Leanne Gray. Her class was full every time.

16 Points
1. Do you do that level of documentation with each and all latent prints in each and every case?
Asked and answered. The level of documentation depends on the clarity of the print. That is how ambiguous is the ride detail? More ambiguity=more documentation.

16 Points, thank you. Finally the question I have been posing all this time. I agree 110% that the more difficulty of the latent (level of complexity) then different levels of documentation. Thank-You I appreciate that.

16 Points
2. If you do not use that level for each latent of each case are you able to choose the level of documentation for the type of case being examined?
No, the type of case is irrelevant. The documentation should be proportional to the difficulty of the latent print comparison. However, if you want to limit examinations of minor property crimes, I'm OK with that. For example, only examining AFIS quality prints.

Again—Thaks “the documentation should be proportional to the difficulty of the latent print”. I agree. Different types of complexity = different types of documentation. Not sure what you mean by only examining AFIS quality prints.

16 Points
3. If you allow LPE to choose the level and type of documentation are you not losing your transparency shield because some of them will go the fluff route.
Hey, you can fluff any comparison and conspire with your co-workers/supervisors to commit fraud. Hopefully there is some integrity left amongst LPEs. Then the technical reviewer and/or supervisor should be assessing the amount and quality of documentation (they should be doing that anyway).

My aim on this one was that to maintain a high level of transparency one would need in depth SOP’s and lots of documentation. Yes, most of the LPE’s have integrity, but I am not fearful of the most, but the one or two in the barrel. Kind of missed my aim on this question. I will try better next time.

16 Points
4. Do you document the number of friction ridge events in agreement between the two images. A lot do not because of the “No Scientific Basis For A Set Number Of Points”. But if you do not----then you are not transparent.
Wow, people really get carried away with this whole "No scientific basis for a set number of points." The Standardization Committee stated, there was no basis for a point standard "at that time". They also recommended a federally funded study to determine the distribution of minutiae in the population so that we would have some statistically significant value associated with the "rarity of features". Go back and read the report of the Standardization Committee if you don't believe me.

Oh I agree with you. It is just that I know of some examiners in agencies do not record the amount of agreement and I know others that do.

Now “rarity” is a topic I really like. Rarity of features/formations or rarity of relationship, but not everyone agrees. I mean rarity was discussed by Locard and Ashbaugh talks about it, but did you know that SWGFAST does not have it in their glossary and it is certainly not mentioned in their methodology. When I was on SWGFAST I made a suggestion about rarity but was told they did not even want to go there. Their is a book titled “Courtroom Testimony for the Fingerprint Expert” and the author is retired FBI LPE and this comes from page 9 “Question: Is more weight placed on some characteristics than on others? Answer: No. Each characteristic has an equal value. I don’t know about you---but I do not believe that.

16 Points Said

Anyhow, so because there is no point standard, does that mean you didn't use any minutiae to make your comparison? And third level detail is not always present in a latent of value for identification. These IDs tend to be based largely (though perhaps not exclusively) on . . . points. So by indicating what minutiae were used you are actually . . . documenting the basis of your identification? Huh, who would have thunk it? (Some examiners look at "ridges in sequence", myself included. You still use minutiae, otherwise all the ridges would have an identical sequence. You can still count the number of ridges in sequence and/or mark the ridges in sequences used to make the ID).

Hmmm-I think you lost me unless you are talking about people that go into court and cannot give the court an accounting of what they found in agreement---an example I believe is in Haavard vs US. If this is what you mean then I agree. Going into court and not being able to give them an accounting of what you found in agreement is ----well I do not have a word for it other than it would be a negative connotation. If that is not the point you were trying to make, then rephrase it. But I do agree that you should give an accounting in court. I do it all the time.

16 Points Said

When you see FRE during the back and forth, what verbiage do you use to document it? “Well I just did not see it the first time around”, “Well it was hiding in kind of a smudge area”, “Well I just had a bad day this morning—we all have those”. Sure. But to my knowledge, most agencies that annotate latent lifts use a color coded system to differentiate between initial analysis and minutiae marked during comparison.

That is interesting. I know of no agencies that do that. Color coding I mean.

16 Points Said

So I'm confused here. Yes, I do get confused . . . quite easily in fact. If you don't document the analysis and the "back and forth" as you call it, how do you know the method was applied incorrectly? When is it OK to look at a latent impression, say "Look there's some background noise", then look at the known and say "Oh no, that's 3rd level detail . . ." And when is that not OK? Or it's just the subjective opinion of the examiner? How does a tech reviewer know that you didn't mark your 3rd level detail until after seeing the known? Or are you saying it doesn't matter?

Quite simple, we do not conduct technical reviews on the methodology we conduct our technical reviews on the conclusions of the examiner (and the SOP’s developed under the criteria of ASCLD-LAB). To be honest if you were to drop a case file on my desk and ask that I check for methodology, I would not know where to begin. I would check the conclusions---those of non-sufficiency---those of exclusion, those of identifications and those of inconclusive. I guess from your point of view for us it does not matter what the methodology was, the conclusion is the basis for the report that is to be generated. I also do not know of any other agencies in this neck of the woods that check for methodology, they all check the conclusions. Having read several posts on this topic on this forum some would say we get lucky 1,500 a year, but I think we are very reliable. No one has found an error on us yet.

16 Points Said

Again, this process really applies to areas of ambiguity. If you have ridge detail with high clarity, then going "back and forth" doesn't really matter because you can't reinterpret the latent based on the known. It really is just that simple. Good luck to all.

I guess here we have a differing point of view. For you it would be reinterpreting the latent based on the known—for me (and others I have talked to) it is additional information (knowledge) gained through observations. Is that not one of the ways that we gain information and knowledge is by observation? A very in place scientific process. Simple I am not sold on yet but I do wish you good luck.

[16_Points]

I do not see it as changing my “Analysis”, but see it more as an observation that provided me with additional information. If it was not in the same relationship with the other “Points” or “Events” then it would remain just background clutter. It is not the event in and of itself but it’s the relationship with other “Points” that both images share.

16 Points Said

Why? Because this is an excellent example of circular reasoning. I marked the "background noise" in the latent because in the known print I saw an incipient ridge. You basically made my point for me.

From everything that I have read so far, additional information gathered or obtained by observations is not circular reasoning. Generally circular occurs when no new information is added. New information was added here by the observations made. Remember relationship is a major key but not the only one.

This sounds like semantics. You see "background noise" in the latent, but after looking at the known print you call it an incipient ridge. And if there weren't any incipient ridges in the known, then what you saw in the latent would stay background noise. Calling it "new information" based on an "observation" is deceptive at best.

In reality, what evidence do you have that the "detail" in the latent is actually an incipient ridge. Answer, the corresponding area of the known contains an incipient ridge hence the "detail" in the latent print must also be an incipient ridge. This is true ONLY if the latent print was made by the same area of friction ridge skin as the known exemplar (identification). But, if you're using the incipient to support your conclusion of Identification, we have a problem.

You're justification for calling the detail in the latent print an incipient ridge is that the latent and known are from the same source. But you need the incipient ridge to reach your identification threshold. Do you see the problem? (I'll try and draw a picture that illustrates this later). You want to call the incipient in the latent but to do that you need to identify the latent to the known print but you can't identify the print without the incipient ridge. Sounds like circular reasoning to me.

You would have a better chance arguing that you didn't see the ridge detail at all in the latent print. That is, you "missed" it. After looking at the known print, you "found" it in the latent print. Even then, this type of ridge detail, if of reduced quality, carries less weight as evidence of an identification.

There are a lot of examples out there and I have been collecting them for several years. Not to take anything away from Mr. Langeburg as his annotations are good, but I believe the best so far (at least to my liking) is M. Leanne Gray. Her class was full every time.

I have no idea what Leanne Gray teaches.

16 Points Said

When you see FRE during the back and forth, what verbiage do you use to document it? “Well I just did not see it the first time around”, “Well it was hiding in kind of a smudge area”, “Well I just had a bad day this morning—we all have those”. Sure. But to my knowledge, most agencies that annotate latent lifts use a color coded system to differentiate between initial analysis and minutiae marked during comparison.

That is interesting. I know of no agencies that do that. Color coding I mean.

I should clarify that I am aware of agencies outside the US that do this.

Quite simple, we do not conduct technical reviews on the methodology we conduct our technical reviews on the conclusions of the examiner (and the SOP’s developed under the criteria of ASCLD-LAB). To be honest if you were to drop a case file on my desk and ask that I check for methodology, I would not know where to begin. I would check the conclusions---those of non-sufficiency---those of exclusion, those of identifications and those of inconclusive. I guess from your point of view for us it does not matter what the methodology was, the conclusion is the basis for the report that is to be generated. I also do not know of any other agencies in this neck of the woods that check for methodology, they all check the conclusions. Having read several posts on this topic on this forum some would say we get lucky 1,500 a year, but I think we are very reliable. No one has found an error on us yet.

Verification is a separate examination using the same methods to check the conclusions.

Technical review is a check on the initial examiners use of methodology. For example, reviewing a charted enlargement of an identification. Is the ridge detail marked sufficient for identification? Is the ridge detail marked in the latent and known actually in agreement. Are there "red flags" or other problem areas that require notes/annotations/explanation?

If you had contemporaneous documentation of the examination in the case folder, you would know where to begin. No one is asking for a level of documentation that is scientifically unreasonable. If what we do is scientific, then our documentation should participate in that same scientific standard. I don't claim to have all the answers. I only propose that contemporaneous documentation is possible, reasonable, and incalculably useful for safeguarding against contextual bias and providing a basis for comprehensive technical review and quality control.

[kevin]

Technical review is a check on the initial examiners use of methodology. For example, reviewing a charted enlargement of an identification. Is the ridge detail marked sufficient for identification? Is the ridge detail marked in the latent and known actually in agreement. Are there "red flags" or other problem areas that require notes/annotations/explanation?

16,

Do you do a system where there is 1) an examiner, 2) verifier, 3) technical reviewer, 4) admin review (et al)???

[minu]

It seems that distinguishing ridge events found in analysis and comparison is aimed for preventing bias (false interpretation of poor quality print using good quality print). But the bias is more related to the confidence of ridge events and no ridge events, rather than the exact phase (analysis or comparison) when ridge events are found. A ridge event may be marked after comparison simply because there is no necessity to mark it in the analysis stage.

So, why not just mark the confidence for ridge events and no ridge events? Here the confidence is only based on the analyzed image. It can be marked at any phase as long as it is based on the LPEs' consensus on confidence (If the confidence is limited to 2-4 levels, I guess that for most latents, it is possible to find a consensus among LPE).

If there is any ridge event that is vague but important for identification, it is better to mark it in a way different from other clear ridge events. If there is any area in the latent, which seems to have a vague ridge event that is not included in the exemplar, but agreements in other area make you feel this is an identification, it is better to mark that area.

This will be useful for the LPE himself, the technical reviewer, and future AFIS.

I think a technical reviewer (I assume his/her job is to check the proper practice of ACE) should care more the confidence of ridge events and no ridge events, rather than the exact phase where the ridge events are found. A technical reviewer may check if his/her confidence on the marked ridge events by the first LPE is similar to the first LPE, and if there is any area where ridge events are missed.

The current AFIS may use little information on the confidence of ridge events and no ridge events. But the future AFIS will make use of it. Actually, NIST is testing the capability of AFIS vendors in utilizing extended features (such as ridge skeleton, quality map, ridge flow map, etc).

Jianjiang Feng
MSU