Beta
×

Welcome to the Slashdot Beta site -- learn more here. Use the link in the footer or click here to return to the Classic version of Slashdot.

Thank you!

We are sorry to see you leave - Beta is different and we value the time you took to try it out. Before you decide to go, please take a look at some value-adds for Beta and learn more about it. Thank you for reading Slashdot, and for making the site better!

Ask Slashdot: How To Catch Photoshop Plagiarism?

timothy posted about 2 years ago | from the when-google-images-isn't-enough dept.

Education 284

First time accepted submitter jemenake writes "A friend of mine teaches electronic media (Photoshop, Premiere, etc.) at a local high-school. Right now, they're doing Photoshop, and each chapter in the book starts with an 'end result' file which shows what they're going to construct in that chapter, and then, given the basic graphical assets (background textures, photos, etc.), the students need to duplicate the same look in the final-result file. The problem, of course, is that some students just grab the final-result file and rename it and turn it in. Some are a little less brazen and they rename a few layers, maybe alter the colors on a few images, etc. So, it becomes time-consuming for her to open each file alongside the final-result file to see if it's 'too perfect.'" How to look for images closer than they should be to the original? Read on for more details. jemenake continues: "When I first discovered that she was doing this, my first reaction was that there's got to be some automated way of catching the cheaters. Of course, my first idea of just doing MD5 hashes of each file won't work, since most kids alter the file a little bit.

A second idea I had was to alter the final-result file in a way that isn't obvious, like removing someone's shoelace, mis-spelling a word in the background, or removing/adding some dust-specks. (I know map publishers and music transcribers use this trick to catch copiers). But this still requires that she look for the alteration in each file. I'd think that Photoshop, after all these years, would have some kind of scripting language which also supports some digital watermarking, but I've just never dabbled in that realm.

And, of course, I guess another solution would be for her to not provide the end-result file in Photoshop format, but to export it as a flat image. But I'm still intrigued by the notion of being able to "fuzzily" compare two photoshop files or images to find the ones which are too similar in certain aspects (color histograms, where the edges are, level of noise, whatever).

Anybody else have any clever ideas for this?"

Sorry! There are no comments related to the filter you selected.

Invent your own exercises (5, Insightful)

Anonymous Coward | about 2 years ago | (#41970879)

That's what a teacher is supposed to do anyway.

Re:Invent your own exercises (1)

TheMiddleRoad (1153113) | about 2 years ago | (#41970985)

Or just provide a similar base image for all to use. They can still use the textures and extras.

She could try... (5, Informative)

AliasMarlowe (1042386) | about 2 years ago | (#41971603)

Perceptual Image Diff [sourceforge.net] and Find Image Dupes [freecode.com] might be helpful. If she runs finddupes with a threshhold of .99 or so, then it is likely just trigger on nearly exact copies. At least, it should narrow down the ones she has to inspect in more detail. On the other hand, pdiff will detect exact or nearly exact copies by specifying how many pixels are allowed to differ (so it can be fooled by addition of random noise). While pdiff is available for Windows as well as Linux, it seems that finddupes is Linux only.

Re:Invent your own exercises (-1, Troll)

InvisibleClergy (1430277) | about 2 years ago | (#41971423)

Just like how Computer Science professors should all invent their own languages so that students can't just copy solutions from the internet?

Re:Invent your own exercises (-1, Troll)

Anonymous Coward | about 2 years ago | (#41971529)

Bad analogy, faggot. Your analogy would be appropriate if he implied that teachers should all develop their own image editing software. We do expect CS professors to write their own assignment problems.

Re:Invent your own exercises (1)

K. S. Kyosuke (729550) | about 2 years ago | (#41971535)

Analogy FAIL. The equivalent in the arts to what you're describing would probably be inventing new art styles. Say hi to post-pointilism and surrealist-mannerism.

Re:Invent your own exercises (1, Insightful)

im_thatoneguy (819432) | about 2 years ago | (#41971479)

Or view exercises for what they are... exercises. The test/final project is what it is. Maybe your students don't need the exercises because they already have a strong grasp of the task.

Re:Invent your own exercises (1)

Anonymous Coward | about 2 years ago | (#41971517)

Or require students to also hand in the intermediate steps for the homework just like old school math.

She should know this if she's teaching photoshop (5, Interesting)

Anonymous Coward | about 2 years ago | (#41970889)

Paste images into the source image as new layers, adjust layer mode to "difference" and look for the similarities. Done.

Re:She should know this if she's teaching photosho (2)

trnk (1887028) | about 2 years ago | (#41971485)

Upvotes if I had them - this is exactly what the difference blend mode does. You could even record the flatten/paste original/adjust blend mode/save as jpeg operation as an action run it on the folder of student images using the batch processor to produce a nice little set of comparison images all at once.

Re:She should know this if she's teaching photosho (2)

3dr (169908) | about 2 years ago | (#41971491)

Bingo. Pixel differencing will show which pixels...are different...which will show gradient differences (as broad areas of different pixels), layer positioning differences (as lines), etc. Otherwise, I think it's pretty obvious that one does not provide the students with a final .psd with all the layers intact. At the least, any provided file should be a flattened version (PNG or JPG of decent quality) with a watermark... There's just so many ways to thwart this. Does the PS instructor *know* image manipulation?

Final result should just be a flattened image? (4, Insightful)

Anonymous Coward | about 2 years ago | (#41970893)

Why not just flatten the final result into a simple image? The students can still see what the end result is supposed to look like, but they obviously can't just hand in that file.

Re:Final result should just be a flattened image? (1)

Joce640k (829181) | about 2 years ago | (#41971029)

Why not just flatten the final result into a simple image?

You know how I know you didn't even bother to read the entire question...?

Re:Final result should just be a flattened image? (2)

green1 (322787) | about 2 years ago | (#41971237)

Just because it was posted as part of the question doesn't mean it's not a good answer. It certainly seems to be the easiest, and most effective way to detect the problem quoted.

MOD PARENT UP! - Re:Final result should just be (5, Insightful)

corychristison (951993) | about 2 years ago | (#41971121)

Why not just flatten the final result into a simple image? The students can still see what the end result is supposed to look like, but they obviously can't just hand in that file.

Offer flat JPG in medium quality as an "end result". Maybe even include a digital metadata watermark?

Require high quality JPEG and PSD for assignment. First check for metadata watermark, then compare quality of JPEG. If it looks too close then open up the PSD and check the layers.

Re:Final result should just be ascii (1)

maestroX (1061960) | about 2 years ago | (#41971507)

Easy to diff, easy to see!

Use a different starting point (0)

Anonymous Coward | about 2 years ago | (#41970903)

Why not start with a different image? For that matter each kid could start with a different image so they can't cheat off each other. Harder to grade, I know, but that's why teachers get paid the big bucks :-)

compare the files (0)

Anonymous Coward | about 2 years ago | (#41970921)

You can always run a binary compare of the files using a diff tool. Anything the students do will show up as a delta. So if there are no deltas, the students copied the original or are creating a pixel perfect copy.

Don't Give Them The File? (2, Insightful)

Anonymous Coward | about 2 years ago | (#41970927)

How about simply not giving them the final file? Why not a printed copy?
If they must have an electronic version of the picture give them a low res thumbnail version.
Project the image on a screen and tell them to draw that.

Your problem is that you are over thinking the tech angle when low tech methods will be super effective.

Also, ErrorLevelAnalisys (0)

Anonymous Coward | about 2 years ago | (#41971035)

This site does a good job of showing you what has been manipulated in an image.
www.errorlevelanalysis.com [errorlevelanalysis.com]

Re:Also, ErrorLevelAnalisys (1)

Yaur (1069446) | about 2 years ago | (#41971261)

they have been offline for some time now...

simple solution (5, Informative)

wbr1 (2538558) | about 2 years ago | (#41970937)

distribute the final file as a watermarked png only. Require assignments to be turned in a multi layered psd files. Problem solved.

Re:simple solution (1)

MadCow42 (243108) | about 2 years ago | (#41971095)

Or if they want to distribute the final file for reference purposes, simply distribute it at lower resolution than what you require the students to do. Problem solved.

MadCow.

Re:simple solution (0)

Anonymous Coward | about 2 years ago | (#41971123)

distribute the final file as a watermarked png only. Require assignments to be turned in a multi layered psd files. Problem solved.

What is to stop the kids doing that as a final step, and giving their assignments to others, who also add the watermark as a final step? You need to think before you suggest something.

Re:simple solution (2)

green1 (322787) | about 2 years ago | (#41971319)

The stated problem was how to stop people from copying the example file and handing it in as their own work. The suggested solution solves that problem perfectly. (unless you think it's easy to de-watermark a png and then separate it out in to believable layers in a psd file? (or at least easier than just doing the assignment))

How to stop students from cheating by copying each other is a completely different problem, and luckily, not what the submitter asked. (because there really isn't a "good" way in that case...)

Re:simple solution (3, Interesting)

ColdWetDog (752185) | about 2 years ago | (#41971393)

This should work and would be trivial to do.

If you want to spend more time on it, make sure that the student's copy of Photoshop is set to record history in the metadata.

Then you can go through and look at every step they made. I do this on some images so I can figure out what the hell I did to get that effect three years later. It takes up little space - it's just text.

But a more boring way to spend a day would be hard to create.

Re:simple solution (0)

Anonymous Coward | about 2 years ago | (#41971475)

If they're Asian, scrutinize it to be sure they did their work. Let the white kids do whatever they like. Offer to help the Black and Latino kids do it or tell them they can just hand in the final copy if they need to. If they have a learning disability, full marks for just showing up to class.

That would apparently work in Virginia.

Extra step (1)

Anonymous Coward | about 2 years ago | (#41970941)

Add an extra step before or after a specific step in the process. Make it unique based on each students name, or student id, or something so it cant be copied and shared. That what iv had happen in a similar class.

Don't give out the answers (2)

Plasmoid (8367) | about 2 years ago | (#41970949)

Provide the book resources as a tutorial but get the students to do something different for the actual assignment. It could be as simple as swapping a few textures or effects. A blur on a cat will look very different to a blur on a dog even though the technique is the same.

Created Date (0)

Anonymous Coward | about 2 years ago | (#41970959)

Look at the Created Date on the file properties and see if it matches the "end result" created date.

Re:Created Date (2)

Joce640k (829181) | about 2 years ago | (#41971069)

Not a bad idea. That would probably catch most of them...

High resolution (0)

Anonymous Coward | about 2 years ago | (#41970961)

Just ask for a high resolution uncompressed result. This works well if the original is a 96 dpi jpg, and is compressed.

Instead (3, Interesting)

Sparticus789 (2625955) | about 2 years ago | (#41970965)

She could also not show the students a picture of the final project. She could just give them a list like:

1. Remove one set of shoelaces.
2. Add a bird in the sky
3. Add a portrait of Spock in the background.
etc.

Re:Instead (3, Funny)

Anne_Nonymous (313852) | about 2 years ago | (#41971171)

More like:

1. Remove the stretch marks
2. Make the boobs bigger
3. Get rid of that wart thing with the hair growing out of it.
etc.

Re:Instead (4, Funny)

dmacleod808 (729707) | about 2 years ago | (#41971189)

See this is where you went wrong. Ever project should ALWAYS include a portrait of Spock in the background, anything less is just not logical.

Re:Instead (4, Funny)

93 Escort Wagon (326346) | about 2 years ago | (#41971515)

It certainly jazzed up our family reunion photo!

suggestion (1)

Tastecicles (1153671) | about 2 years ago | (#41970969)

supply the desired end result to the students in hardcopy, ask for their results in electronic format. Oh, and hide the SCANNERS!

FindImageDupes (5, Informative)

Ken_g6 (775014) | about 2 years ago | (#41970971)

From the manpage:

findimagedupes [jhnc.org] compares a list of files for visual similarity.

To calculate an image fingerprint:

          1) Read image.
          2) Resample to 160x160 to standardize size.
          3) Grayscale by reducing saturation.
          4) Blur a lot to get rid of noise.
          5) Normalize to spread out intensity as much as possible.
          6) Equalize to make image as contrasty as possible.
          7) Resample again down to 16x16.
          8) Reduce to 1bpp.
          9) The fingerprint is this raw image data.

To compare two images for similarity:

          1) Take fingerprint pairs and xor them.
          2) Compute the percentage of 1 bits in the result.
          3) If percentage exceeds threshold, declare files to be similar.

Of course, you shouldn't take its suggestions at face value every time, but it should help narrow your search for cheats.

Re:FindImageDupes (2, Insightful)

Anonymous Coward | about 2 years ago | (#41971223)

Given that the students are supposed to reproduce the images, I guess they will get a high visual similarity, unless they failed.

Re:FindImageDupes (0)

Anonymous Coward | about 2 years ago | (#41971321)

Your fingerprint space is 256 images.

Google search 101... (1)

Anonymous Coward | about 2 years ago | (#41970973)

http://pixelnovel.com/comparepsd/
first result I found... and it looks to be free.

Re:Google search 101... (0)

Anonymous Coward | about 2 years ago | (#41971593)

THis is slashdot. You'll only see ask slashdot questions like,

What VM software do I use?
How do I search the internet for a PSD comparison tool?

This site is one big epic troll.

OP already has the answer? (2)

Zironic (1112127) | about 2 years ago | (#41970977)

You already mention the solution, why be silly about it? Just watermark the images and hand them out as jpegs, not photoshop files.

You could obviously watermark each individual layer if you wanted to give the photoshop files, but why would you want to do that?

easy (0)

Anonymous Coward | about 2 years ago | (#41970979)

ask them to do something similar with *other* pictures that they find on the net, then assemble.

other questions ?

Learning vs Working (1)

dloolb (159254) | about 2 years ago | (#41971013)

The students appear to be treating the class as work, instead of a learning experience.

"The problem, of course, is that some students just grab the final-result file and rename it and turn it in."
Well, those students will not do well on a test designed in the same manner using a set of teacher supplied images.

Personally, I think the teacher is trying to do too much, if she is comparing images looking for small differences. Let the test tell you who LEARNED and who didn't.

Simple Solution (4, Insightful)

hubang (692671) | about 2 years ago | (#41971017)

The solution is simple:

Give a token homework grade (like ~ 10%) for participating and make everything in the final grade else be based on original projects and tests. Make the students use given files.

Then, if they cheat, they only cheat themselves.

Re:Simple Solution (1)

HornWumpus (783565) | about 2 years ago | (#41971149)

Came to say this. Don't even collect homework/classwork. Give them 10% for attendance. Don't take attendance. Just give them 10%

If they can't do it at the end, they will fail. If they didn't ask for help, tough. Treat them like adults. They will ether rise to it or learn what doesn't work.

Re:Simple Solution (3, Interesting)

Anonymous Coward | about 2 years ago | (#41971463)

"If they can't do it at the end, they will fail. If they didn't ask for help, tough. Treat them like adults. They will ether rise to it or learn what doesn't work."

While that is a fair approach, I've found that I get much higher success rates in the end if I have some kind of feedback during the period they are working. If they don't *know* they're way off track, or don't *know* they aren't putting in enough effort, then, sure, they'll fail at the end like they should, but there will be some missed opportunities where honest and interested students would have increased their effort had they realized the course demanded more.

Inevitably, there will be some students that are incorrigible. I can't do much for those. But a mid-term "reality check" evaluation worth, say, half as much as the final evaluation, is really useful. It also means that students who don't want to put in the required effort can bail out early (which means I don't have to waste my time with them the rest of the term either -- a win-win as far as I'm concerned).

Re:Simple Solution (2)

supercrisp (936036) | about 2 years ago | (#41971553)

As a teacher, I'd like to say that this method really, really worked for me at a major midwestern university. Then I moved to the South and tried the same method; it did not work. I'm not saying it's a regional difference, perhaps just admissions policies. But I'm at a second Southern university now, and I'm surprised students even wipe themselves, as they'll do little else if they don't receive a grade for it. Maybe this teacher works with similar students, ones for whom only high-stakes grading is sufficient motivation to lift a finger. (Can you tell I just had to sit thru a long faculty meeting? I'm pissy as all get-out.)

Re:Simple Solution (0)

Anonymous Coward | about 2 years ago | (#41971379)

Bingo! The problem is with trying to grade something where the performance is irrelevant.

Test the students instead.

If you can't figure out how to do that, stop pretending to be a teacher.

Ummmmm (5, Informative)

cultiv8 (1660093) | about 2 years ago | (#41971033)

2 mins of googling and I found this: ComparePSD [pixelnovel.com] .

ComparePSD compares two Adobe Photoshop PSD files for you and highlights the differences. Layer by layer. Effect by effect. Simple. And did we mention that ComparePSD is absolutely free?

Re:Ummmmm (1)

Maximum Prophet (716608) | about 2 years ago | (#41971301)

This is good but still time consuming.

There are other algorithms that work on the final result to tell you how much of the photo was changed.

Since she has the originals, something like this would work: Auto compare every final picture with its original and produce a number, percent changed. Then use ComparePSD to compare the submission that are the same % changed.

The problem is that if she is too specific in her instructions for the assignment, then everyone in the class that tries will have the same changes. If she's too inspecific, then someone can claim they deserve an 'A' for work that's not what was intended.

Of course, once the students wise up to this, they will start using ComparePSD to compare theirs friend's work to the modified copy, just to make sure that there are enough changes.

Pop Quiz! (0)

Anonymous Coward | about 2 years ago | (#41971047)

Next time, just give them a pop quiz. Give them a base image for them to work on and do not give them the "end result" photo. Show on a projector and screen what the final image should look like without naming the tools, etc.

Give them 1 hour to do it.

That ought to ween out the cheaters from the do-gooders.

Good Students' Results (1)

Anonymous Coward | about 2 years ago | (#41971051)

As an OCD overachiever, I'd like to recommend the instructor not select a test protocol which would make the students who make an effort to duplicate the outcome on their own end up pooled with the students who know the rename command.

Perhaps you could require saved copies of the files on the way towards the final product from each student.

Complete photoshopped multi-layer works don't come out of nowhere; you could ask each student to save one copy of their work at the end of each section they are following in the text.

None of those partially completed works exist for the students who use 'rename;' all exist for the students who are doing the assignment.

Print out each set of data as thumbnails on a sheet of paper so you can glance them over.

Posting anon since I'm already a "cheater" because my work is "too good," but I'd rather not have others with talent suffer the same indignities.

Complete project? (0)

Anonymous Coward | about 2 years ago | (#41971059)

Don't hand out the completed project. You don't really get any where else in education.

history (1)

Tsu-na-mi (88576) | about 2 years ago | (#41971063)

A quick review of the history should tell you a lot. If it's exactly duplicated with a few tweaks at the end, it's a dead giveaway.

Re:history (4, Informative)

ColdWetDog (752185) | about 2 years ago | (#41971453)

Yep. PS does include a history function that can be written to the metadata. Bonus points for pulling out the metadata stream, running some regex on it and deciding if it was legit.

Extra bonus points for not including the history in the image given to the students. And requiring it for a grade.

Easy: Have each student sign original image (0)

Anonymous Coward | about 2 years ago | (#41971067)

Have each student watermark the original image with their name and flatten the image. Their name will have to have passed through all of the transformation steps to look right on the final image. Each student's end result will then be unique.

What? (1)

smith6174 (986645) | about 2 years ago | (#41971079)

What are you looking for, some fancy machine learning algorithm to compare the files? Just make the assignments something that doesn't have an answer available and be done with it.

Re:What? (1)

Maximum Prophet (716608) | about 2 years ago | (#41971373)

I think the problem is that while the assignment might be new, everyone in the same class gets the same original file, and has to make changes to the same specifications. I'd be supprise if some students didn't legitamately have the same end result.

One poster suggested that each student gets a slightly different original file. Something like different house numbers, but in the same font and color. If two students submit the same house number, you know they copied.

Simple (0)

Anonymous Coward | about 2 years ago | (#41971083)

Don't grade the students around the chapter images. Create a separate image using the same techniques and print it out on a sheet of paper. Students must recreate it, using what they've learned.

Stenography? (0)

Anonymous Coward | about 2 years ago | (#41971093)

Perhaps embed some stenography in each layer of the image? Not sure if that's possible with photoshop though. But going under that theory (if it's possible), you could just run each file through the same program you used to add the stenographic text. If it opens any text, they cheated.

Unfortunately, the only program I know of that adds stenography is that slight changes anywhere in the image will wipe it out, so they would need to outright cheat.

Could you not just give them a printout of what the end result should look like? Generally speaking, scanning a printed image will never look as 'clean' as the original. Even if they have it ridiculously high quality scanned, they'd still have to resize it or at least adjust the file size somehow so it's not 10 megs.

Re:Stenography? (1)

queequeg1 (180099) | about 2 years ago | (#41971385)

Steganography?

grade em... (0)

Anonymous Coward | about 2 years ago | (#41971101)

So, it becomes time-consuming for her to open each file alongside the final-result file to see if it's 'too perfect.'

There are several ways to mitigate that. In class tests. Custom assignments for each student.

You are NOT going to get away with 'auto grading' for something like that, at this time. Auto grading works good for multi-choice, and t/f. Not so much where there is any sort of 'creative' input. You are just going to have to knuckle down and grade each one.

Have each student do something unique. Such as having to put their name in the pic somwhere with a particular effect...

Statistical Answer (0)

Anonymous Coward | about 2 years ago | (#41971105)

Create a script which accepts two images of the same size and resolution.

For each pixel, there are 3 RGB values from 0 to 255

Find the euclidian norm of the variance of the two pixels with this equation:
R1 = red in pic 1, R2 = Red in pic 2, G1 = green in pic 1, G2 = green in pic 2, B1 = blue in pic 1, B2 = blue in pic 2

Sn = sqrt((R2-R1)^2+(G2-G1)^2+(B2-B1)^2)

Now, find the sum of all Sn (one value for each pixel.) That will be the images "Variance Score".

Graph all the variance scores on a bar chart with bins of appropriate sizes. It should quickly become apparent who is cheating.

Use an XOR filter (0)

Anonymous Coward | about 2 years ago | (#41971113)

Flatten the student's image.
Take the reference image and add it (flattened as well) as another layer.
Set the blend type of the layer to XOR.

If the image is identical or near identical, you'll either get black, or a pretty uniform field of specks where compression noise or whatever has taken place.

If the image is different, you'll see brighter areas wherever the differences between the images are.

If the student copied the image, it will be very similar. If they copied the image and modified it slightly, you'll see the modifications. If they copied it and applied filters to the whole image, you'll see bright areas across the entire image, but not so much locally.

If the student did the assignment correctly, you would expect to see bright areas around the parts of the image that were supposed to be added/changed by the student, and relatively little around the source area.

If you could write a script that did that process, (takes flattened student image, takes flattened answer image, outputs an image that is the xor of the two images) which I'm sure is possible, you could just scroll through the output images until you found one that looks suspicious and then look at it in more detail.

I wish... (0)

Anonymous Coward | about 2 years ago | (#41971115)

I wish someone had taught us Photoshop or any other useful 21th century stuff in high-school.

Re:I wish... (1)

ColdWetDog (752185) | about 2 years ago | (#41971511)

I wish someone had taught us Photoshop or any other useful 21th century stuff in high-school.

So do the rest of us who have to look at the awful Photoshop [psdisasters.com] stuff that floats around the Internet these days.

Personal experience. (1)

alexbgreat (1422591) | about 2 years ago | (#41971119)

Don't provide the end result, only the resources to create it. Print out all of the work. Printing is important because: a) it gives you a physical representation of the work, signed by the student. b) it allows you to thumb through them quickly and spot the duplicates (read: cheaters) c) allows you to prove this cheating relatively easily to administration, and give examples of non-cheating to compare to. Even though the students are all working towards a common goal, every image will come out different. Colors won't be exact, positions different, cut-lines different. Every image has a "signature", which makes duplicates and highly similar images (read: cheating with obfuscation) stand out when you physically look at each set of images individually.

Picking out the cheaters is much easier than it seems. Don't overthink this.

Resultion is your freind. (0)

Anonymous Coward | about 2 years ago | (#41971125)

I would suggest that she only provides a lower (at least a 1/4) resolution file to her students while requiring a full resolution on tuned in projects. Also adding compression to loose detail is another trick. This should make the copies easy to spot. Speaking of which why is she giving them layered photo shop files instead of a small jpg. They should figure out what layer to make on their own.Also, I never get any requests to Photoshop that come with more then a written or verbal description. Usually I do a quick proto-drawings if there is potential for confusion. But maybe she is start them off with a monkey see monkey do type learning. She could also hide small color dots in layers or use patterns with slight differences to create water marks even rotating them 90 degrees would work. Hell why not just use the water mark plugin or draw rough strips over the picture. That won't stop cross-student coping though. However, she could even give different students different patterns to make sure they are doing the work themselves.

Crop the example result files (1)

JazzHarper (745403) | about 2 years ago | (#41971143)

Students must turn in the full image. Much simpler than watermarking.

Reverse Image Search (0)

Anonymous Coward | about 2 years ago | (#41971165)

Reverse image searching can find images 'like' another image. Try: http://www.tineye.com/

Meta data (1)

Anonymous Coward | about 2 years ago | (#41971199)

1. In Final.psd go to: File > File info and under keywords type: iamacheater (or any other unique string)
2. Distribute file
3. Receive student files and import in Lightroom
4. Search for string iamacheater to find cheaters

wrong problem (0)

Anonymous Coward | about 2 years ago | (#41971211)

if she doesn't know, perhaps she shouldn't teach photoshop ...

be smarter (1)

rubycodez (864176) | about 2 years ago | (#41971213)

for one assignment students turn in their own pictures with themselves in it near some assigned object. Later, another assignment has them work with their picture toward some given result on the object, with them still in the picture.

Re:be smarter (1)

Maximum Prophet (716608) | about 2 years ago | (#41971403)

for one assignment students turn in their own pictures with themselves in it near some assigned object. Later, another assignment has them work with their picture toward some given result on the object, with them still in the picture.

It's a good chance that the more photogenic students will get the highest grades. There are several studies about grad student grant proposals that back this up.

So (0)

Anonymous Coward | about 2 years ago | (#41971217)

[quote] So, it becomes time-consuming for her to open each file alongside the final-result file to see if it's 'too perfect.[/quote]

Script it. Compare the layer names, do a CRC check on the layers themselves.

I'm sure such a thing already exists.

Subtract (0)

Anonymous Coward | about 2 years ago | (#41971245)

Subtract turned in image from the Demo result.
Average the result if its 0 its the same, if its less than some critical value then have a look see whats different.

you might be able to use OpenImagIO (https://sites.google.com/site/openimageio/) to script it, not sure how it might deal with layers.

You in denial? (-1)

Anonymous Coward | about 2 years ago | (#41971255)

You wanna fight god, too? Fucken jews.

God says...
Go, carry them to the city.

20:41 And as soon as the lad was gone, David arose out of a place
toward the south, and fell on his face to the ground, and bowed
himself three times: and they kissed one another, and wept one with
another, until David exceeded.

20:42 And Jonathan said to David, Go in peace, forasmuch as we have
sworn both of us in the name of the LORD, saying, The LORD be between
me and thee, and between my seed and thy seed for ever. And he arose
and departed: and Jonathan went into the city.

21:1 Then came David to Nob to Ahimelech the priest: and Ahimelech was
afraid at the meeting of David, and said unto him, Why art thou alone,
and no man with thee? 21:2 And David said unto Ahimelech the priest,
The king hath commanded me a business, and hath said unto me, Let no
man know any thing of the business whereabout I send thee, and what I
have commanded thee: and I have appointed my servants to such and such
a place.

If you must provide the end result... (0)

Anonymous Coward | about 2 years ago | (#41971281)

Have the students record the action of them creating the final result and have them play it back to recreate the project.

You can cut the resolution of the sample by half, compare dimensions of the end result. (if they up sampled their image you should be able to tell very easily).

Similarly, you can compress the heck out of the end result file and then compare file sizes of their work.

CS6 has digimark digital watermarking capability you could use on it and then write a simple script to check for the watermark.

Convert the demo to a different color space/bit depth (they probably won't realize it is different) and compare.

Change all of the elements in the project file to something silly (change colors and pictures).

I think .psd supports metadata, put something in the metadata that says "I cheated".

You could just SAY that you have some crazy cheat-catching checking capability and assume the kids will take your word for it. (Then just check the smart kids)

Re:If you must provide the end result... (0)

Anonymous Coward | about 2 years ago | (#41971335)

Most of these would be simple to just check all the files at once, look for the odd-ball. But really, most teachers know that the trick to grading is not to grade every thing. Just pick a handful to keep the class honest.

Alter the assignment slightly (0)

Anonymous Coward | about 2 years ago | (#41971317)

The instructor should add a unique instruction to each assignment. I.e. if the task is to change the text on a billboard in a photo, ask them to put different text in, rather than what's referenced in the text. The rest of the tutorial remains the same. Or if they are changing the color of a bird, ask them to make the beak purple instead of green, etc.

This way the people who don't pay attention and don't follow the steps will have the wrong result, but the rest of the assignment is the same. A clever instructor should be able to pick a step to change that is hard to work around by altering the example.

Not viable (0)

Anonymous Coward | about 2 years ago | (#41971325)

http://www.codeproject.com/Articles/374386/Simple-image-comparison-in-NET [codeproject.com]

There seems to be some resources available online, if you write some code and run your images through this you should be able to quickly assess the differences. But IMHO you should just get the students to do the assignments on different images. Perhaps even individual assets for each of them. If I were one of them, I'd just blur the image ever so slightly, somewhat mess with the levels and crop a few pixels off the edges and no matter how good your algorithm, it will fail to recognize plagiarism. On the other hand, perhaps the sneaky little bastards that go that far deserve to pass? ;) In an unrelated note, as a pretty experienced Photoshop user, I find it that the ability to creatively choose and apply the right tool to obtain a desired effect is very valuable, and as such, I generally don't think much of those step-by-step tutorials.

The first question I would raise is... (0)

Anonymous Coward | about 2 years ago | (#41971331)

Is this a class on Using Photoshop or is this a class on Digitially editing and enhancing photos? If the class is specific to Photoshop, then distribute the final product in a flat file for comparison. Otherwise, teach the students how to use tools to enhance their own photos. In this day and age, every student should have access to digital photos of some sort, either from their own devices (camera, cell phone, etc), or online.

My son ran into a problem with one teacher in a high school web design class. The classroom systems were preloaded with Dreamweaver. The class was basic web design. My son chose to use a simple text editor instead, and was told he would fail if his homework assignments weren't done in Dreamweaver. When I pointed out to the teacher that the program cost ~$900, he changed the requirements. If the class had been specific to Dreamweaver, then I would have asked the teacher to consider the costs involved when assigning homework. My son ended up getting his final class project on the school web site - it was an html/css/javascript page that allowed other students to create simple web pages, complete with formatting and colorization via popup color wheel. All written in Notepad. That was 2004.

Make answering this question their 1st assignment (1)

davidwr (791652) | about 2 years ago | (#41971367)

Don't make it a big part of their grade. After all, you aren't teaching forensics here.

But by making copy-detection an early-on element of the course you'll make them aware that you know how to detect copies.

Also make it clear early-on that this class is designed to teach them a useful, marketable skill and that if they cheat, they won't have learned the skill and if enough of them cheat and don't get caught, YOU won't know to slow down the pace of instruction. As a result, the whole class may "pass" knowing a lot less than they would if nobody cheated.

Re:Make answering this question their 1st assignme (1)

Maximum Prophet (716608) | about 2 years ago | (#41971447)

Also make it clear early-on that this class is designed to teach them a useful, marketable skill and that if they cheat, they won't have learned the skill and if enough of them cheat and don't get caught, YOU won't know to slow down the pace of instruction. As a result, the whole class may "pass" knowing a lot less than they would if nobody cheated.

Some artists/journalists would argue that *any* use of photoshop is cheating. Albeit, it's not copying, but you are bending reality. (Not that photo manipulation is new to photoshop. Trick photography started just a few years after photograpy itself was invented, and even realist painters didn't paint the "real", reality)

Re:Make answering this question their 1st assignme (1)

davidwr (791652) | about 2 years ago | (#41971559)

Many if not most of Photoshop's uses aren't in journalism per se, they are in fields like marketing, advertising, and other places where the viewer isn't expecting to see an authentic recording of a real event.

Think of photo-editing tools like this the same way you think of the tools a recording studio sound engineer uses in after the recordings are made but before the final master is declared final.

Criteria (2)

Firewheels (252266) | about 2 years ago | (#41971401)

> So, it becomes time-consuming for her to open each file alongside the final-result file to see if it's 'too perfect.'"

How is it that she's grading these? One would assume the grade depends on similarity to the target image or the layers embedded in the file [1] which should be dependent on comparing the student file against the master.

Or is it yet another "Best try" scheme? "You tried, Timmy, so I give you an 'A'".

[1] Does Photoshop embed the history inside the file? It's been awhile since I've worked with it.

Flatten The Example. (1)

TheSpoom (715771) | about 2 years ago | (#41971407)

And, of course, I guess another solution would be for her to not provide the end-result file in Photoshop format, but to export it as a flat image. But I'm still intrigued by the notion of being able to "fuzzily" compare two photoshop files or images to find the ones which are too similar in certain aspects (color histograms, where the edges are, level of noise, whatever).

If you provide the kids with the end result, and they need to turn in an end result for grading, you're fighting a losing battle. I could personally get around everything you did to try to protect your "example" PSD, and I'm relatively certain that I could have done so at the age of your students. Just give them the flattened image, it's enough for them to see what it should look like.

You should still, of course, use one of the methods mentioned in sibling posts to compare submissions for too much similarity since you'll inevitably have a group of students who figure they should "pool their resources" and submit the same PSD. A quick way to get around the laziest kids would be to md5sum every submission and flag those with the same md5sum. Don't immediately accuse the kids, just make them do the same exercise in person, in front of you. If they can't, they were probably cheating.

Computer Science and programming professors have been dealing with these issues for years; perhaps seek out one of them and ask how they do code submission grading. There are a lot of similarities.

Do what I do for my textures (2)

s13g3 (110658) | about 2 years ago | (#41971435)

Do what I do for my textures, and embed a "watermark" of your signature or something similar deep into the final image where it can't / won't be seen by anybody who doesn't know where or what to look for, in multiple places where the pixels are conducive to such masquerading. It's almost a form of steganography, where the message to be sent is a verification of the authors' identity and claims of original work.

I do mine in such a way that even if I leave one such image that can be readily seen, there are at least a half dozen more than cannot be found without a side-by-side comparison of source and production images with and without the "watermarks" (impossible without someone getting hold of my .PSD's). Keep the true "source" .psd for yourself, create another for disbursing to students that contains several "watermarks" with an extreme level of transparency well-blended into many or all of the layers so they'll have an example .psd to "reverse engineer", and then separately give them the actual un-watermarked original source images, which they should then be expected to chuse to assemble the final image themselves. You might even put an entirely separate watermark into the source images, so you can check to see which watermarks the submitted image has, as opposed to checking only for the source mark.

If they put in enough time and effort to actually successfully circumvent this technique by finding and either eliminating or duplicating all the various marks, then they've probably got the requisite skills to pass the original challenge... at least if you do it the way I do.

My "signature" is in at least 3 places in this image [photobucket.com] , buried deep in different layers with heavy transparency masks, and it would have to be altered drastically to be guaranteed to remove all traces of it.

Software (1)

craigminah (1885846) | about 2 years ago | (#41971457)

A few years back I used iPhoto and it had facial recognition software built-in. When I went through training it, it mis-labeled faced but it did so along family lineage. For example, it would think my dad was me or vice-versa.

I currently have a program called PhotoSweeper (http://overmacs.com/photosweeper/) which uses five different methods to find duplicate images. It doesn't use facial recognition but instead it compared the bitmaps and/or histograms with a user-changeable threshold (e.g. identify really close matches or kind of close matches). Very accurate and it would work if you had flat images without layers.

A Different Approach (1)

Kevinoid (931628) | about 2 years ago | (#41971461)

I disagree with the approach of flattening, printing, or otherwise destroying information in the final-result file, because there can be a lot of learning value for the students in having the solution. The approach of manipulating the image in some way and attempting to detect that modification in the result could work, although it seems like a lot of effort.

Instead, could you require the students to submit the intermediate results as well? That way you have more evidence that the students actually performed the steps. Also, if there is any variation in the steps, it gives you more information about possible copying between students if all intermediate results are the same in addition to the final result.

Not an answer (0)

Anonymous Coward | about 2 years ago | (#41971473)

This doesn't answer your question as posed, but she shouldn't have made available the final xcf/psd, just a flattened export of it. Either way, the kids are cheating themselves if they don't make an attempt to do it from scratch, and it'll show when she tests them on how to apply the principles used to recreate the image.

If she still wants to give them the final image in that format, then she could create the provided pieces and provided final image to differ unnoticably, and then tell the student they may not use any of the parts of the provided final image. If her invisible watermarks / patterns of slightly off-colored pixels show up / etc, then they took a layer from the final and lose credit.

Another way could be to use tools they don't have available to create a unique or advanced effect, and if they're able to replicate it, then you need to look a little closer to determine whether they were clever in a good or bad way.

If she can check levels and stats to compare the images, so can they, and they'll probably be tweaking theirs until they match hers, so that's probably not the best way to detect fraud if she doesn't want to alienate the perfectionists and over achievers. I'd say ruling out the possibility of copying would be the simplest option, but perhaps she has her reasons for not doing that?

Difference Filter + Histogram for Numeric Measure (1)

AaronLS (1804210) | about 2 years ago | (#41971477)

You can use a difference filter, which will produce a ratio of dark/light based on the amount of difference, and then a histogram to get a more quantitative view of how much of the image is different, rather than analyzing with your eyes. You could do this on both flattened composite comparison, as well as layer by layer(maybe trying all combinations of layers and picking out the X number that are closest, where X is the number of layers in the final image). I'm not sure what the capabilities of Photoshop plugins are, but this seems like it could be built to create something more quantitative. Then the instructor could sort them and eyeball the ones with the highest "closeness" scores. Any positives of potential plagiarism should be verified carefully by the instructor though.

ah, it's " time-consuming" (0)

Anonymous Coward | about 2 years ago | (#41971483)

very typical teacher attitude - it's "time-consuming" to examine every image. God forbid she earns the money she's paid.
The "problem" is not students cheating but most likely they are not being taught in class - sit here for 40 minutes and
do boring exercises in the book. That's not teaching, that's babysitting.
Photoshop is about creativity and not about repeating steps in a book. Let students pic their own subject matter
and work on images until they get them looking right.
Of course they need to know all of the basics and build on top of that -
if they gave up learning the basics, there's something seriously wrong with the book, teacher or maybe even students' attitudes.
I wouldn't worry about cheating since bullshit will eventually catch-up to them in college, if they make it that far.

Introduction to Time Lapse (3, Interesting)

mrbene (1380531) | about 2 years ago | (#41971505)

It's probably annoying for all involved, but just like the "show your work" in math classes, you can request a "show your work" equivalent via screen-cast. And the students will learn a bit about screen-casting.

Alternatively, request a picture of each step.

Easy (0)

Anonymous Coward | about 2 years ago | (#41971539)

Show them the image in Photoshop and tell them to recreate it in Gimp.

Why are schools teaching people specifically to use Adobe software anyway?

Cheaters don't prosper (0)

Anonymous Coward | about 2 years ago | (#41971569)

Or... spend the time doing marking the existing very slow way (perhaps even with the shoelack suggestion), but completely fail any student who cheats, make them an example with lots of detention.

Word will soon get around!

*Sometimes the simplest answer is the right one.

Practical Exam (0)

Anonymous Coward | about 2 years ago | (#41971599)

We added a "Practical" section to exams in our media classes. basically the teacher (And proctor) would go around to each student and have them demonstrate the process that was in question. If the student had immaculate homework but demonstrated zero ability in class, we would have a discussion.

Load More Comments
Slashdot Login

Need an Account?

Forgot your password?