Slashdot Mirror
Ask Slashdot: How To Catch Photoshop Plagiarism?
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?"
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?"
That's what a teacher is supposed to do anyway.
Paste images into the source image as new layers, adjust layer mode to "difference" and look for the similarities. Done.
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.
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.
distribute the final file as a watermarked png only. Require assignments to be turned in a multi layered psd files. Problem solved.
Silence is a state of mime.
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.
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.
You don't exist. Go away. --SysVinit Halt
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.
sudo make me a sandwich
supply the desired end result to the students in hardcopy, ask for their results in electronic format. Oh, and hide the SCANNERS!
Operation Guillotine is in effect.
From the manpage:
findimagedupes 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.
(T>t && O(n)--) == sqrt(666)
http://pixelnovel.com/comparepsd/
first result I found... and it looks to be free.
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?
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.
The electric yellow has got me by the brain banana
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.
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?
sysadmins and parents of newborns get the same amount of sleep.
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.
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.
I've built up so much character I have an alter-ego
Not a bad idea. That would probably catch most of them...
No sig today...
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.
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.
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.
Students must turn in the full image. Much simpler than watermarking.
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
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.
they have been offline for some time now...
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.
Knowledge is how to play a game, intelligence is how to win, wisdom is knowing what game to play.
Steganography?
> 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.
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.
It's better to vote for what you want and not get it than to vote for what you don't want and get it.
- E. Debs
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, 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.
"Inveniemus Viam Aut Faciemus" 'We will find a way... Or we will make one!' --Hannibal of Carthage
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)
All ideas^H^H^H^H^Hprocesses in this post are Patent Pending. (as well as the process of patenting all postings)
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.
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.
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.
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.
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?
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 to diff, easy to see!
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 stuff that floats around the Internet these days.
Faster! Faster! Faster would be better!
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.
Knowledge is how to play a game, intelligence is how to win, wisdom is knowing what game to play.
Perceptual Image Diff and Find Image Dupes 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.
Those who can make you believe absurdities can make you commit atrocities. - Voltaire
Let them grade each other. Of course you will need to anonymize the submissions so that only the creator knows that the picture is his.
If builders built buildings the way programmers wrote programs, then the first woodpecker would destroy civilization.
I went to design school. Only the very worst teachers made us do these kinds of "copy" exercises. I remember this class about illustrator brushes, I never used what they thought me, until years later, when I needed it and I just had to look it up again anyways. Just tell them to create whatever, (something useful, like.. a forum sig or give them a theme) but tell them that they have to include specific elements. layer masks or whatever, and that they will be graded not on creativity per se, but on the skill with which they applied the techniques they learned. Better way of learning, gets them more involved.. more work, but also more fun for the teacher..
Also, doesn't the extended edition have some advanced quantitative analysis tools? I'm not sure what exactly is their scope, but when it comes to calculating differences between images, this sounds like it could be of some help.
Ezekiel 23:20
Option 1 — Quick and manual: DIFFERENCE Blending!!!!
Each file should have the same dimensions.
Flatten student work. Drag it into the final-result file with a shift-click-and-drag, it will automatically align since they have the same dimensions.
Set the blending mode on the layer with the student work DIFFERENCE.
Anything different between the two will automatically jump right out. If the differences are very subtle add levels or curves adjustment layer on the top.
Option 2 — Metadata
Within the "File Info..." you can set the copyright, creator, origin and a host of other fields. It's located within the File drop-down. Needless to say, the file info from student work will be different and they won't know what fields you've changed.
Option 3 — Plug-Ins
There are many watermarking options available with Photoshop. Is the Digimarc filter standard with Photoshop? It's just been there for the last few years that I'm unsure if my agency installs it or not.
Option 4 — All the above plus anything else from the OP and other contributors.
Hope this helps!
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.
Ummm, maybe I'm missing something here? She should be looking at the file, right? What's the harm in checking one little thing as she's looking at it? I mean, how else is she going to be sure they actually did it properly without checking it anyway? Zero solutions should allow her to skip checking the homework entirely, if it does, its kind of missing the point.
While the flat no-layered file is the obvious solution, it will have an unintended side effect of not acting as a fall-back guide for students who get stuck. I don't think she'll particularly save any time when she has more students needing individual help, whereas the sample will clearly show how certain things are done without intervention.
I use this http://www.duplicate-finder.com/photo.html to find image files which are similar (they don't have be identical). Does not work with PSD files though. Maybe the files can be exported to PNG, etc?
I really shouldn't have used someone else's email address for this account.
I agree - a blank slate is the best, in the real world you have to be creative.
Just state that they should have a picture with an animal, a beverage and a well-known landmark and that some types of transitions and effects are expected. But then also state that they aren't limited to that but can do something completely different as long as they have a certain number of effects in the image.
And to make sure that they don't copy an existing image they should provide the source images used too.
If builders built buildings the way programmers wrote programs, then the first woodpecker would destroy civilization.
We require that they take screenshots while working, and they submit the screenshots as a multi-page PDF.
Also, as everyone else is saying, don't distribute the final file. If the files come from a 3rd party (like lynda.com) then add a few more steps onto the tutorial.
Dear Slashdot: next time you want to mess with the site, add a rich-text editor for comments.
For one thing, you can prevent plagiarism by not asking students for plagiarism. You're giving students a file and then asking them to duplicate it. That's pretty much the definition of plagiarism and, frankly, probably of very little educational benefit.
The teacher needs to stop trying to figure out ways to catch people cheating on an exercise designed for cheating and start teaching the damn course. Teaching doesn't just mean lecturing and assigning exercises out of some book, it means developing exercises, homework problems, and exams from scratch as well.
-1 disagree is not a modifier for a reason. -1 troll, flaimbait, redundant, overrated are NOT acceptable substitutes.
binary compare? Open image, save as JPG. Completely different contents to the file. Unless the world switches to lossless images all around and bandwidth makes some pretty big leaps.
Exactly. The OP illustrates multiple failings with the teacher (exercises that are nothing more than copying, attempting to avoid actually looking at images, etc.) and then attributes the problem to the students while jumping over the fact that the students probably aren't learning a god damned thing in class.
-1 disagree is not a modifier for a reason. -1 troll, flaimbait, redundant, overrated are NOT acceptable substitutes.
Depending on what the exercises are, they might not leave much room for slight differences. "Construct a rounded rect button with 12 px radius corners with a vertical gradient from #RRGGBB to #RRGGBB and a 15% drop-shadow with radius of 7px offset by 14px at 120."
A better solution is to digitally watermark the solution files, or if the students have access to the solution files independently (i.e. they came with the book on a CD), watermark the source files and require students to start with those rather than the ones from the CD. Honestly the book ought to have watermarked the solution files already.
Alternatively alter the exercise enough that the students can't depend on the provided solution. On the whole, this is not different from any other course where you're given problems to solve, and also given the solutions. You're going to by-and-large be depending on the students for honesty, if you don't feel you can do so (and that's probably safest), alter the parameters so the provided solution and the correct solution do not agree.
Slay a dragon... over lunch!
If she does not want them to use her assets, Embed each asset with a visible watermark. Come on, is this photoshop teacher a newbie?
Do not look at laser with remaining good eye.
How can the product be "too perfect" when the products should be identical? Ok, date stamps and maybe embeded path names will be different, but the image itself, starting from the same sources, doing the same things, should be the same.
Maybe the problem is that the students are copying things from the book to start with? Shouldn't it be easy to tell that a student has not done this, when his image product is of a golfball and the book example was an umbrella?
In any case, doesn't the teacher have to look at the homework to grade it anyway?
I'm pretty sure it's a simple task to use imagemagick and bash/python/language of choice to make a map of binary differences comparing each image with the results file and with each of the other. Crude attempts to copy the result or another students image should be fairly easy to spot. Also, watermarks or lowering the resolution on the final results file might also help.
When asked why, the answer is almost always: "It's 2014".
Why does this sound like a stock image supplier trying to find machine-modified infringing images using a web crawler so that they can bludgeon the people publishing the modified images, who have not paid a license fee, with a copyright infringement lawsuit?
I'm just saying, a good answer to the OP's question is going to mean the ability to use the answer in this fashion.
Just give them an 'A+'. When they later discover that they spent thousands and they still suck at photoshop, well, that's thier problem.
Honestly, the best way to catch the cheaters is to test them. If they aren't doing the work, they aren't learning the steps. So hold a few tests throughout the term and make the tests worth more than the assignments. Show them, not give them but show them on an overhead projector or using a large photo, the end result that is required. They have 30 minutes to produce it.
Rather than having them copy the output. Give them each a different set of art assets, and have them each turn out an original work using the lessons -taught and shown- by the example. Then there's no copying, and they might actually learn something.
This is basically the digital equivalent of "show your work" on a math test. If you want to see how well the students are grasping certain concepts, tell them to include an audio track in which they describe what they're doing and *why* -- e.g. "lightening this layer now because I did *blah blah* and messed it up previously".
Anyone who is good enough to fake the screencast convincingly probably doesn't need this class. And if you're really concerned about people using a ringer to do their work, have an in-person exam for each student at the end of the term. You won't need more than 10 min each. Tailor each test to the skills demonstrated in their videos -- that way if someone got a hotshot to tear through the work for them, in person you can easily see if they don't know, say, how to use keyboard shortcuts to invert a selection, etc..
coding is life
You could just save all of the steps as a Photoshop Action. Just start recording at the step after the watermark is placed. As long as every student's file has the same resolution, the cheaters only need put in a watermark and then load and run the action.
Your friend is teaching directly from textbooks and is asking students to do otherwise when they complete assignments?
Try actually teaching:
Go over the written material
Show an example (You know, the one that comes with the book that you are already using)
Create a similar but different project for the students to do based on the lesson in the book .
Grade and critique.
Listen to parents raise hell.
ZOMG.
Teacher complains she must examine each students work in turn.
1. get a new job
2. give the students something actually creative to do and ask them to show how they did it.
- http://www.milkme.co.uk
I know this probably isn't the answer that you're looking for, but nothing constructive can be taught in photoshop if the expected final result is intended to be exactly the same. Students should be encouraged to use the techniques taught in the image creation process, but not replicate it pixel-perfect.
How about not giving them access to the file itself?
Give them a jpeg, and make them turn in a photoshop file. Problem solved. The photoshop file will have all the layers etc. Possibly even a history of alterations made?
As long as the assignment is not for teaching softening by blurring effects, some regions should be untouched and still be available for verification. All this creation and verification processing could be automated fairly easily by any capable software engineer.
If you must give them the final version in component form for some reason, how about reducing its scale so that they'd have to seriously enlarge and degrade it in order to produce the required final result?
Haven't read through every comment, so don't know if it's already been said, but why not just look at the date that the file was created? I have a feeling that the chance of them just turning in the 'end result' file would be bigger than the chance that they're copying it into a new file.
What i do, however, is have the students create a unique product for any assessment. This is really the only way to assess that the student understands the process. Even if he or she creates a final product from the book, does that mean they understand the process or just how to follow instructions. Clearly following instructions is important, but the minmimum one should expect for learning is the ability to tweak. After all, that is what many software developers do. Take a bit of code and modify it to do the specific task. In this case, for instance, start with a personal photo and then practice whatever skills are needed on it. This makes grading harder, but maybe not.
"She's a scientist and a lesbian. She's not going to let it slide." Orphan Black
If I'm not mistaken, Photoshop has had "Digimarc" watermarking in it since forever (under Filters). You only get a "demo" ID in the base install (you have to pay for your own personal ID), but the demo would likely be sufficient since you're only really interested in watermarked sample vs. original work. The watermark itself is practically invisible, and meant to be resistant to the sorts of minor edits you're talking about.
When you open the file, go into File -> File Info. Type a number you know like a phone number into the City entry of the Origin section.
Then check this on all files entered. That one is a simple cheat catch against the lazy. It won't stop someone copying layers to a new file.
Why not just say they have to show their work by submitting a file saved periodically during the process?
Calculating the joint entropy or the mutual information between the images should do the trick.
... it's appropriation
They'll learn their lesson that they shouldn't have cheated in class once they get out into the real world and get assigned real projects.
If you disagree with me on social issues, then it's pretty clear that you are a narrow-minded bigot.
Can't you just give them the final product image in a low resolution, but the work images in high resolution and demand a high resolution final product?
Couldn't she simply slip a large amount of bit information into one of the layers? Put a high resolution photograph in a background layer at 100% transparency. This will substantially increase the file size beyond what the students would be producing. Then when she gets the assignments, sort by file size, and pick the ones off the top that are a few MBs too large.
Really?
Why make it so complicated?
How about simply assigning a project that uses a different base image than the example from the textbook?
Bite the bullet a few times and identify the copy images (maybe by secretly marking it somehow as others have said) and FAIL the students who turn in a copy.
There are several reasonably good ways to accomplish this, but the most obvious is to give them the end result image in a format that does not support layers (e.g., png). The students, of course, must turn in their assignment in layered format, on the "show your work" principle.
This doesn't prevent students from working together, of course.
Cut that out, or I will ship you to Norilsk in a box.
A stock image supplier would already be smart enough to use a watermark, no?
I am not really here right now.
Because it very well could be just that. The teacher wants the students to do the creative work, but the teacher does not want to put the work into the grading portion of the student-teacher contract, eh? It's kind of like requirinig "Turn-it-in" to check for plagiaristic turns-of-phrase so that the teacher can off-load or out-source a first-pass of the grading to somewhere else. Maybe there's a service to be built on the web for this project: turn-in-your-psds.com which takes images to compare and charges a service fee to tell you the similarity index (si-silicon-i-better-trademark this right now! silicon, graphics, silicon graphics, yeah nobody would have ever used that phrase ;) ) and returns a first-pass version of is this student cheating...
accept a "study group" could together show each other their work and then grade eachother higher resulting in a higher grade and others lower resulting in a higher grade for themselves
---Saying gnome 3 is better than windows 8 not so much a compliment as it is damning with light praise.
Just put the "original" on the top layer, set the layer effect to "difference"
All identical pixels will be black, all others will show up as a rainbow looking effect. A perfect match creates a solid black image.
It's fun to do this with NASA photos of the moon, some of them have lens flares added from photoshop. :) (you can recreate some the lensflares pixel perfect...)
Add dust to the final image. Make it four or five specks - even at 1% opacity works. Then, write a script to -Open "pixelspecs.config" containing information about the pixel color at each 1px speck of dust -Iterate through the "studentprojects" directory ---Open the file ---Loop through and check how many of the pixels match the color ---If it's more than two, print the file name: they plagiarized! Not too hard to do, could be done in less than 50 lines of code.
"I am a writing teacher and I am having a problem catching my students cheating. The assignment is to write Romeo and Juliet. The problem is that some of them just take a copy of the text off the internet and hand that in! The clever ones insert a few typos, which means I can't just do a byte-by-byte comparison. What can I do to determine if that's what they did, instead of sitting down and rewriting the story themselves from memory, or (for the less gifted ones) typing it while looking at a paperback copy of the play?"
http://alternatives.rzero.com/
- to gimp & g'mic fourier watermark each resource before putting them back into their layers in Photoshop. But if you are really hot, you just export the psd directly from gimp. When they give it back, you flatten and run fourier analysis. If the watermark appears, you know they are cheating. Then just sudo them a new sandwich from the console on your MacBook.
Give all the students individual assignments, based on their student registration number.
First number 1, use this texture; 2, use that texture and so on.
They could copy-and-change, but that would still be a lot of work. Make sure that would be more work than just doing the assignments.
An other trick is to point out the risks of getting caught.
But the more interesting thing is indeed to write something to analyse the files. :)
I made something to compare Maya files. That worked quite nicely, but I'm not sure if the students liked it. I know I did though
Privacy is terrorism.
Steganography? Then if you see the hidden stuff, you know it's a copy.
Let them use their own base images! And then let them do something creative with them!
One of the least interesting and least creative classes I took in art school was one that was about producing photorealistic oil paintings based on photographs. The class was 99% about mechanical technique, and to hell with creativity... which seems to be the theme of the class being taught here. So be it. But at least the instructor let us pick our own photographs to replicate! So we'd have an interest in what we were doing. And even if he had never checked on our progress along he way (like would happen in any worthwhile "learn how to ____" class), he would know whether we had done the work, because each of our paintings was a) unique, and b) matched the photograph we'd had approved at the start of the assignment. Plagiarism wasn't even a question, and not just because we were working in traditional physical media.
All of these suggestions for how to identify plagiarism through technological measures are missing the point. The problem isn't "how to catch a cheat", but "how to give students an assignment that they will have a reason to bother doing in the first place".
http://alternatives.rzero.com/
You can simply change the example solution so it can be caught by a teacher actually examining the work of the student. Yet you are still looking for an automatic solution? Let the teacher continue to be a teacher and take your geek points and continue being a geek.
We should learn what we need to know about issues, before we decide what we need to feel about them.
I will assume the end result file isn't in a format that allows them to "undo" the steps? So why not have them provide each step they took as well as their end result? That way they can't just copy and past the end result without showing their workings.
Um ... yeah. The more I think about it, the more I agree that this is likely the case.
As it's been mentioned several times here already, it's as simple as providing a flattened JPG or PNG as the assignment and requiring that students submit a completed PSD. Any competent "Photoshop teacher" would know this.
Leave it to Slashdot to come up with a miriad of Rube Goldberg solutions to a simple problem.
Duh. Flatten the final picture. You will still have outliers who can make it look convincingly different enough but it won't be as easy as giving them access to each individual layer for them to fudge up.
Also consider embedding tiny watermarks in the image. Little sets of pixels at specific coordinates. Not specifically exact colors but something you would be able to identify on close inspection. Something like a small 6x6 checkerboard grid of alternating light colored pixels in a light area of the document that wouldn't be easily seen by your students but you could zero in on and verify.
Um, can't you just watermark the images?
Either something visible, or something not visible, that you can check.
Why you didn't think of that, I have no idea, Mr. "I Teach Photoshop". I mean, it's not like websites haven't been watermarking their images since the 90's.
Be seeing you...
I think you should tell them at the outset "It's really easy for you to cheat on your assignments. That's also a horrible way to learn. I've got an honor system, don't cheat. If you do cheat you will learn less, and therefore be wasting your own time." If you need something to base grades on, you need something else that you can watch then do or they can't cheat on somehow.
San Francisco Photographers
Why are you giving the students the final photoshop file with all the layers? Just give them a jpg with all the layers compressed, and put a big fat watermark on it so they can't use it for anything.
require each student to turn on their undo to a bazillion undo's .
have them turn in the finished photoshop project, the entire thing
then the teacher should be able to undo all the way back to the original step by step and then replay all the steps (using redo) to get to the final again.
Just hold down ctrl-Z and watch
Works in GIMP anyway.
Uh huh, guess you didn't really learn how to use them.
The point was to LEARN, so when you get in the real world, you can develop a solution where pointers (or whatever structure/technique/language feature is most appropriate) makes the most sense.
Yeah, right.
The teacher wants the students to do the creative work, but the teacher does not want to put the work into the grading portion of the student-teacher contract, eh?
I think that's a little unfair (and I'm the OP, by the way... and I'm going to address some of the other criticisms, here, so don't take all of this as being directed toward you). That's like saying that you're being lazy by using a testing suite to do your unit tests on the code you're working on. So, stop being lazy and go back to testing your code by hand.
My personal opinion is that grading is a Q/A step in the teaching process. Teaching is when you're actually "generating product", in the sense that you're putting knowledge into brains where it previously wasn't. Grading is just checking to see if it actually worked and, every minute you're grading is a minute you're not teaching. So, it seems to me that, the more you can automate the grading (just like automating testing of your code), the more you can focus on actually producing (ie, adding knowledge to those malleable little brains).
Also, keep in mind that she teaches about 4-5 very un-related courses (video production, electronic media, freehand drawing, printing techniques), each with it's own set of students who either A) are trying to contrive ways to avoid doing the actual work or B) want to actually learn, but who can't follow directions for shit (I used to grade for math, physics, and CS in college, and, even at that level, it's amazing at how hard students seem to make it for you to give them the credit they deserve). What you end up with is about 25 homework submissions, none of which look alike, and you've got to figure out which ones are properly demonstrating the learned skills and which ones are just blowing it off. The submissions which look almost perfect require some painstaking attention because the kid is either really good (and deserves 100%) or they just copied the end result (and deserves a 0), so the stakes are higher with those than with the other submissions. Now, multiply all that work by 4-5 classes of kids. That's why she was at the school, grading, through the whole 3-day weekend. Fun way to spend your long weekend, eh? Now, what was that again about lazy teachers, basking in the warm glow of union protection while they run their feet through the sand at the beach?
So that's where I came in. When I saw her comparing images, side by side... overlaying them, adjusting transparencies to find differences, my first reaction was (due to the "hubris" and "impatience" traits of programmers) "Hey, I'm a programmer. I can write you an application which just compares all of them and tells you which ones are identical faster than you can drag-and-drop them into the app". Then, she could get on to thinking up cool projects for the next week. But, alas, as I thought about it, MD5's wouldn't work (as the students sometimes change names of layers or make other trivial changes). And, also, the "impatience" programmer trait kicked in and I said "Somebody has to have solved this problem before". Now, we did realize that she could just provide a flat image of the target result, but the problem still stuck in my mind, since I'm a programmer, but fuzzy image comparison is not my forte, so I figured I'd ask. This wasn't started by her asking me to find some way for her to automate her grading so she could duck out early and hit the nail salon. This was started by me, reflexively seeking a way to automate a boring, labor-intensive process (like unit-testing) so that she could get on to the creative parts of teaching and also because I was curious about open-source image processing stuff out there, these days.
And not the file itself.
Closest match gets the highest grade..
Unless precision is an absolute requirement, give the example file at half the resolution of the wanted result. They will still get the gist of what is supposed to be done, with the desired layers and all, but won't be able to submit it as-is.