 |
 | Home | Contact Us | FAQ | Search & Site Map | Link to Us |
 | Sign In | Join | Other 45 Sites in Network |
Java Forum / GUI / January 2005fast image compare?
Whats the best way to compare 2 images to see if they are exactley the same? Specifically, a BufferedImage and a sun.awt.windows.WImage. My plan is to convert the WImage into a BufferedImage. Then verify that they are the same size. Then iterate thru each pixel rgb value. Is there a better way? Tx, Darrel > Whats the best way to compare 2 images to see if they are exactley the > same? [quoted text clipped - 4 lines] > they are the same size. Then iterate thru each pixel rgb value. Is there > a better way? you could compare say 1000 random choosed pixels  SignatureAndrey Kuznetsov http://uio.dev.java.net Unified I/O for Java http://reader.imagero.com Java image reader http://jgui.imagero.com Java GUI components and utilities spam0@imagero.com.invalid wrote in comp.lang.java.gui: >> Whats the best way to compare 2 images to see if they are exactley the >> same? <snip> > you could compare say 1000 random choosed pixels ...and get random results SignatureAntti S. Brax Rullalautailu pitää lapset poissa ladulta http://www.iki.fi/asb/ http://www.cs.helsinki.fi/u/abrax/hlb/ "Disconnect this cable to shorten, re-connect to lengthen." -- Instructions on Logitech's USB mouse extension cord. >>> Whats the best way to compare 2 images to see if they are exactley the >>> same? > <snip> >> you could compare say 1000 random choosed pixels > > ...and get random results no, because it is just for quick check - if some pixel are different then you know that images are not same. Comparing random pixels can help if images has big similar regions. After that, if all pixels had same value, you can compare pixel for pixel. SignatureAndrey Kuznetsov http://uio.dev.java.net Unified I/O for Java http://reader.imagero.com Java image reader http://jgui.imagero.com Java GUI components and utilities Please set the Followup-To header when crossposting. riekhof@lvcm.com wrote in comp.lang.java.gui: > Whats the best way to compare 2 images to see if they are exactley the same? > [quoted text clipped - 3 lines] > they are the same size. Then iterate thru each pixel rgb value. Is there a > better way? Are you asking for a better way to convert an image to a BufferedImage or a better algorithm for check the pixels? I'd recommend converting the images into byte arrays and then comparing them instead of going through an API every time you compare a pixel. SignatureAntti S. Brax Rullalautailu pitää lapset poissa ladulta http://www.iki.fi/asb/ http://www.cs.helsinki.fi/u/abrax/hlb/ "Disconnect this cable to shorten, re-connect to lengthen." -- Instructions on Logitech's USB mouse extension cord. > Please set the Followup-To header when crossposting. > [quoted text clipped - 14 lines] > then comparing them instead of going through an API every > time you compare a pixel. It might be worth checking a few pixcels using the API first - you will eliminate most non-matching images very quickly, and save many full conversions. >> Please set the Followup-To header when crossposting. >> [quoted text clipped - 19 lines] > eliminate most non-matching images very quickly, and save many full > conversions. Or use a small byte array of approx. one scan line. > Whats the best way to compare 2 images to see if they are exactley the > same? [quoted text clipped - 6 lines] > > Tx, Darrel You might perform a hash function on each image, and then compare the resulting hash values. That might be a problem if both images are structured differently: you would have to convert each to a common format first. regards, Dar7yl > You might perform a hash function on each image, and then compare the > resulting hash values. How do you imagine that to be faster than comparing the actual data (which you're going to have to do anyway if the hashcodes match)? >> You might perform a hash function on each image, and then compare the >> resulting hash values. > > How do you imagine that to be faster than comparing the actual data > (which you're going to have to do anyway if the hashcodes match)? You just have to choose a hash function that is cheaper than the compare function you would use. For instance, just add the binary values a raster line at a time. You can shortcut at any time as well. Even if you have to repeat the total compare after the hashes match, you would save time if the number of negative compares is significantly higher than the postives. regards, Dar7yl. >>>You might perform a hash function on each image, and then compare the >>>resulting hash values. [quoted text clipped - 9 lines] > would save time if the number of negative compares is significantly higher > than the postives. Not if the total compare aborts with a negative result as soon as it detects a difference, which any sane implementation will do. > You might perform a hash function on each image, and then compare the > resulting hash values. That might be a problem if both images are > structured differently: you would have to convert each to a common format > first. for hash function you can use any digest method, e.g. md5. if your images are in database, you can also save hash values for every image and this could save you much time... (because hash computed only once for each image) SignatureAndrey Kuznetsov http://uio.dev.java.net Unified I/O for Java http://reader.imagero.com Java image reader http://jgui.imagero.com Java GUI components and utilities > Whats the best way to compare 2 images to see if they are exactley the same? > [quoted text clipped - 5 lines] > > Tx, Darrel One way to do this is with Fourier Transform. By comparing the magnitude and phase of the transform you can determine if one image is the same as another only under some type of transform such as scale rotation, or translation. Here are some sites that may be useful: http://www.dai.ed.ac.uk/HIPR2/fourier.htm http://www.mathworks.com/access/helpdesk/help/toolbox/images/transfo5.shtml http://www.cs.unm.edu/~brayer/vision/fourier.html Google for: Fourier-Based Image Registration Techniques & Fourier-Mellin Transforms for Image Registration It is popular in research because it has possibilities with image watermarking as well as image recognition for robots and the like. Another thing you could to is simply take one image and subtract the other. This would probably be the easiest. I know Jimage can do this and I am sure there are alot of others. basically subtract pixel for pixel and take the absolutes value of the result. The resultant image should be white were the images match and colored were the images do not. You can then count the non white pixels or do a image histogram of the results. Mark Free MagazinesGet these publications absolutely FREE for up to 12 months. There are no hidden fees and no obligation. Simply choose a title, complete the application form and submit it. Read more ...
|
Reply to this Message
|
 Sign In
 Join
 My Latest Posts My Monitored Threads My Blog My Photo Gallery My Profile My Homepage Start New Thread Enable EMail Alerts Rate this Thread
|
©2008 Advenet LLC Privacy Policy - Terms of Use
This website includes both content owned or controlled by Advenet as well as content owned or controlled by third parties.