how to compare set of images in java using pixel based image comparision metric based on mean squared error?

Question

In my project i have a set of images. I need to compare them. Each pixel from one image is compared to the pixel at the same location in all other images in the dataset. Aft

Answer 1

You can do it using Catalano Framework. There's several metrics to compare image, including mean square error.

Example:

FastBitmap original = new FastBitmap(bufferedImage1);
FastBitmap reconstructed = new FastBitmap(bufferedImage2);

ObjectiveFidelity o = new ObjectiveFidelity(original, reconstructed);

// Error total
int error = o.getTotalError();

//Mean Square Error
double mse = o.getMSE();

//Signal Noise Ratio
double snr = o.getSNR();

//Peak Signal Noise Ratio
double psnr = o.getPSNR();

All these metrics are based in the book: Computer Imaging: Digital Image Analysis and Processing - Scott E Umbaugh.

Next version (1.3) will contains Derivative SNR.

Answer 2

  File file1 = new File(args[0]);
  File file2 = new File(args[1]);

  boolean compareResult = FileUtils.contentEquals(file1, file2);
  System.out.println("Are the files are same? " + compareResult);

Apache Commons IO library to perform this comparison (download commons-io-2.4.jar)

Answer 3

Right now your code will compare always the file 2000.png with 2002.png. First refactor your comparison code into a method:

CompareResult compare(BufferedImage  img1,BufferedImage  img2) { ... }

Declare a class CompareResult to group the result info into one Message:

class CompareResult {
     long pixMatched;
     long pixVar;
     ...
}

Then expect a list of image files from commandline and compre them each to each other:

for(int i=0;i<args.length();i++) {
    BufferedImage img1 = ImageIO.read(args[i]);
    for (int j=i+1;j<args.length();j++) {
        BufferedImage img2 = ImageIO.read(args[j]);
        CompareResult r = compare(img1,img2);
        printInfo(r);
    }
}