Takes a panoramic image that is in a 'fisheye' view and unwraps it. See http://terra-azure.org/?loc=articles/programming/java/polar2cartesian for an explanation

polar2cartesian.java

package ajmas74.experimental.graphics2d;

import java.awt.Image;
import java.awt.image.BufferedImage;
import java.awt.image.ColorModel;
import java.awt.image.DataBuffer;
import java.awt.image.DataBufferInt;
import java.awt.image.DirectColorModel;
import java.awt.image.PixelGrabber;
import java.awt.image.Raster;
import java.awt.image.WritableRaster;
import java.io.File;
import java.io.IOException;

import javax.imageio.ImageIO;

/**
 * See http://terra-azure.org/?loc=articles/programming/java/polar2cartesian
 * @author andre-john mas
 */
public class RolloutImage {

	/**
	 * This takes a panoramic image that is in a 'fisheye' view and unwraps it. The method
	 * take an 'innerRadius' and an 'outerRadius' parameter. If your image contains a 
	 * central area, that should be ignored, specify the radius, in pixels, of the area
	 * to be ignored, in the 'innerRadius' parameter, otherwise you can specify '0' as value.
	 * 
	 * 
	 * @param image the image to convert
	 * @param innerRadius the inner radius, representing the area with no significant data
	 * @param outerRadius the radius to which the image extends to
	 * @param newWidth the width of the new image
	 * @return converted image
	 */
    public static BufferedImage convertPolarImageToCartesian(Image image,
            int innerRadius, int outerRadius, int newWidth) {
        
        int imgWidth = image.getWidth(null);
        int imgHeight = image.getHeight(null);

        // Get this pixels from the original image //
        int[] packedPixels = new int[imgWidth * imgHeight];
        PixelGrabber pixelgrabber = new PixelGrabber(image, 0, 0, imgWidth,
                imgHeight, packedPixels, 0, imgWidth);
        try {
            pixelgrabber.grabPixels();
        } catch (InterruptedException e) {
            throw new RuntimeException();
        }

        int width = newWidth;
        int height = outerRadius - innerRadius;
        int size = width * height;
        int[] pixelData = new int[size];

        // Get the center point of the image //
        int xc = imgWidth / 2;
        int yc = imgHeight / 2;

        // Do the conversion //
        for (int r = innerRadius; r < outerRadius; r++) {
            // t = theta in degrees
            for (int t = 0; t < width; t++) {
                // theta in radians
                double tr = Math.toRadians((t * 1.0) / width * 360);
                int x = (int) Math.round(xc + r * Math.cos(tr));
                int y = (int) Math.round(yc + r * Math.sin(tr));

                int rDiff = r - innerRadius;
                int h = width * (height - (rDiff + 1));
                pixelData[h + t] = packedPixels[imgWidth * y + x];

            }
        }

        DataBuffer dataBuf = new DataBufferInt(pixelData, size);

        ColorModel colorModel = new DirectColorModel(32, 0x00ff0000,
                0x0000ff00, 0x000000ff);
        int[] masks = { 0x00ff0000, 0x0000ff00, 0x000000ff };
        WritableRaster raster = Raster.createPackedRaster(dataBuf, width,
                height, width, masks, null);

        BufferedImage convertedImage = new BufferedImage(colorModel, raster,
                false, null);

        return convertedImage;

    }
    
    public static Image loadImage(String filePath) throws IOException {
        return ImageIO.read(new File(filePath));        
    }

    public static void main(String[] args) throws Exception {
        // TODO Auto-generated method stub
        Image image = ImageIO.read(new File("data/circular-panormaic.jpg"));
        
        BufferedImage convertedImage = convertPolarImageToCartesian(
                image, 1, 510, 510);
        
        File outputfile = new File("data/out.png");
        ImageIO.write(convertedImage, "png", outputfile);
    }

}