Converting a Bitmap to a Byte Array does not produce the expected results

Question

I have the following code in my backend. I need an equivalent of this in Android Bitmap.

public byte[] extractBytes (String ImageName) throws IOException {
    // open image
    File imgPath = new File(ImageName);
    BufferedImage bufferedImage = ImageIO.read(imgPath);

    // get DataBufferBytes from Raster
    WritableRaster raster = bufferedImage .getRaster();
    DataBufferByte data   = (DataBufferByte) raster.getDataBuffer();

    return ( data.getData() );
}

I have tried the following:

1) Using Bitmap.compress() function

src.compress(Bitmap.CompressFormat.JPEG, 100, stream);
byte[] byteArray = stream.toByteArray();

2) Using Bitmap.copyPixelsToBuffer() function

int size = src.getRowBytes() * src.getHeight();
ByteBuffer byteBuffer = ByteBuffer.allocate(size);
src.copyPixelsToBuffer(byteBuffer);
byte[] byteArray = byteBuffer.array();

Both of the methods are giving me wrong results which don't match with my backend output.

I have the following code in my backend for Component Labelling Algorithm.

BlobDemo.java

public class BlobDemo
{
    static ListDisplayPanel gui2 = new ListDisplayPanel();
    public BlobDemo(String filename)
    {
        // Load Source image
        BufferedImage srcImage = null;
        try {
           File imgFile = new File(filename);
           srcImage = javax.imageio.ImageIO.read(imgFile);
        }
        catch (IOException ioE) {
          System.err.println(ioE);
          System.exit(1);
       }

        int width = srcImage.getWidth();
        int height = srcImage.getHeight();

        // Get raw image data
        Raster raster = srcImage.getRaster();
        DataBuffer buffer = raster.getDataBuffer();

        int type = buffer.getDataType();
        if (type != DataBuffer.TYPE_BYTE)
        {
            System.err.println("Wrong image data type");
            System.exit(1);
        }
        if (buffer.getNumBanks() != 1)
        {
            System.err.println("Wrong image data format");
            System.exit(1);
        }

        for(int i=0;i<10;i++){
            for(int j=0;j<10;j++){
                int pix = srcImage.getRGB(i, j);
                Color mycolor = new Color(pix);
                int red = mycolor.getRed();
                int green = mycolor.getGreen();
                int blue = mycolor.getBlue();
                String val ="("+red+","+green+","+blue+")";
                System.out.print(val);
            }
            System.out.println();
        }

        DataBufferByte byteBuffer = (DataBufferByte) buffer;
        byte[] srcData =( (DataBufferByte) (buffer)).getData();

        // Sanity check image
        if (width * height * 3 != srcData.length) {
            System.err.println("Unexpected image data size. Should be RGB image");
            System.exit(1);
        }

        // Output Image info
        System.out.printf("Loaded image: '%s', width: %d, height: %d, num bytes: %d\n", filename, width, height, srcData.length);

        // Create Monochrome version - using basic threshold technique
        byte[] monoData = new byte[width * height];
        int srcPtr = 0;
        int monoPtr = 0;

        while (srcPtr < srcData.length)
        {
            int val = ((srcData[srcPtr]&0xFF) + (srcData[srcPtr+1]&0xFF) + (srcData[srcPtr+2]&0xFF)) / 3;
            monoData[monoPtr] = (val > 128) ? 0 : (byte) 0xFF;

            srcPtr += 3;
            monoPtr += 1;
        }       

        byte[] dstData = new byte[srcData.length];

        // Create Blob Finder
        BlobFinder finder = new BlobFinder(width, height);

        ArrayList<BlobFinder.Blob> blobList = new ArrayList<BlobFinder.Blob>();
        finder.detectBlobs(monoData, dstData, 0, -1, (byte)0, blobList);

        // List Blobs
        System.out.printf("Found %d blobs:\n", blobList.size());
        for (BlobFinder.Blob blob : blobList){
            System.out.println(blob);
        }
    }

    public static void main(String[] args)
    {
        System.out.println("Blob Finder Demo");
        new BlobDemo("/home/maruthi/Desktop/images.jpg");
    }
}

BlobFinder.java

public class BlobFinder
{
    private byte[][] COLOUR_ARRAY = {{(byte)103, (byte)121, (byte)255},
                                                {(byte)249, (byte)255, (byte)139},
                                                {(byte)140, (byte)255, (byte)127},
                                                {(byte)167, (byte)254, (byte)255},
                                                {(byte)255, (byte)111, (byte)71}};

    private int width;
    private int height;

    private int[] labelBuffer;

    private int[] labelTable;
    private int[] xMinTable;
    private int[] xMaxTable;
    private int[] yMinTable;
    private int[] yMaxTable;
    private int[] massTable;

    static class Blob
    {
        public int xMin;
        public int xMax;
        public int yMin;
        public int yMax;
        public int mass;

        public Blob(int xMin, int xMax, int yMin, int yMax, int mass)
        {
            this.xMin = xMin;
            this.xMax = xMax;
            this.yMin = yMin;
            this.yMax = yMax;
            this.mass = mass;
        }

        public String toString()
        {
            return String.format("X: %4d -> %4d, Y: %4d -> %4d, mass: %6d", xMin, xMax, yMin, yMax, mass);
        }
    }

    public BlobFinder(int width, int height)
    {
        this.width = width;
        this.height = height;

        labelBuffer = new int[width * height];

        // The maximum number of blobs is given by an image filled with equally spaced single pixel
        // blobs. For images with less blobs, memory will be wasted, but this approach is simpler and
        // probably quicker than dynamically resizing arrays
        int tableSize = width * height / 4;

        labelTable = new int[tableSize];
        xMinTable = new int[tableSize];
        xMaxTable = new int[tableSize];
        yMinTable = new int[tableSize];
        yMaxTable = new int[tableSize];
        massTable = new int[tableSize];
    }

    public List<Blob> detectBlobs(byte[] srcData, byte[] dstData, int minBlobMass, int maxBlobMass, byte matchVal, List<Blob> blobList)
    {
        if (dstData != null && dstData.length != srcData.length * 3)
            throw new IllegalArgumentException("Bad array lengths: srcData 1 byte/pixel (mono), dstData 3 bytes/pixel (RGB)");

        // This is the neighbouring pixel pattern. For position X, A, B, C & D are checked
        // A B C
        // D X

        int srcPtr = 0;
        int aPtr = -width - 1;
        int bPtr = -width;
        int cPtr = -width + 1;
        int dPtr = -1;

        int label = 1;

        // Iterate through pixels looking for connected regions. Assigning labels
        for (int y=0 ; y<height ; y++)
        {
            for (int x=0 ; x<width ; x++)
            {
                labelBuffer[srcPtr] = 0;

                // Check if on foreground pixel
                if (srcData[srcPtr] == matchVal)
                {
                    // Find label for neighbours (0 if out of range)
                    int aLabel = (x > 0 && y > 0)           ? labelTable[labelBuffer[aPtr]] : 0;
                    int bLabel = (y > 0)                        ? labelTable[labelBuffer[bPtr]] : 0;
                    int cLabel = (x < width-1 && y > 0) ? labelTable[labelBuffer[cPtr]] : 0;
                    int dLabel = (x > 0)                        ? labelTable[labelBuffer[dPtr]] : 0;

                    // Look for label with least value
                    int min = Integer.MAX_VALUE;
                    if (aLabel != 0 && aLabel < min) min = aLabel;
                    if (bLabel != 0 && bLabel < min) min = bLabel;
                    if (cLabel != 0 && cLabel < min) min = cLabel;
                    if (dLabel != 0 && dLabel < min) min = dLabel;

                    // If no neighbours in foreground
                    if (min == Integer.MAX_VALUE)
                    {
                        labelBuffer[srcPtr] = label;
                        labelTable[label] = label;

                        // Initialise min/max x,y for label
                        yMinTable[label] = y;
                        yMaxTable[label] = y;
                        xMinTable[label] = x;
                        xMaxTable[label] = x;
                        massTable[label] = 1;

                        label ++;
                    }

                    // Neighbour found
                    else
                    {
                        // Label pixel with lowest label from neighbours
                        labelBuffer[srcPtr] = min;

                        // Update min/max x,y for label
                        yMaxTable[min] = y;
                        massTable[min]++;
                        if (x < xMinTable[min]) xMinTable[min] = x;
                        if (x > xMaxTable[min]) xMaxTable[min] = x;

                        if (aLabel != 0) labelTable[aLabel] = min;
                        if (bLabel != 0) labelTable[bLabel] = min;
                        if (cLabel != 0) labelTable[cLabel] = min;
                        if (dLabel != 0) labelTable[dLabel] = min;
                    }
                }

                srcPtr ++;
                aPtr ++;
                bPtr ++;
                cPtr ++;
                dPtr ++;
            }
        }

        // Iterate through labels pushing min/max x,y values towards minimum label
        if (blobList == null) blobList = new ArrayList<Blob>();

        for (int i=label-1 ; i>0 ; i--)
        {
            if (labelTable[i] != i)
            {
                if (xMaxTable[i] > xMaxTable[labelTable[i]]) xMaxTable[labelTable[i]] = xMaxTable[i];
                if (xMinTable[i] < xMinTable[labelTable[i]]) xMinTable[labelTable[i]] = xMinTable[i];
                if (yMaxTable[i] > yMaxTable[labelTable[i]]) yMaxTable[labelTable[i]] = yMaxTable[i];
                if (yMinTable[i] < yMinTable[labelTable[i]]) yMinTable[labelTable[i]] = yMinTable[i];
                massTable[labelTable[i]] += massTable[i];

                int l = i;
                while (l != labelTable[l]) l = labelTable[l];
                labelTable[i] = l;
            }
            else
            {
                // Ignore blobs that butt against corners
                if (i == labelBuffer[0]) continue;                                  // Top Left
                if (i == labelBuffer[width]) continue;                              // Top Right
                if (i == labelBuffer[(width*height) - width + 1]) continue; // Bottom Left
                if (i == labelBuffer[(width*height) - 1]) continue;         // Bottom Right

                if (massTable[i] >= minBlobMass && (massTable[i] <= maxBlobMass || maxBlobMass == -1))
                {
                    Blob blob = new Blob(xMinTable[i], xMaxTable[i], yMinTable[i], yMaxTable[i], massTable[i]);
                    blobList.add(blob);
                }
            }
        }

        // If dst buffer provided, fill with coloured blobs
        if (dstData != null)
        {
            for (int i=label-1 ; i>0 ; i--)
            {
                if (labelTable[i] != i)
                {
                    int l = i;
                    while (l != labelTable[l]) l = labelTable[l];
                    labelTable[i] = l;
                }
            }

            // Renumber lables into sequential numbers, starting with 0
            int newLabel = 0;
            for (int i=1 ; i<label ; i++)
            {
                if (labelTable[i] == i) labelTable[i] = newLabel++;
                else labelTable[i] = labelTable[labelTable[i]];
            }

            srcPtr = 0;
            int dstPtr = 0;
            while (srcPtr < srcData.length)
            {
                if (srcData[srcPtr] == matchVal)
                {
                    int c = labelTable[labelBuffer[srcPtr]] % COLOUR_ARRAY.length;
                    dstData[dstPtr] = COLOUR_ARRAY[c][0];
                    dstData[dstPtr+1]   = COLOUR_ARRAY[c][1];
                    dstData[dstPtr+2]   = COLOUR_ARRAY[c][2];
                }
                else
                {
                    dstData[dstPtr] = 0;
                    dstData[dstPtr+1]   = 0;
                    dstData[dstPtr+2]   = 0;
                }

                srcPtr ++;
                dstPtr += 3;
            }
        }
        return blobList;
    }
}

I need to convert the above code to Android. I am stuck here.

for(int i=0;i<10;i++) {
    for(int j=0;j<10;j++) {
        int pix= src.getPixel(i,j);
        int red = Color.red(pix);
        int green= Color.green(pix);
        int blue = Color.blue(pix);
        int alpha = Color.alpha(pix);
        String val ="("+red+","+green+","+blue+")";
        System.out.print(val);
    }
    System.out.println();
}

When I print these values, I get totally different results.

I also tried the following.

int[] pix1=new int[width*height];
src.getPixels(pix1, 0, width, 0, 0, width , height );
byte[] pixels= new byte[pix1.length*3];
for (int i = 0; i < pix1.length; i++) {
    int td = pix1[i];
    int tind = i * 3;
    pixels[tind++] = (byte) ((td >> 0) & 0xFF);
    pixels[tind++] = (byte) ((td >> 8) & 0xFF);
    pixels[tind++] = (byte) ((td >> 16) & 0xFF);
}

Example:

Backend Code:

for(int i=0;i<10;i++){
    for(int j=0;j<10;j++){
        int pix = srcImage.getRGB(i, j);
        Color mycolor = new Color(pix);
        int red = mycolor.getRed();
        int green = mycolor.getGreen();
        int blue = mycolor.getBlue();
        String val ="("+red+","+green+","+blue+")";
        System.out.print(val);
    }
    System.out.println();
}

Backend Output:

(253,253,253)(254,254,254)(254,254,254)(254,254,254)(254,254,254)(254,254,254)(254,254,254)(254,254,254)(254,254,254)(254,254,254)
(254,254,254)(255,255,255)(255,255,255)(255,255,255)(255,255,255)(255,255,255)(255,255,255)(255,255,255)(255,255,255)(255,255,255)
(254,254,254)(255,255,255)(255,255,255)(255,255,255)(255,255,255)(255,255,255)(255,255,255)(255,255,255)(255,255,255)(255,255,255)
(254,254,254)(255,255,255)(255,255,255)(255,255,255)(255,255,255)(255,255,255)(255,255,255)(255,255,255)(255,255,255)(255,255,255)
(254,254,254)(255,255,255)(255,255,255)(255,255,255)(255,255,255)(255,255,255)(255,255,255)(255,255,255)(255,255,255)(255,255,255)
(254,254,254)(255,255,255)(255,255,255)(255,255,255)(255,255,255)(255,255,255)(255,255,255)(255,255,255)(255,255,255)(255,255,255)
(254,254,254)(255,255,255)(255,255,255)(255,255,255)(255,255,255)(255,255,255)(255,255,255)(255,255,255)(255,255,255)(255,255,255)
(254,254,254)(255,255,255)(255,255,255)(255,255,255)(255,255,255)(255,255,255)(255,255,255)(255,255,255)(255,255,255)(255,255,255)
(254,254,254)(255,255,255)(255,255,255)(255,255,255)(255,255,255)(255,255,255)(255,255,255)(255,255,255)(255,255,255)(255,255,255)
(254,254,254)(255,255,255)(255,255,255)(255,255,255)(255,255,255)(255,255,255)(255,255,255)(255,255,255)(255,255,255)(255,255,255)

Android Code:

for(int i=0;i<10;i++ ){
    for(int j=0;j<10;j++){
        int pixel = test.getPixel(i,j);
        int red = Color.red(pixel);
        int green = Color.green(pixel);
        int blue= Color.blue(pixel);
        String val ="("+red+","+green+","+blue+")";
        System.out.print(val);
    }
    System.out.println();
}

Android Output:

(253,253,253)(254,254,254)(254,254,254)(253,253,253)(254,254,254)(254,254,254)(254,254,254)(254,254,254)(254,254,254)(254,254,254)
(254,254,254)(254,254,254)(255,255,255)(254,254,254)(254,254,254)(254,254,254)(254,254,254)(254,254,254)(255,255,255)(255,255,255)
(254,254,254)(255,255,255)(255,255,255)(254,254,254)(255,255,255)(255,255,255)(255,255,255)(255,255,255)(255,255,255)(255,255,255)
(253,253,253)(254,254,254)(254,254,254)(254,254,254)(254,254,254)(254,254,254)(254,254,254)(254,254,254)(254,254,254)(254,254,254)
(254,254,254)(254,254,254)(255,255,255)(254,254,254)(255,255,255)(255,255,255)(254,254,254)(255,255,255)(255,255,255)(255,255,255)
(254,254,254)(254,254,254)(255,255,255)(254,254,254)(255,255,255)(255,255,255)(254,254,254)(255,255,255)(254,254,254)(254,254,254)
(254,254,254)(254,254,254)(255,255,255)(254,254,254)(254,254,254)(254,254,254)(254,254,254)(254,254,254)(255,255,255)(255,255,255)
(254,254,254)(254,254,254)(255,255,255)(254,254,254)(255,255,255)(255,255,255)(254,254,254)(255,255,255)(254,254,254)(254,254,254)
(254,254,254)(254,254,254)(255,255,255)(254,254,254)(255,255,255)(255,255,255)(254,254,254)(255,255,255)(255,255,255)(255,255,255)
(254,254,254)(254,254,254)(255,255,255)(254,254,254)(255,255,255)(255,255,255)(254,254,254)(255,255,255)(255,255,255)(255,255,255)

How can I get the same byte array in Android as given in the original function?