What steps will reproduce the problem? 1.First, thanks for create great AForge.net free library in image processing field. 2.I use the code "new ApplyMask(imageMask).ApplyInPlace(image);" in my project. 3.I use too the code "new MaskedFilter(new Sepia(), image).ApplyInPlace(image);" and "image" is an "AForge.Imaging.UnmanagedImage" that by "ref" pass to my mask filter. 4.static void ApplyBlobMaskingOnImage(ref AForge.Imaging.UnmanagedImage image, ref System.Drawing.Rectangle[] rects) { if (rects == null || rects.Length == 0) return;

        var imageMask = AForge.Imaging.UnmanagedImage.Create(
            image.Width, image.Height, image.PixelFormat);

        foreach (var rect in rects)
            if (rect != System.Drawing.Rectangle.Empty)
                AForge.Imaging.Drawing.FillRectangle(imageMask, rect, System.Drawing.Color.White);

// I currently use this instead of: //new Invert().ApplyInPlace(imageMask); //new Subtract(imageMask).ApplyInPlace(image);

        // Bug
        new ApplyMask(imageMask).ApplyInPlace(image);
        //new MaskedFilter(new Sepia(), image).ApplyInPlace(image);
    }

What is the expected output? What do you see instead? *When I pass an image in grayscale(8bpp) everything is ok, But pass a RGB(24bpp) image, unexpected region is show.

What version of the product are you using? 2.2.5

Please provide any additional information below.

One thing to note from your code: //new MaskedFilter(new Sepia(), image).ApplyInPlace(image); This looks very strange to me. You use same image as mask image, as input image for processing. I would not expect this.

The code around ApplyMask() seem to be valid. Did you try doing these steps first in IPLab application and see if it all works (just to exclude coding errors): http://www.aforgenet.com/projects/iplab/

I believe you are telling about 24 bpp and 8 bpp input image, not mask. Since mask can be only 8 bpp (you would get exception otherwise).

Very appreciate for your prompt response.

Correct: //new MaskedFilter(new Sepia(), imageMask).ApplyInPlace(image );

Sorry about this issue, cause I use this function:

    static void ApplyBlobMaskingOnImage(ref

AForge.Imaging.UnmanagedImage image, ref System.Drawing.Rectangle[] rects) { if (rects == null || rects.Length == 0) return;

        var imageMask = AForge.Imaging.UnmanagedImage.Create(
            image.Width, image.Height, *image.PixelFormat*);

        foreach (var rect in rects)
            if (rect != System.Drawing.Rectangle.Empty)
                AForge.Imaging.Drawing.FillRectangle(imageMask, rect,

System.Drawing.Color.White);

        new ApplyMask(imageMask).ApplyInPlace(image);
        //new MaskedFilter(new Sepia(), imageMask).ApplyInPlace(image);

}

But I don't care that I must use Gray image mask as overlay instead RGB image mask even for RGB image! and "ApplyMask" or "MaskedFilter" don't throw exception about this!

. . . var imageMask = AForge.Imaging.UnmanagedImage.Create( image.Width, image.Height, System.Drawing.Imaging.PixelFormat.Format8bppIndexed); . .

.

Be nice if the filters check for overlay that be Grayscale.

Very very good and thanks. Good luck. Hashem.

The documentation clearly says about the fact that mask must be 8bpp: http://www.aforgenet.com/framework/docs/html/fbcfe698-f9a2-a672-c334-309831030643.htm

The code tells an exception should be thrown if it is not 8bpp: https://code.google.com/p/aforge/source/browse/trunk/Sources/Imaging/Filters/Other/ApplyMask.cs

You are right and this issue is discarded, I'm sorry. But I couldn't find the try catch blocks in my code that prevent to appear this exception. All of them content has this command:"MessageBox.Show(ex.Message);".

​____________________________________________________ ​​ if ( ( maskImage != null ) && ( maskImage.PixelFormat != PixelFormat.Format8bppIndexed ) ) { throw new ArgumentException( "The mask image must be 8 bpp grayscale image." ); }

            *maskImage **= value;*
            unmanagedMaskImage = null;
            mask = null;

​ _____________________________________________________​ I found it!!!! ​Bug is exist! ​Check is before set value!​

​

​

On Thu, Aug 13, 2015 at 5:59 PM, Hashem Zavvari wrote:

Yes, good spot. Somehow the silly bug has survived.