Opencv desaturation

demo:

Game character recognition , Simple preprocessing
First, prepare a picture for character recognition

Direct use OCR Recognition will identify a lot of junk information , At this time, we need to preprocess the image

It can be seen from the color scale above that the left side should be the junk information behind the background board , On the right is the text information we need , Just remove the background information to get the text information to be processed
Desaturate the image before threshold operation

        static Mat Preprocessing(string file)
        {
    
            var img = Cv2.ImRead(file,ImreadModes.AnyColor);
            if (img.Empty())
                return null;
            Mat result = Mat.Zeros(img.Size(), img.Type());
            img.Desaturation(result);
            //  threshold 
            Cv2.Threshold(result, result, 100, 255, ThresholdTypes.Binary);
            return result;
        }

Acquired image , The background looks good after discarding the background information

Use the desert for this image OCR operation , Fairly good recognition , It is only necessary to limit the recognition range, and the result should be correct , It is very fast when there are few words in the font

Look at the results of online character recognition , slow

Baidu API High precision version , It's slow , The Standard Version return is empty my God

Desaturation

The formula ：

Gray = ( Math.max(Red, Green, Blue) + Math.min(Red, Green, Blue) ) / 2

public static class ImageHelper
{
    
    /// <summary>
    ///  Image desaturation 
    /// </summary>
    /// <param name="src"></param>
    /// <param name="dst"></param>
    public static unsafe void Desaturation(this Mat src, Mat dst)
    {
    
        System.Diagnostics.Debug.Assert(dst != null, "dst not set to an instance of an object.");
        System.Diagnostics.Debug.Assert(src.Channels() >= 3, "src.Channels < 3");
        System.Diagnostics.Debug.Assert(src.Type() == dst.Type(), "src.Type != des.Type");
        System.Diagnostics.Debug.Assert(src.Size() == dst.Size(), "src.Size != des.Size");

            int height = src.Rows;
            int width = src.Cols;
            var channels = src.Channels();

            Parallel.For(0, height, i =>
            {
    
                var ptr = (byte*)src.Ptr(i).ToPointer();
                var dstPtr = (byte*)dst.Ptr(i).ToPointer();
                for (var j = 0; j < width; j++)
                {
    
                    var ptrOffset = j * channels;
                    var max = Math.Max(Math.Max(ptr[ptrOffset], ptr[ptrOffset + 1]), ptr[ptrOffset + 2]);
                    var min = Math.Min(Math.Min(ptr[ptrOffset], ptr[ptrOffset + 1]), ptr[ptrOffset + 2]);
                    var color = (byte)((max + min) * 0.5f);
                    dstPtr[ptrOffset] = color;
                    dstPtr[ptrOffset + 1] = color;
                    dstPtr[ptrOffset + 2] = color;
                }
            });
    }
}