Original: Win8 Metro (C #) Digital image processing--2.57 image binary with one-dimensional maximum entropy method
[ function name ]
One-dimensional maximum entropy method image binary writeablebitmap entropymaxthsegment ( writeablebitmap src"
[ algorithm description ]
One-dimensional maximum entropy method image segmentation is to use the gray distribution density function of image to define the information entropy of image, by optimizing certain entropy
The criterion obtains the threshold value corresponding to the maximum entropy, thus the method of image segmentation.
Algorithm process:
1, for a grayscale image, the gray range is [0,l-1], the minimum gray level of the image is obtained , themaximum gray level max;
[ function Code ]
< Span style= "Font-family:calibri" >
<summary>//Entropy Max method of image segmention. </summary>//<param name= "src" >the source iamge.</param>//<returns></ret urns> public static WriteableBitmap entropymaxthsegment (WriteableBitmap src)////One-dimensional entropy maximum threshold segmentation { if (src! = null) {int w = src. Pixelwidth; int h = src. Pixelheight; WriteableBitmap dstimage = new WriteableBitmap (w, h); byte[] temp = src. Pixelbuffer.toarray (); Byte[] Tempmask = (byte[]) temp. Clone (); Defining grayscale image information storage variables int[] Srcdata = new int[w * h]; Define a threshold variable int Th = 0; Define the histogram storage variable int[] histogram = new int[256]; Define the Entropy value variable double Ht = 0.0; Double Hl = 0.0; Double sigma = 0.0; Define grayscale Maximum variables int max = 0; int min = 255; Define temporary variable Double t = 0.0, pt = 0.0, Tempmax = 0.0; int TEMPV = 0; for (int j = 0; J < H; j + +) {for (int i = 0; i < W; i++) { TEMPV = (int) ((double) tempmask[i * 4 + J * W * 4] * 0.114 + (double) tempmask[i * 4 + 1 + J * W * 4 ] * 0.587 + (double) tempmask[i * 4 + 2 + J * W * 4] * 0.299); Srcdata[i + J * W] = TEMPV; histogram[tempv]++; if (TEMPV > Max) {max = TEMPV; } if (TEMPV < min) {min = TEMPV; }}} for (int i = min; i < Max; i++) {t = (double) ((Double) Histogram[i]/(Double) (w * h)); if (T > 0.00000001) {Hl + =-T * MATH.LOG10 (T); } else continue; } for (int i = min; i < Max; i++) {t = (double) (double) histogram[ I]/(double) (w * h)); PT + = t; if (T > 0.00000001) {Ht + =-T * MATH.LOG10 (T); Sigma = Math.log10 (PT * (1-PT)) * ht/pt + (HL-HT)/(1-PT); if (Sigma > Tempmax) {tempmax = (int) sigma; Th = i; }} else continue; } for (int j = 0; J < H; j + +) { for (int i = 0; i < W; i++) {Temp[i * 4 + J * W * 4] = temp[i * 4 + 1 + j * W * 4] = temp[i * 4 + 2 + J * W * 4] = (byte) (Srcdata[i + j * W] < Th? 0:255); }} Stream Stemp = DstImage.PixelBuffer.AsStream (); Stemp.seek (0, Seekorigin.begin); Stemp.write (temp, 0, W * 4 * h); return dstimage; } else {return null; } }
Win8 Metro (C #) Digital image processing--2.57 image binary with one-dimensional maximum entropy method