圖像分割技術(shù)的matlab實現(xiàn)

1、s and Settings/Administrator/桌面/代碼/第23章/代碼23-s and Settings/Administrator/桌面/代碼/第23章/代碼23-figure, imshow(f),title(Original figure,imshow(PF),title(Prewitt figure,imshow(RF),title(Roberts n of Gaussian (LoG) figure,imshow(CF),title(Canny s and Settings/Administrator/桌面/代碼/第23章/代碼23-1.txt2011-5-16 s a

2、nd Settings/Administrator/桌面/代碼/第23章/代s and Settings/Administrator/桌面/代碼/第23章/代碼23-% 使用垂直Sobel算子，自動選擇閾VSFATfigure, imshow(f),title(Original figure,imshow(VSFAT),title(Sobel Filter - Automatic figure,imshow(SFST),title(Sobel Filter (Horizontal and Vertical); % 顯示邊緣探測圖像s45=-2 -1 0;-1 0 1;0 1 2;figure,

3、imshow(SFST45),title(SobelFilter(45sm45=0 1 2;-1 0 1;-2 -1 0;figure,imshow(SFSTM45),title(Sobel Filter (-45 s and Settings/Administrator/桌面/代碼/第23章/代碼23-2.txt2011-5-16 s and Settings/Administrator/桌面/代碼/第23章/代碼23-Is and Settings/Administrator/桌面/代碼/第23章/代碼23-I=rotI=e(I,33,crop); % 圖像旋轉(zhuǎn)，該函數(shù)具體用法在本書13.

4、3.3有介紹fig1=BW=H,theta,rho = figure, imshow(imadjust(mat2gray(H),XData,theta,YData,rho,.% 設(shè)置顯示屬性 xlabel(theta(degrees),ylabel(rho); axis on, axis normal, hold on; P = houghpeaks(H,5,threshold,ceil(0.3*max(H(:); x = theta(P(:,2);y=s and Settings/Administrator/桌面/代碼/第23章/代碼23-3.txt2011-5-16 s and Setti

5、ngs/Administrator/桌面/代碼/第23章/代碼23-全局閾值 s and Settings/Administrator/桌面/代碼/第23章/代碼23-全局閾值 display(Threshold(T)-Iterative); figure,imshow(f),title(Original title(Global Thresholding - Iterative Method); % 圖像標題display(Threshold(T) - Otsus Method); title(Global Thresholding - Otsus Method); display(Thre

6、shold(T) - Local Thresholding); s and Settings/Administrator/桌面/代碼/第23章/代碼23-4.txt2011-5-16 s and Settings/Administrator/桌面/代碼/第23章/代碼23-title(Original Image); title(Negative Image); title(Watershed - Distance s and Settings/Administrator/桌面/代碼/第23章/代碼23-title(Original Image); title(Negative Image);

7、 title(Watershed - Distance ed - Watershed and original sq=sqrt(imfilter(fd,h,replicate).2+imfilter(fd,h,replicate).2); % title(Watershed-Gradient); ed - Watershed and original title(Watershed-MarkerControlled); title(Watershed-GradientandMarkerControlled); ed - Watershed (GM) and original image);s

8、and Settings/Administrator/桌面/代碼/第23章/代碼23-5.txt2011-5-16 s and Settings/Administrator/桌面/代碼/第23章/代碼23-s and Settings/Administrator/桌面/代碼/第23章/代碼23-if numel(s)=1 for k=1:length(s1)gtitle(Segmented Image - Region Growing);s and Settings/Administrator/桌面/代碼/第23章/代碼23-6.txt2011-5-16 s and Settings/Admi

9、nistrator/桌面/代碼/第23章/s and Settings/Administrator/桌面/代碼/第23章/代碼23-I=8(11112361121456111177711116652022202212320222220547202220209124020222020131415Ss and Settings/Administrator/桌面/代碼/第23章/代碼23-7.txt2011-5-16 s and Settings/Administrator/桌面/代碼/第23章/代碼23-I=s and Settings/Administrator/桌面/代碼/第23章/代碼23-

10、I=Sblocks = fordim=512256128643216842numblocks=length(find(S=dim); if (numblocks 0)values=8(1),dimdimvalues(2:dim,2:dim,:) = blocks = blocks(end,1:end)=blocks(1:end,end)=1; figure, s and Settings/Administrator/桌面/代碼/第23章/代碼23-8.txt2011-5-16 s and Settings/Administrator/桌面/代碼/第23章/代碼23-he = imread(he

11、stain.png); Image courtesy of Alan Partin, Johns Hopkins University, . cform=ab = double(lab_he(:,:,2:3);nrowss and Settings/Administrator/桌面/代碼/第23章/代碼23-he = imread(hestain.png); Image courtesy of Alan Partin, Johns Hopkins University, . cform=ab = double(lab_he(:,:,2:3);nrows=size(ab,1); ncols=si

12、ze(ab,2);ab = resh nColors=3;% 重復(fù)聚類3次，以避免局部最小pixel_labels=ols); title(image labeled by cluster index); segmented_images = cell(1,3);rgb_label = repmat(pixel_labels,1 1 3); for k = 1:nColorscolor = he; color(rgb_label=k)=0;segmented_imagesk=color; title(objects in cluster 1); title(objects in cluster

13、 2); title(objects in cluster 3);mean_cluster_value = mean(cluster_center,2); tmp, idx = sort(mean_cluster_value); blue_cluster_num = idx(1);L=blue_idx = find(pixel_labels = blue_cluster_num); L_blue = L(blue_idx);is_light_blue=nuclei_labels=nuclei_labels(blue_idx(is_light_blue=false) = nuclei_labels = repmat(nuclei_labels,1 1 3); blue_nuclei = he;s and Settings/Administrator/桌面/代碼/第23