Chromadist.docVIP

Chromadist: function [cb, cr] = ChromaDist(filename) % return the chromatic components of the image %回返图像的彩色的成份 % low pass filtering is carried out to remove noise % 低点途径过滤是被运的外面对距离噪音 im = imread(filename); imycc = rgb2ycbcr(im); lpf = 1/9 * ones(3); cb = imycc(:,:,2); cb = filter2(lpf, cb); cb = reshape(cb, 1, prod(size(cb))); cr = imycc(:,:,3); cr = filter2(lpf, cr); cr = reshape(cr, 1, prod(size(cr))); ColorDistPlot.m : function color = ColorDistPlot(cb, cr) % plot the chromatic color distribution of the image % 情节彩色的彩色分配那图像 chroma = zeros(256); cb = round(cb); cr = round(cr); for i = 1:length(cb) chroma(cb(i), cr(i)) = chroma(cb(i), cr(i)) + 1; end surf(chroma) SegmentSkin.m function [skin1, skin2, optimalThreshold] = SegmentSkin(filename, bmean, rmean, brcov) % Assume the skinmodel.m is run % 承担 skinmodel.m 是奔跑 % Produce two images, skinlikelihood greyscale image, skin1 % 农产品二个图像, skinlikelihood greyscale 图像, 皮肤 1 % and skin segment binary image, skin2 % 而且皮肤片段二进位的图像, 皮肤 2 im = imread(filename); imycbcr = rgb2ycbcr(im); dim = size(im); skin1 = zeros(dim(1), dim(2)); for i = 1:dim(1) for j = 1:dim(2) cb = double(imycbcr(i,j,2)); cr = double(imycbcr(i,j,3)); x = [(cb-bmean); (cr-rmean)]; skin1(i,j) = exp(-0.5* x*inv(brcov)* x); end end lpf= 1/9*ones(3); skin1 = filter2(lpf,skin1); skin1 = skin1./max(max(skin1)); % Adaptive Thresholding % 适合的门槛 previousSkin2 = zeros(i,j); changelist = []; for threshold = 0.55:-0.1:0.05 skin2 = zeros(i,j); skin2(find(skin1threshold)) = 1; change = sum(sum(skin2 - previousSkin2)); changelist = [changelist change]; previousSkin2 = skin2; end % Finding the optimal threshold % 发现最佳的门槛 [C, I] = min(changelist); optimalThreshold = (7-I)*0.1 skin2 = zeros(i,j); skin2(find(skin1optimalThreshold)) = 1; figure(1) imshow(skin1, [0 1]); figure(2) imshow(skin2, [0 1]); skinmodel.m % 32500 skin samples from 17 color images will be used here to determine the %32500 皮肤样品从 17种颜色图