基于匹配滤波器的血管检测matlab仿真【包括程序操作视频】

function I_corr = getCorrForAllPixels(kernel, I, resolution, threshold, show) %此函数旨在获取中每个像素的相关值 %图像。 %尝试：首先翻转内核，以便我们可以实现 %与之卷积，这是每个像素的有效相关性 %kernel=翻转（翻转（kernel，2））； %现在可以将此内核视为筛选器 % I2 = imfilter(I, kernel); cellArr = generateRotKernels(kernel, resolution); % NOTE: Access cell array elements as cellArr{1, 1} imgArr = cell(1, length(cellArr)); for i=1:length(cellArr) imgArr{1, i} = imfilter(I, cellArr{1, i}); end [m, n] = size(I); % The imgArr elements should have the same size as I I_corr = zeros(size(I)); for i=1:m for j=1:n valArr = zeros(1, length(cellArr)); for k=1:length(cellArr) valArr(1, k) = imgArr{1, k}(i, j); end if(max(valArr) > threshold) % WHY THE THRESHOLD?? - to classify as blood vessel or not I_corr(i, j) = max(valArr); end end end % Visualising what each kernel does to the image if(show ~= false) figure for i=1:length(imgArr) subplot(3, length(imgArr)/3, i), imshow(imgArr{1, i}), title([num2str((i-1)*resolution), ' degrees']); end end % Visualising each kernel if(show ~= false) figure for i=1:length(cellArr) subplot(3, length(cellArr)/3, i), imshow(mat2gray(cellArr{1, i})), title([num2str((i-1)*resolution), ' degrees']); end end end