芯片复制MMBEBHE代码
| 芯片复制input_image = imread('图片地址.tif'); % Step 1 计算每个阈值级别的 AMBE hist = imhist(input_image); % 计算直方图 total_pixels = numel(input_image);% 获取图像中的总像素数n ambe_values = zeros(256, 1); % 初始化数组以便储存AMBE的值 % 芯片复制根据阈值将输入的直方图一分为二 for threshold =[] lower_hist = hist(1:threshold + 1); upper_hist = hist(threshold + 2:end); % 检查是否有任何直方图为空 if isempty(lower_hist) || isempty(upper_hist) ambe_values(threshold + 1) = inf; %将AMBE设置为无限大 else lower_cdf = cumsum(lower_hist); upper_cdf = cumsum(upper_hist);% 计算nk % 按像素总数归一化Pr（rk）=nk/n lower_cdf_normalized = lower_cdf / total_pixels; upper_cdf_normalized = upper_cdf / total_pixels; % 计算两个区域的平均强度 mean_lower = sum((0:threshold) .* lower_hist) / sum(lower_hist); mean_upper = sum((threshold + 1:255) .* upper_hist) / sum(upper_hist); % 计算绝对平均亮度误差 （AMBE） ambe_values(threshold + 1) = abs(mean_upper - mean_lower); end end % STEP 2 找到产生最小 MBE 的阈值水平 XT [~, optimal_threshold] = min(ambe_values); XT = optimal_threshold - 1; % Step 3 芯片复制根据找到的 XT将输入直方图一分为二，独立均衡化 % 根据最佳阈值分割图像 lower_region = input_image <= XT; upper_region = input_image > XT; % 两个区域单独直方图均衡化 lower_eq = histeq(input_image(lower_region), 256); upper_eq = histeq(input_image(upper_region), 256); % 合并两个均衡化区域得出最终结果 result_image = input_image; result_image(lower_region) = lower_eq; result_image(upper_region) = upper_eq; |
芯片解密
