1 簡(jiǎn)介

為了實(shí)現(xiàn)對(duì)圖像的降噪，兼顧全局整體增強(qiáng)和局部細(xì)節(jié)增強(qiáng)，首先定義輸入圖像的對(duì)比度度量模型和亮度度量模型，然后構(gòu)造標(biāo)量權(quán)重映射，并進(jìn)行歸一化處理，接著綜合利用高斯金字塔和拉普拉斯金字塔，制定融合規(guī)則，利用融合金字塔重構(gòu)圖像，實(shí)現(xiàn)圖像增強(qiáng)。

2 部分代碼

% 對(duì)圖像進(jìn)行灰度世界白平衡處理% 再進(jìn)行直方圖自動(dòng)對(duì)比度調(diào)整function ?output = simple_color_balance(image)num = 255;r = image(:, :, 1);g = image(:, :, 2);b = image(:, :, 3);Ravg = mean(mean(r));Gavg = mean(mean(g));Bavg = mean(mean(b));avgRGB = [Ravg, Gavg, Bavg];grayValue = (Ravg + Gavg + Bavg)/3;scaleValue = grayValue./avgRGB;% newI(:,:,1) = scaleValue(1) * r;% newI(:,:,2) = scaleValue(2) * g;% newI(:,:,3) = scaleValue(3) * b;% % figure,imshow(newI);satLevel = ?0.001 * scaleValue;[m,n,p] = size(image);imgRGB_orig = zeros(p, m*n);for i = 1 : p ? imgRGB_orig(i, : ) = reshape(double(image(:, :, i)), [1, m * n]);endimRGB = zeros(size(imgRGB_orig));%直方圖對(duì)比度調(diào)整for ch = 1 : p ? ?q = [satLevel(ch), 1 - satLevel(ch)]; ? ?tiles = quantile(imgRGB_orig(ch, :), q); ? ?temp = imgRGB_orig(ch, :); ? ?temp(find(temp < tiles(1))) = tiles(1); ? ?temp(find(temp > tiles(2))) = tiles(2); ? ?imRGB(ch, :) = temp; ? ?pmin = min(imRGB(ch, :)); ? ?pmax = max(imRGB(ch, :)); % fac(A) = Amin + (A - Alow) * (Amax -Amin)/(Ahigh - Alow)% 對(duì)于8bit圖像，Amin = 0，Amax = 255 ? ?imRGB(ch, :) ?= (imRGB(ch, :) - pmin) * num /(pmax - pmin);endoutput = zeros(size(image));for i = 1 : p ? ? ? ?output(:, :, i) = reshape(imRGB(i, :), [m, n]); endoutput = uint8(output);end

3 仿真結(jié)果

4 參考文獻(xiàn)

[1]馬永強(qiáng), 王順利, 孫偉,等. 基于高斯金字塔和拉普拉斯金字塔融合的圖像對(duì)比度增強(qiáng)算法研究[J]. 信息與電腦, 2018(4):3.

博主簡(jiǎn)介：擅長(zhǎng)智能優(yōu)化算法、神經(jīng)網(wǎng)絡(luò)預(yù)測(cè)、信號(hào)處理、元胞自動(dòng)機(jī)、圖像處理、路徑規(guī)劃、無(wú)人機(jī)等多種領(lǐng)域的Matlab仿真，相關(guān)matlab代碼問(wèn)題可私信交流。

部分理論引用網(wǎng)絡(luò)文獻(xiàn)，若有侵權(quán)聯(lián)系博主刪除。