概況：

灰度變換通過(guò)對(duì)原圖像素值重新分配實(shí)現(xiàn), 目的是使圖像中表現(xiàn)較暗的像素值, 通過(guò)灰度變換函數(shù)映射的方法使較暗的像素值增大, 這樣圖像的亮度就提高了。增強(qiáng)處理并不能使原始圖像信息增加, 其結(jié)果只能增強(qiáng)對(duì)某種信息的辨別能力, 而這種處理有可能損失一些其他信息。但是, 只要提高了圖像的視覺(jué)特性, 增強(qiáng)圖像的目的就達(dá)到了。

幾個(gè)概念

1、灰度：對(duì)于通常所謂的黑白圖像，把黑色和白色之間按對(duì)數(shù)關(guān)系分為若干等級(jí)稱(chēng)為灰度。灰度分為256階，用灰度表示的圖像稱(chēng)作灰度圖.在圖像中用0~255表示，0是全黑，255是全白

2、對(duì)比度：對(duì)比度值一幅圖像中敏感區(qū)域最亮的白和最暗的黑之間的不同亮度層級(jí)的測(cè)量，差異范圍越大代表對(duì)比月大。好的對(duì)比率120:1就可以容易的顯式生動(dòng)、豐富的色彩，當(dāng)對(duì)比率達(dá)到300:1時(shí)便可以支持各階的顏色。

(1)線(xiàn)性變換:

通過(guò)建立灰度映射來(lái)調(diào)整源圖像的灰度。

k>1增強(qiáng)圖像的對(duì)比度;k=1調(diào)節(jié)圖像亮度，通過(guò)改變d值達(dá)到調(diào)節(jié)亮度目的;0

i = imread('theatre.jpg');

i = im2double(rgb2gray(i));

[m,n]=size(i);

%增加對(duì)比度

Fa = 1.25; Fb = 0;

O = Fa.*i + Fb/255;

figure(1), subplot(221), imshow(O);

title('Fa = 1.25, Fb = 0, contrast increasing');

figure(2),subplot(221), [H,x]=imhist(O, 64);

stem(x, (H/m/n), '.');

title('Fa = 1.25, Fb = 0, contrast increasing');

%減小對(duì)比度

Fa =0.5; Fb = 0;

O = Fa.*i + Fb/255;

figure(1), subplot(222),imshow(O);

title('Fa = 0.5, Fb = 0, contrast decreasing');

figure(2), subplot(222), [H,x] = imhist(O, 64);

stem(x, (H/m/n), '.');

title('Fa = 0.5, Fb = 0, contrast decreasing');

%線(xiàn)性亮度增加

Fa = 0.5; Fb = 50;

O = Fa.*i + Fb/255;

figure(1), subplot(223), imshow(O);

title('Fa = 0.5, Fb = 50, brightness control');

figure(2), subplot(223), [H,x]=imhist(O,64);

stem(x, (H/m/n), '.');

title('Fa = 0.5, Fb = 50, brightness control');

%反相顯示

Fa = -1; Fb = 255;

O = Fa.*i + Fb/255;

figure(1), subplot(224), imshow(O);

title('Fa = -1, Fb = 255, reversal processing');

figure(2), subplot(224),[H,x]=imhist(O, 64);

stem(x, (H/m/n), '.');

title('Fa = -1, Fb = 255, reversal processing');

(2)對(duì)數(shù)變換：

增強(qiáng)低灰度，減弱高灰度值。

i = imread('theatre.jpg');

i = rgb2gray(i);

i = double(i);

out1 = log(1+i)/0.065;

out2 = log(1+i)/0.035;

out1(find(out1>255)) = 255;

out2(find(out2>255)) = 255;

out1 = uint8(out1);

out2 = uint8(out2);

(3)冪次變換：

次數(shù)小于1時(shí)，增強(qiáng)低灰度，減弱高灰度;次數(shù)大于1時(shí)增強(qiáng)高灰度，減弱低灰度。

i = rgb2gray(imread('theatre.jpg'));

i = double(i);

y1 = 255*(i/255).^2.5;

y2 = 255*(i/255).^0.4;

y1 = uint8(y1);

y2 = uint8(y2);

(4) 指數(shù)變換：

增強(qiáng)高灰度，減弱低灰度。

i = imread('theatre.jpg');

i = rgb2gray(i);

i = double(i);

y1 = 1.5.^(i*0.070)-1;

y2 = 1.5.^(i*0.050)-1;

y1(find(y1>255)) = 255;

y2(find(y2>255)) = 255;

y1 = uint8(y1);

y2 = uint8(y2);

(5)灰度拉伸：

有時(shí)圖像灰度集中在某小塊區(qū)域，需要改變圖像對(duì)比度。

i = imread('theatre.jpg');

i = rgb2gray(i);

L = imadjust(i,[ ],[50/255;150/255]);

J = imadjust(L,[50/255;150/255 ],[20/255;230/255]);

(6)灰度均衡：

i = rgb2gray(imread('theatre.jpg'));

LC = imadjust(i,[ ],[50/255;150/255]);

HE1 = histeq(LC);%均衡函數(shù)

(7)直方圖規(guī)定化：

實(shí)現(xiàn)局部的灰度均衡。

img = rgb2gray(imread('theatre.jpg'));

img_ref = rgb2gray(imread('rpic.jpg'));%參考圖，按照這個(gè)的的直方圖進(jìn)行規(guī)定化

[hgram, x] = imhist(img_ref);

J = histeq(img, hgram);