MATLAB中對(duì)比兩幅圖像各種直方圖的程序-

1、MATLAB中對(duì)比兩幅圖像各種直方圖的程序文章來(lái)源:中國(guó)科學(xué)院計(jì)算技術(shù)研究所作者:劉偉民- 該文章講述了MATLAB中對(duì)比兩幅圖像各種直方圖的程序. MATLAB中對(duì)比兩幅圖像各種直方圖的程序,程序如下。figure(1;hold on;aa = 'e:d1.jpg'bb = 'e:d4.jpg'imhist(rgb2gray(imread(aa;hold off;figure(2;hold on;imhist(rgb2gray(imread(bb;hold off;imga = imread(aa;imgb = imread(bb;figure(3;hold

2、on;title('imga1'imhist(imga(:,:,1;hold off;figure(4;hold on;title('imga2'imhist(imga(:,:,2;hold off;figure(5;hold on;title('imga3'imhist(imga(:,:,3;hold off;figure(6;hold on;title('imgb1'imhist(imgb(:,:,1;hold off;figure(7;hold on;title('imgb2'imhist(imgb(:,:,

3、2;hold off;figure(8;hold on;title('imgb3'imhist(imgb(:,:,3;hold off;3 matlab圖像配準(zhǔn)的方法-醫(yī)學(xué)圖像配準(zhǔn)目前,圖像配準(zhǔn)技術(shù)在醫(yī)學(xué)領(lǐng)域應(yīng)用得相當(dāng)廣泛。醫(yī)學(xué)圖像配準(zhǔn)具有很重要的臨床應(yīng)用價(jià)值。對(duì)各種使用相同或者不同的成像手段所獲得的醫(yī)學(xué)圖像進(jìn)行配準(zhǔn)不僅可以應(yīng)用于醫(yī)療診斷,也可以應(yīng)用于手術(shù)計(jì)劃的制定、治療計(jì)劃的制定、病理情況的跟蹤和治療效果的評(píng)價(jià)等各個(gè)方面。下面給出一個(gè)具體的圖像配準(zhǔn)的例子來(lái)形象說(shuō)明配準(zhǔn)的應(yīng)用?！?】醫(yī)學(xué)圖像配準(zhǔn)。(1 讀取圖像數(shù)據(jù)。因?yàn)樵磮D像以矩陣形式存在一個(gè)二進(jìn)制的文件里,用fread可將其

4、讀取到變量矩陣中。將讀取文件編制成一個(gè)子函數(shù)(RTIread.m。源代碼如下:function imMatrix=RTIread(FILENAME,SIZE%RTIread Read the image matrix from binary "Registration Test Image" file.% imMatrix=RTIread(FILENAME,SIZE opens the file FILENAME, and reads the% number of elements specified by SIZE.% FILENAME is a string conta

5、ining the name of the file to be opened.% Valid entries for SIZE are:% N read N elements into a column vector.% inf read to the end of the file.% M,N read elements to fill an M-by-N matrix, in column order.% N can be inf, but M can't.% It returns the image matrix.fid=fopen(FILENAME,'r'im

6、Matrix=fread(fid,SIZE,'uint8=>uint8'fclose(fid;運(yùn)行以下代碼讀取圖像矩陣:base=RTIread('sag1.tif',256,256;input=RTIread(sag2.tif,256,256;(2 選取匹配點(diǎn)。根據(jù)預(yù)定的配準(zhǔn)方法,選定足夠的匹配點(diǎn)對(duì)。運(yùn)行下列代碼:cpselect(input,base; %please select 15 points for test.注意:這里允許選擇15個(gè)點(diǎn)進(jìn)行測(cè)試。操作很簡(jiǎn)單,只需注意選點(diǎn)要均勻布開(kāi),以增加其代表性。選定完畢,再選擇FileSave Points

7、 to Workspace命令將數(shù)據(jù)保存到工作區(qū)中。Workspace立刻多出兩個(gè)N×2的數(shù)組(其中N為選定的匹配點(diǎn)對(duì)數(shù),分別為input_points和base_points。(3 利用十字相關(guān)法調(diào)整選定了的匹配點(diǎn)。如果對(duì)上面一個(gè)步驟的選點(diǎn)工作不滿意,可以進(jìn)行這一步。如果感覺(jué)尚可,則跳過(guò)這一步,以減小運(yùn)算量。運(yùn)行代碼:input_points_corr = cpcorr(input_points,base_points,input,base;%優(yōu)化選擇點(diǎn)的作用input_points_corr為優(yōu)化后在輸入圖片的對(duì)應(yīng)匹配點(diǎn)。(4 計(jì)算變換公式的參數(shù)。利用cp2tform選定變換類型

8、(配準(zhǔn)方法,計(jì)算變換參數(shù)。以下只需選定一種即可。% (1 not Fine-tune pointsTlinear = cp2tform(input_points,base_points,'linear conformal'Taffine = cp2tform(input_points,base_points,'affine'Tprojective = cp2tform(input_points,base_points,'projective'Tpolynomial2 = cp2tform(input_points,base_points,'polynomial',2;Tpolynomial3 = c