算法实现Alg4-Infrared small target detection via multi-direction line reconstruction ···

《Infrared small target detection via multi-direction line reconstruction and information entropy-induced suppression 》

Infrared physics & technology 

impact factor：1.713

引用：Shang K, Sun X, Tian J, et al. Infrared small target detection via line-based reconstruction and entropy-induced suppression[J]. Infrared Physics & Technology, 2016, 76:75-81.

适合简单图像，复杂云层不ok~

clear all;
close all;
clc;

img=imread('E:\A 研究生学习\我的论文~\image\6.jpg');
n=size(size(img));
if n(1,2)==2;
    im_gray=img;
else im_gray=rgb2gray(img);
end
figure;
imshow(im_gray,'Colormap',jet(255));



f=im_gray;
[m,n]=size(f);
lm=2;
D=f;
  for p=1:12;
      seta=p*pi/6;
    for l=2:2:lm
     for j=lm+1:n-lm
        for i=lm+1:m-lm 
          D(i,j,l,p)=f(floor(i+lm*cos(seta)),floor(j+lm*sin(seta)));
        end
     end
     D(:,:,p)=min(D(:,:,2,p),D(:,:,l,p));
    end
    maxD=max(D(:,:,1),D(:,:,p));
  end
  final=f-maxD;
  figure;
  final=imcrop(final,[lm+1,lm+1,n-2*lm-1,m-2*lm-1]);
  imshow(final,[]);

  figure; %final
img_1=imresize(final,1); %图像放缩
[m3,n3]=size(img_1);
[x,y]=meshgrid(1:1:n3,1:1:m3);
for i=1:m3
    for j=1:n3
        R(i,j)=img_1(i,j);
    end
end
mesh(double(x),double(y),double(R));
xlabel('x');
ylabel('y');

原图：

伪色彩图像：

线性残差：

没有做熵，没有算最后用熵加权的final result。