【MATLAB数字图像处理】将灰度图生成伪彩色图像

àì夳堔傛蜴生んèń 2021-12-22 11:57 4600阅读 1赞

有时在Matlab中我们需要将灰度图像添加彩色变成伪彩色图像。

通常做法为：

I=imread('lenna.jpg');
    I=rgb2gray(I);
    figure
    imshow(I);
    rgb = ind2rgb(gray2ind(I,255),jet(255));
    figure
    imshow(rgb)

灰度图像：

![watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L2xodzE5OTMxMjAx_size_16_color_FFFFFF_t_70][]

添加伪彩色后的图像：

![watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L2xodzE5OTMxMjAx_size_16_color_FFFFFF_t_70 1][]

有时我们需要将255替换为灰度图像中的颜色数。如果您不知道灰度图像中的颜色数量，可以通过以下方式轻松找到：

n = size(unique(reshape(im,size(im,1)*size(im,2),size(im,3))),1);

如果im已经是RGB图像，那么处理起来有点过于复杂。如果你不介意RGB图像以uint8而不是double的形式出现，可以使用以下更快的数量级同样可以实现添加伪彩色的效果：

rgb = label2rgb(gray2ind(im,255),jet(255));

还有一种办法实现对灰度图像添加伪彩色，就是将亮度低的映射为蓝色(冷色)，亮度高的映射为红色（暖色）。  
方法是分别用三个矩阵保存，三个通道的灰度值。每个通道的灰度值，有不同的映射函数。

对于：

<table style="width:500px;"> 
 <tbody> 
  <tr> 
   <td>&nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; 输入灰度级</td> 
   <td>输出彩色</td> 
  </tr> 
  <tr> 
   <td>&nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp;&nbsp; 0～63 &nbsp;1/4 &nbsp;</td> 
   <td>&nbsp;蓝色</td> 
  </tr> 
  <tr> 
   <td>&nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; 64～127 2/4 &nbsp;</td> 
   <td>&nbsp;紫色</td> 
  </tr> 
  <tr> 
   <td>&nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp;&nbsp; 128～191 3/4 &nbsp; &nbsp;</td> 
   <td>&nbsp;黄色</td> 
  </tr> 
  <tr> 
   <td>&nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; 192～255 &nbsp;4/4&nbsp;</td> 
   <td>&nbsp;红色</td> 
  </tr> 
 </tbody> 
</table>

clc;
    clear;
    grayImage=rgb2gray(imread('D:\Matlab test\lenna.jpg'));
    figure,imshow(grayImage);
    [row,col]=size(grayImage);
    range=255;%每个通道的最大灰度值
    
    R = zeros(row,col);
    G = zeros(row,col);
    B = zeros(row,col);
    
    for i=1:row
         for j=1:col
            if grayImage(i,j)<=range/4% [0,64]偏蓝
                R(i,j)=0;
                G(i,j)=4*grayImage(i,j);
                B(i,j)=range;
                else if grayImage(i,j)<=range/2% (64,128] 偏紫
                     R(i,j)=0;
                     G(i,j)=range;
                     B(i,j)=-4*grayImage(i,j)+2*range;
                    else if grayImage(i,j)<=3*range/4% (128, 192]
                            R(i,j)=4*grayImage(i,j)-2*range;
                            G(i,j)=range;
                            B(i,j)=0;
                        else
                            R(i,j)=range;
                            G(i,j)=4*(range-grayImage(i,j));
                            B(i,j)=0;
                        end
                    end
            end
         end
    end
    
    out = zeros(row,col);
    
    for i=1:row
        for j=1:col
                  out(i,j,1)=R(i,j);
                  out(i,j,2)=G(i,j);
                  out(i,j,3)=B(i,j);
        end
    end
    out=out/256;
    figure,imshow(out);

运行程序，效果如图：

![watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L2xodzE5OTMxMjAx_size_16_color_FFFFFF_t_70 2][]

参考博文：1）[https://blog.csdn.net/yangyangyang20092010/article/details/51168630/][https_blog.csdn.net_yangyangyang20092010_article_details_51168630]

2）[https://blog.csdn.net/qjt19950610/article/details/89575611][https_blog.csdn.net_qjt19950610_article_details_89575611]

[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L2xodzE5OTMxMjAx_size_16_color_FFFFFF_t_70]: /images/20211220/99b8c2b8ccfe402e9008068771058867.png
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L2xodzE5OTMxMjAx_size_16_color_FFFFFF_t_70 1]: /images/20211220/e72631cc18f3457490de84d4faf0f991.png
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L2xodzE5OTMxMjAx_size_16_color_FFFFFF_t_70 2]: /images/20211220/76919d6da2054c77a5ebdd6634f2e187.png
[https_blog.csdn.net_yangyangyang20092010_article_details_51168630]: https://blog.csdn.net/yangyangyang20092010/article/details/51168630/
[https_blog.csdn.net_qjt19950610_article_details_89575611]: https://blog.csdn.net/qjt19950610/article/details/89575611