How to do it...

The conversion of images between different color spaces is easily done through the use of the cv::cvtColor OpenCV function, as follows:

1. Let's convert the input image to the CIE L*a*b* color space at the beginning of the process method:

cv::Mat ColorDetector::process(const cv::Mat &image) { 
 
     // re-allocate binary map if necessary 
     // same size as input image, but 1-channel 
     result.create(image.rows,image.cols,CV_8U); 
 
 
     // Converting to Lab color space  
     cv::cvtColor(image, converted, cv::COLOR_BGR2Lab); 
 
 
     // get the iterators of the converted image  
     cv::Mat_<cv::Vec3b>::iterator it=  
                 converted.begin<cv::Vec3b>(); 
     cv::Mat_<cv::Vec3b>::iterator itend=  
                 converted.end<cv::Vec3b>(); 
     // get the iterator of the output image  
     cv::Mat_<uchar>::iterator itout= result.begin<uchar>(); 
 
     // for each pixel 
     for ( ; it!= itend; ++it, ++itout) { 
     ... 

1. The converted variable contains the image after color conversion. In the ColorDetector class, it is defined as a class attribute:

class ColorDetector { 
 
  private: 
     // image containing color converted image 
     cv::Mat converted; 

1. You also need to convert the input target color. You can do this by creating a temporary image that contains only one pixel. Note that you need to keep the same signature as in the earlier recipes; that is, the user continues to supply the target color in RGB:

     // Sets the color to be detected 
     void setTargetColor(unsigned char red,  
           unsigned char green, unsigned char blue) { 
 
         // Temporary 1-pixel image 
         cv::Mat tmp(1,1,CV_8UC3); 
         tmp.at<cv::Vec3b>(0,0)= cv::Vec3b(blue, green, red); 
 
         // Converting the target to Lab color space  
         cv::cvtColor(tmp, tmp, cv::COLOR_BGR2Lab); 
 
         target= tmp.at<cv::Vec3b>(0,0); 
     } 

If the application of the preceding recipe is compiled with this modified class, it will now detect the pixels of the target color using the CIE L*a*b* color model.