反向投影

一种记录给定图像中的像素点如何适应直方图模型像素分布的方式。

反向投影就是首先计算某一特征的直方图模型，然后使用模型去寻找图像中存在的该特征。

backproject是直接取直方图中的值，即以灰度为例，某种灰度值在整幅图像中所占面积越大，其在直方图中的值越大，backproject时，其对应的像素的新值越大（越亮），反过来，某灰度值所占面积越小，其新值就越小。

源图像 ——> 直方图 ——> backproject图像

calcBackProjectPatch，整个是基于块的形式，利用直方图做匹配，类似于模板匹配，只不过这些模板转换为直方图，而原图中以某点为基准，抠出来作对比的部分也转换为直方图，两个直方图作匹配，匹配的结果作为此点的值。 结果会是一张概率图，概率越大的地方，代表此区域与模板的相似度越高。

Application example

If you have a histogram of flesh color (say, a Hue-Saturation histogram ) —— 皮肤颜色的直方图 , then you can use it to find flesh color areas in an image:

1，Let’s say you have gotten a skin histogram (Hue-Saturation) based on the image below. The histogram besides is going to be our model histogram (which we know represents a sample of skin tonality). You applied some mask to capture only the histogram of the skin area: 模板肤色直方图，需要用mask来提取出需要的肤色区域

    

2，let’s imagine that you get another hand image (Test Image) like the one below: (with its respective histogram):

  

3，What we want to do is to use our model histogram (that we know represents a skin tonality) to detect skin areas in our Test Image. Here are the steps ：

 

• 投影之前要用mask把非兴趣区域的值设置为0

  1. In each pixel of our Test Image (i.e. ), collect the data and find the correspondent bin location for that pixel (i.e. ). 对test image 的每个像素点，找到其直方图中对应的bin值

  2. Lookup the model histogram in the correspondent bin - - and read the bin value. 读取此bin值在模板直方图中对应的像素值

  3. Store this bin value in a new image (BackProjection). Also, you may consider to normalize the model histogramfirst, so the output for the Test Image can be visible for you.  归一化直方图等，使这些值在新的backprojection 图像中可使

  4. Applying the steps above, we get the following BackProjection image for our Test Image:

     

  5. In terms of statistics, the values stored in BackProjection represent the probability that a pixel in Test Image belongs to a skin area, based on the model histogram that we use. For instance in our Test image, the brighter areas are more probable to be skin area (as they actually are), whereas the darker areas have less probability (notice that these “dark” areas belong to surfaces that have some shadow on it, which in turns affects the detection). 在反向投影图像中每一个像素点代表了其属于皮肤区域的概率，越亮的地方（像素值越高）的地方可能性越大

void mixChannels(const Mat* src, size_t nsrcs, Mat* dst, size_t ndsts, const int* fromTo, size_t npairs)

Parameters:

• src – input array or vector of matricesl; all of the matrices must have the same size and the same depth.
• nsrcs – number of matrices in src.
• dst – output array or vector of matrices; all the matrices must be allocated; their size and depth must be the same as in src[0].
• ndsts – number of matrices in dst.
• fromTo – array of index pairs specifying which channels are copied and where; fromTo[k*2] is a 0-based index of the input channel in src, fromTo[k*2+1] is an index of the output channel in dst; the continuous channel numbering is used: the first input image channels are indexed from 0 tosrc[0].channels()-1, the second input image channels are indexed from src[0].channels() tosrc[0].channels() + src[1].channels()-1, and so on, the same scheme is used for the output image channels; as a special case, when fromTo[k*2] is negative, the corresponding output channel is filled with zero .
• npairs – number of index pairs in fromTo.

 

Mat rgba( 100, 100, CV_8UC4, Scalar(1,2,3,4) );Mat bgr( rgba.rows, rgba.cols, CV_8UC3 );Mat alpha( rgba.rows, rgba.cols, CV_8UC1 );// forming an array of matrices is a quite efficient operation,// because the matrix data is not copied, only the headersMat out[] = { bgr, alpha };// rgba[0] -> bgr[2], rgba[1] -> bgr[1],// rgba[2] -> bgr[0], rgba[3] -> alpha[0]int from_to[] = { 0,2, 1,1, 2,0, 3,3 };mixChannels( &rgba, 1, out, 2, from_to, 4 );

通道拆分，交换一步到位，好nb的方法

Code

#include "stdafx.h"#include "opencv2/imgproc/imgproc.hpp"#include "opencv2/highgui/highgui.hpp"#include <iostream>using namespace cv;using namespace std;/// Global VariablesMat src; Mat hsv; Mat hue;int bins = 25;/// Function Headersvoid Hist_and_Backproj(int, void* );/** @function main */int main( int argc, char** argv ){  /// Read the image  src = http://www.mamicode.com/imread("zanxin.jpg", 1 );  /// Transform it to HSV  cvtColor( src, hsv, CV_BGR2HSV );  /// Use only the Hue value  hue.create( hsv.size(), hsv.depth() );  int ch[] = { 0, 0 };	// from 0 to 0  mixChannels( &hsv, 1, &hue, 1, ch, 1 );  /// Create Trackbar to enter the number of bins  char* window_image = "Source image";  namedWindow( window_image, CV_WINDOW_AUTOSIZE );  createTrackbar("* Hue  bins: ", window_image, &bins, 180, Hist_and_Backproj );	// bins —— 0到180  Hist_and_Backproj(0, 0);  /// Show the image  imshow( window_image, src );	// 窗口初始图像  /// Wait until user exits the program  waitKey(0);  return 0;}/** * @function Hist_and_Backproj * @brief Callback to Trackbar */void Hist_and_Backproj(int, void* ){  MatND hist;  int histSize = MAX( bins, 2 );  float hue_range[] = { 0, 180 };  const float* ranges = { hue_range };  /// Get the Histogram and normalize it  calcHist( &hue, 1, 0, Mat(), hist, 1, &histSize, &ranges, true, false );	  normalize( hist, hist, 0, 255, NORM_MINMAX, -1, Mat() );  /// Get Backprojection  MatND backproj;  calcBackProject( &hue, 1, 0, hist, backproj, &ranges, 1, true );  /// Draw the backproj  imshow( "BackProj", backproj );  /// Draw the histogram  int w = 400; int h = 400;  int bin_w = cvRound( (double) w / histSize );  Mat histImg = Mat::zeros( w, h, CV_8UC3 );  for( int i = 0; i < bins; i ++ )     { rectangle( histImg, Point( i*bin_w, h ), Point( (i+1)*bin_w, h - cvRound( hist.at<float>(i)*h/255.0 ) ), Scalar( 0, 0, 255 ), -1 ); }  imshow( "Histogram", histImg );}

Code

#include "stdafx.h"#include "opencv2/imgproc/imgproc.hpp"#include "opencv2/highgui/highgui.hpp"#include <iostream>using namespace cv;using namespace std;/// Global VariablesMat src; Mat hsv;Mat mask;	// 手动选择种子点和阈值，洪范法int lo = 20; int up = 20;const char* window_image = "Source image";/// Function Headersvoid Hist_and_Backproj( );void pickPoint (int event, int x, int y, int, void* );/** * @function main */int main( int, char** argv ){  /// Read the image  src = http://www.mamicode.com/imread("zanxin.jpg", 1 );  /// Transform it to HSV  cvtColor( src, hsv, COLOR_BGR2HSV );  /// Show the image  namedWindow( window_image, WINDOW_AUTOSIZE );  imshow( window_image, src );  /// Set Trackbars for floodfill thresholds  createTrackbar( "Low thresh", window_image, &lo, 255, 0 );  createTrackbar( "High thresh", window_image, &up, 255, 0 );  /// Set a Mouse Callback  setMouseCallback( window_image, pickPoint, 0 );	// 选择种子点  waitKey(0);  return 0;}/** * @function pickPoint */void pickPoint (int event, int x, int y, int, void* ){  if( event != EVENT_LBUTTONDOWN )    { return; }  // Fill and get the mask  Point seed = Point( x, y );  int newMaskVal = 255;  Scalar newVal = Scalar( 120, 120, 120 );  int connectivity = 8;  int flags = connectivity + (newMaskVal << 8 ) + FLOODFILL_FIXED_RANGE + FLOODFILL_MASK_ONLY;  Mat mask2 = Mat::zeros( src.rows + 2, src.cols + 2, CV_8UC1 );  floodFill( src, mask2, seed, newVal, 0, Scalar( lo, lo, lo ), Scalar( up, up, up), flags );  mask = mask2( Range( 1, mask2.rows - 1 ), Range( 1, mask2.cols - 1 ) );  imshow( "Mask", mask );  Hist_and_Backproj( );}/** * @function Hist_and_Backproj */void Hist_and_Backproj( ){  MatND hist;  int h_bins = 30; int s_bins = 32;  int histSize[] = { h_bins, s_bins };  float h_range[] = { 0, 179 };  float s_range[] = { 0, 255 };  const float* ranges[] = { h_range, s_range };  int channels[] = { 0, 1 };  /// Get the Histogram and normalize it  calcHist( &hsv, 1, channels, mask, hist, 2, histSize, ranges, true, false );  normalize( hist, hist, 0, 255, NORM_MINMAX, -1, Mat() );  /// Get Backprojection  MatND backproj;  calcBackProject( &hsv, 1, channels, hist, backproj, ranges, 1, true );  /// Draw the backproj  imshow( "BackProj", backproj );}

OpenCV Tutorials —— Back Projection