基于Opencv的目标检测与跟踪阴影去除算法实现

在运动目标检测中，常常会出现由于光线被遮挡，或场景其他物体的遮挡，在目标附近或场景里出现阴影，阴影的出现对后期目标的正确分割与处理带了很大的不便。如今，国内外已有不少文献来研究这个问题，并且提出了各种各样的阴影去除算法。本文就其中的一种算法提出了一个通用的实现算法，该算法考虑了背景与前景颜色相近的情况，希望能给大家一些帮助：(介绍下算法的思路：算法首先对RGB颜色空间的值进行归一化处理，即：r=R/(R+G+B),g=G/(R+G+B), I=(R+G+B)/3。然后利用背景和当前帧r,g的插值和I的比例来确定阴影区域：

                                                                

详细细节请参考文献：Detecting moving objects,ghosts and shadows in video streams

// shadeImg is a binary image,the value th2 and th3 are chosen empirically here,set th2=0.6 th3=1.5, th4=7
void ShadeDetect(IplImage *currImg, IplImage *bkImg, IplImage *shdadeImg,double th1,double th2,double th3,double th4)
{
       cvZero(shdadeImg);
       unsigned char* currData;
       unsigned char* bkData;
       unsigned char* shadeData;
       int i=0,j=0;
       int height=currImg->height;
       int width=currImg->width;
       double rb=0,gb=0,Ib=0,Rb=0,Gb=0,Bb=0;
       double rt=0,gt=0,It=0,Rt=0,Gt=0,Bt=0;
      //CvScalar cs=cvScalarAll(255);
       for (i=0;i<height;i++)
       {
             currData=(unsigned char*)currImg->imageData+i*currImg->widthStep;
             bkData=(unsigned char*)bkImg->imageData+i*bkImg->widthStep;
             shadeData=(unsigned char*)shdadeImg->imageData+i*shdadeImg->widthStep;
            for (j=0;j<width;j++)
           {
                // current normalized
                  Rt=currData[j*3];
                  Gt=currData[j*3+1];
                  Bt=currData[j*3+2];
                  rt=Rt/(Rt+Gt+Bt);
                  gt=Gt/(Rt+Gt+Bt);
                  It=(Rt+Gt+Bt)/3;
                 // Bk normalized
                 Rb=bkData[j*3];
                 Gb=bkData[j*3+1];
                 Bb=bkData[j*3+2];
                rb=Rb/(Rb+Gb+Bb);
                gb=Gb/(Rb+Gb+Bb);
                Ib=(Rb+Gb+Bb)/3;
              // judge whether is shadeimg
              if (fabs(rt-rb)<=th1 && fabs(gt-gb)<th1 && It/Ib>=th2 && It/Ib<=th3 && fabs(It-Bt)<th4)
             {
                    shadeData[j]=255;
             }  
       }
   }
}

注意：引用或转帖请注明出处：http://cvchina.net/space-uid-423.html 或http://hi.baidu.com/312611432belial/home