为了细致掌握程明明CVPR 2014 oral文章：BING: Binarized Normed Gradients for Objectness Estimation at 300fps的代码，的好好学习opencv库啊，从基础走起。

（1）CV_Assert函数作用：

CV_Assert（）若括号里的表达式值为false。则返回一个错误信息。

（2）Mat使用方法

1、使用准备：

 using  namespace  cv;



2、Mat的声明



Mat m=Mat(rows, cols, type);

Mat m=Mat(Size(width,height), type);

Mat A=Mat(3,4,CV_32FC1);

Mat B=Mat(4,3,CV_32FC1);



3、Mat赋值



vector<Point3f>v;//suppose it is already full

Mat m1=Mat(v,true);//boolean value true is necessary in order to copy data from v to m1

CvMat *p1;

Mat m2=Mat(p1);



4、Mat之间运算



MatC=2*A*B;



Mat C=C.inv();//Now C is its own inverse matrix



Mat D=A.t();//D is the transposed matrix of A



Mat a=Mat(4,1, CV_32FC3);//a is 4x1, 3 channels



Mat b=a.reshape(1);//b is 4x3, 1 channel





5、单通道Mat元素读写



Mat a=Mat(4,3, CV_32FC1);

floatelem_a=a.at<float>(i,j);//access element aij, with i from 0 to rows-1 and j from 0 to cols-1

Point p=Point(x,y);

floatelem_a=a.at<float>(p);//Warning: y ranges from 0 to rows-1 and x from 0 to cols-1





6、多通道Mat元素读写



template<typename _Tp> _Tp& at(int y,int x);              // cxcore.hpp (868)

template<typename _Tp>const _Tp& at(int y,int x)const;    // cxcore.hpp (870)

template<typename _Tp> _Tp& at(Point pt); // cxcore.hpp (869)

template<typename _Tp>const _Tp& at(Point pt)const;       // cxcore.hpp (871)

// defineded in cxmat.hpp (454-468)

typedefVec<float,2>Vec2f;// cxcore.hpp (254)



// we can access the element like this :

Mat m(Size(3,3), CV_32FC2 );

Vec2f& elem = m.at<Vec2f>( row , col );// or m.at<Vec2f>( Point(col,row) );

elem[0]=1212.0f;

elem[1]=326.0f;

float c1 = m.at<Vec2f>( row , col )[0];// or m.at<Vec2f>( Point(col,row) );

float c2 = m.at<Vec2f>( row , col )[1];

m.at<Vec2f>( row, col )[0]=1986.0f;

m.at<Vec2f>( row, col )[1]=326.0f;





7.选取Mat上指定区域方法





Mat src; Rect rect;

Mat dst = src(rect); 或者Mat dst(src,rect);



注意：cv::Mat A;



A.row(i) = A.row(j); // 错误

A.row(i) = A.row(j) + 0; // 正确