2016-05-05 82 views
1

我正在寻找汉明距离帮助下的视频差异。这里是我的代码:OpenCV.norm中的声明错误

#include <iostream> 
#include "opencv2/opencv.hpp" 
using namespace cv; 
int main(int argc, char** argv) 
{ 
    char const* filename = ("video.mp4"); 
    VideoCapture video(filename); 
    Mat frame, temp; 
    for(unsigned int i = 0; i<100; i++) 
    { 
     video >> frame; 
     if (i > 0) 
     { 
      double dist = norm(frame, temp, NORM_HAMMING); 
      cout<<"Dist= "<< dist <<endl; 
     } 
     temp = frame; 
    } 
    return 0; 
} 

的问题是,在

double dist = norm(frame, temp, NORM_HAMMING); 

程序降至

**OpenCV Error**: Assertion failed (normType == NORM_INF || normType == NORM_L1 || normType == NORM_L2 || normType == NORM_L2SQR || ((normType == NORM_HAMMING || normType == NORM_HAMMING2) && src1.type() == CV_8U))in norm, in file /home/andrio/.local/share/Trash/files/build/OpenCV/modules/core/src/stat.cpp, line 3123 terminate called after throwing an instance of 'cv::Exception' what(): /home/andre/.local/share/Trash/files/build/OpenCV/modules/core/src/stat.cpp:3123: error: (-215) normType == NORM_INF || normType == NORM_L1 || normType == NORM_L2 || normType == NORM_L2SQR || ((normType == NORM_HAMMING || normType == NORM_HAMMING2) && src1.type() == CV_8U) in function norm 

UPD:此代码的工作:

double dist = norm(frame, temp, NORM_L2); 
+0

那么,当您调用该函数时,“帧”矩阵的数据类型是什么?实际上是否有任何图像加载?在将其作为参数传递给其他函数之前,您不会尝试进行验证。 –

+0

frame.type()和temp.type()输出为16.这里[OpenCV](http://docs.opencv.org/2.4/modules/core/doc/operations_on_arrays.html)没有说明应该是什么一个数组类型。 –

回答

1

您需要将图像变为灰色

cvtColor(frame, frame, CV_BGR2GRAY);