注意:该程序功能是检测二维码,不是识别,只是在图中定出二维码的位置即可
原图是这样:如果出现这张图片时,程序需要找到二维码
其余图片是这样:
程序步骤: 1.图片缩小 2.灰度化,直方图均衡化,对比度增强,滤波 3.otsu阈值分割 4.五次膨胀 5.轮廓查找,如果轮廓满足一下条件,认为可能为二维码区域,像素面积大于60,长短轴之比小于1.3 6.对疑似区域做判断,因为本实验二维码贴在大概中心位置,找到疑似轮廓的质心,判断质心是不是在图片长宽的三分之一到三分之二之间,如果是认为是二维码区域 7.对二维码区域画成蓝色
代码语言:javascript复制#include <stdio.h>
#include <opencv/highgui.h>
#include <time.h>
#include <opencv2/opencv.hpp>
#include <opencv/cv.h>
#include <iostream>
using namespace std;
using namespace cv;
int main(int argc,char *argv[])
{
Mat srcImage;
Mat resizeImage;
Mat grayImage;
Mat equalizeHistImage;
Mat contrastandbrightImage;
Mat sobelImage;
Mat thresholdImage;
Mat dilateImage;
Mat areaImage;
Mat ellipseImage;
srcImage = imread("3.jpg");
Size dsize = Size(srcImage.cols*0.3,srcImage.rows*0.3);
resize(srcImage, resizeImage,dsize);
cvtColor(resizeImage,grayImage,CV_BGR2GRAY);
equalizeHist(grayImage,equalizeHistImage);
亮度、对比度增强
contrastandbrightImage= Mat::zeros( equalizeHistImage.size(), equalizeHistImage.type() );
for(int y = 0; y < equalizeHistImage.rows; y )
{
for(int x = 0; x <equalizeHistImage.cols; x )
{
contrastandbrightImage.at<uchar>(y,x)= saturate_cast<uchar>(6*(equalizeHistImage.at<uchar>(y,x) ));
}
}
blur(contrastandbrightImage,contrastandbrightImage,Size(3,3));
//Sobel(contrastandbrightImage, sobelImage, CV_8U, 0,1 ,1, 1, 0, BORDER_DEFAULT);
threshold(contrastandbrightImage,thresholdImage, 0, 255, CV_THRESH_OTSU CV_THRESH_BINARY_INV);
Mat element = getStructuringElement(MORPH_RECT, Size(3, 3));
//进行膨胀操作
dilate(thresholdImage, dilateImage, element);
dilate(dilateImage, dilateImage, element);
dilate(dilateImage, dilateImage, element);
dilate(dilateImage, dilateImage, element);
dilate(dilateImage, dilateImage, element);
//dilateImage.copyTo(areaImage);
//vector< vector< Point> > contours;
//findContours(areaImage,contours,CV_RETR_TREE, CV_CHAIN_APPROX_NONE);
//vector<vector<Point> >::iterator itc= contours.begin();
//while (itc!=contours.end())
//{
// if( itc->size()<100)
// {
// itc= contours.erase(itc);
// }
// else
// {
// itc;
// }
//}
//drawContours(areaImage, contours, -1, Scalar(255), CV_FILLED);
//找轮廓 两步判断 一个是大小 一个是长短轴比比 认为面积比60大,长短轴比比1.3小的区域是二维码区域 再做下一步判断
dilateImage.copyTo(ellipseImage);
vector<vector<Point> > twocontours;
vector<Vec4i>twohierarchy;
findContours(ellipseImage,twocontours,twohierarchy,CV_RETR_TREE,CV_CHAIN_APPROX_SIMPLE);
vector<Moments> mu(twocontours.size() );
vector<Point2f> mc( twocontours.size() );
for(int k = 0; k < (int)twocontours.size(); k ) //查找轮廓
{
if (int(twocontours.at(k).size()) <=60)//轮廓面积小于60的不处理
{
drawContours(ellipseImage, twocontours, k, Scalar(0), CV_FILLED);
}
else
{
RotatedRect rRect = fitEllipse(twocontours.at(k));
double majorAxis = rRect.size.height > rRect.size.width ? rRect.size.height : rRect.size.width;
double minorAxis = rRect.size.height > rRect.size.width ? rRect.size.width : rRect.size.height;
float rate = majorAxis/minorAxis;
if (rate<1.3) //长短轴之比小于1.4的轮廓
{
//可能为二维码的区域,判断是否为二维码区域
printf("%fn",rate);
//求区域的质心
mu[k] = moments( twocontours[k], false );
mc[k] = Point2d( mu[k].m10/mu[k].m00 , mu[k].m01/mu[k].m00 );
//打印质心
printf("x=%f,y=%fn",mc[k].x,mc[k].y);
//打印图像的长宽
printf("图像的长%d,图像的宽%dn",resizeImage.cols,resizeImage.rows);
//画出质心
Point center = Point(mc[k].x,mc[k].y);
int r = 10;
circle(resizeImage,center,r,Scalar(255,0,0));
//判断质心是不是在图像中间 三分之一 < 质心 < 三分之二
if ((int)mc[k].x<(int)2*(resizeImage.cols/3)&&(int)mc[k].x>(int)(resizeImage.cols/3))
{
if ((int)mc[k].y<(int)2*(resizeImage.rows/3)&&(int)mc[k].y>(int)(resizeImage.rows/3))
{
drawContours(resizeImage, twocontours, k, Scalar(255,0,0), CV_FILLED);//把轮廓涂成蓝色
printf("蓝色是二维码区域n");
}
}
}
else //长短轴之比大于1.4的轮廓 不是二维码区域
{
drawContours(resizeImage, twocontours, k, Scalar(0,0,255), CV_FILLED);//把轮廓涂成红色
}
}
}
imshow("缩小图",resizeImage);
imshow("灰度图",grayImage);
imshow("对比度和亮度增强",contrastandbrightImage);
//imshow("soble检测",sobelImage);
imshow("二值化结果",thresholdImage);
imshow("膨胀",dilateImage);
//imshow("删除小面积",areaImage);
imshow("长短轴",ellipseImage);
waitKey(0);
}
部分结果图:
无二维码图片的结果图: