毕设之Opencv批量圆拟合

毕设之Opencv批量圆拟合

程序思路：编写read_csv()函数读取图片目录下txt文档，获取各BMP文件绝对路径以及对应圆的圆心坐标、半径参数。读取各BMP图像，转为灰度图，二值化后检测轮廓，然后检测最小圆；并将检测结果和实际结果存入txt文档。

不需调参；对于完美圆的检测优于霍夫圆变换。

功能小结：

1、read_csv()函数

void read_csv(string&csvPath, vector<String>&CirclePath, vector<int>&Circle_x, vector<int>&Circle_y, vector<int>&Circle_r)

csvPath：txt路径

CirclePath：从txt中读取的bmp路径

Circle_x：bmp图像中圆的圆心x坐标

Circle_y：bmp图像中圆的圆心y坐标

Circle_r：bmp图像中圆的半径r

2、创建txt文本，并写入数据

#include <iostream>

#include <sstream>

#include <fstream>

ofstream file("filepath",ios::out);

if (file.is_open())

{

    file <<;

}

file.close();

3、转化为灰度图

cvtColor(srcImage,midImage, COLOR_BGR2GRAY);//转化边缘检测后的图为灰度图

@paramsrc input image: 8-bit unsigned, 16-bit unsigned ( CV_16UC... ), orsingle-precision

floating-point.

@paramdst output image of the same size and depth as src.

@paramcode color space conversion code (see cv::ColorConversionCodes).

@paramdstCn number of channels in the destination image; if the parameter is 0, thenumber of the

channelsis derived automatically from src and code.

4、图像二值化Threshold()

函数 Threshold对单通道数组应用固定阈值操作。该函数的典型应用是对灰度图像进行阈值操作得到二值图像。(cvCmpS也可以达到此目的)或者是去掉噪声，例如过滤很小或很大像素值的图像点。本函数支持的对图像取阈值的方法由 threshold_type确定。

CV_EXPORTS_W double threshold( InputArray src,OutputArray dst,

                               double thresh, double maxval, int type );

/**@brief Applies a fixed-level threshold to each array element.

 

Thefunction applies fixed-level thresholding to a single-channel array. Thefunction is typically

usedto get a bi-level (binary) image out of a grayscale image ( cv::compare couldbe also used for

thispurpose) or for removing a noise, that is, filtering out pixels with too smallor too large

values.There are several types of thresholding supported by the function. They aredetermined by

typeparameter.

 

Also,the special values cv::THRESH_OTSU or cv::THRESH_TRIANGLE may be combined withone of the

abovevalues. In these cases, the function determines the optimal threshold valueusing the Otsu's

orTriangle algorithm and uses it instead of the specified thresh . The functionreturns the

computedthreshold value. Currently, the Otsu's and Triangle methods are implementedonly for 8-bit

images.

 

@paramsrc input array (single-channel, 8-bit or 32-bit floating point).

@paramdst output array of the same size and type as src.

@paramthresh threshold value.

@parammaxval maximum value to use with the THRESH_BINARY and THRESH_BINARY_INVthresholding

types.

@paramtype thresholding type (see the cv::ThresholdTypes).

 

@sa  adaptiveThreshold, findContours, compare,min, max

 */

5、检索轮廓FindContours()

函数FindContours从二值图像中检索轮廓，并返回检测到的轮廓的个数。first_contour的值由函数填充返回，它的值将为第一个外轮廓的指针，当没有轮廓被检测到时为NULL。其它轮廓可以使用h_next和v_next连接，从first_contour到达。

CV_EXPORTS_W void findContours( InputOutputArray image,OutputArrayOfArrays contours,

                              OutputArray hierarchy, int mode,

                              int method,Point offset = Point());

/**@brief Finds contours in a binary image.

 

Thefunction retrieves contours from the binary image using the algorithm @citeSuzuki85 . The contours

area useful tool for shape analysis and object detection and recognition. Seesquares.c in the

OpenCVsample directory.

 

@noteSource image is modified by this function. Also, the function does not takeinto account

1-pixelborder of the image (it's filled with 0's and used for neighbor analysis in thealgorithm),

thereforethe contours touching the image border will be clipped.

 

@paramimage Source, an 8-bit single-channel image. Non-zero pixels are treated as1's. Zero

pixelsremain 0's, so the image is treated as binary . You can use compare , inRange ,threshold ,

adaptiveThreshold, Canny , and others to create a binary image out of a grayscale or color one.

Thefunction modifies the image while extracting the contours. If mode equals toRETR_CCOMP

orRETR_FLOODFILL, the input can also be a 32-bit integer image of labels(CV_32SC1).

@paramcontours Detected contours. Each contour is stored as a vector of points.

@paramhierarchy Optional output vector, containing information about the imagetopology. It has

asmany elements as the number of contours. For each i-th contour contours[i] ,the elements

hierarchy[i][0], hiearchy[i][1] , hiearchy[i][2] , and hiearchy[i][3] are set to 0-basedindices

incontours of the next and previous contours at the same hierarchical level, thefirst child

contourand the parent contour, respectively. If for the contour i there are no next,previous,

parent,or nested contours, the corresponding elements of hierarchy[i] will benegative.

@parammode Contour retrieval mode, see cv::RetrievalModes

@parammethod Contour approximation method, see cv::ContourApproximationModes

@paramoffset Optional offset by which every contour point is shifted. This is usefulif the

contoursare extracted from the image ROI and then they should be analyzed in the wholeimage

context.

 */

6、最小圆逼近minEnclosingCircle()

CV_EXPORTS_W void minEnclosingCircle(InputArray points,

                                      CV_OUT Point2f& center, CV_OUT float& radius );

 

/**@brief Finds a circle of the minimum area enclosing a 2D point set.

 

Thefunction finds the minimal enclosing circle of a 2D point set using aniterative algorithm. See

theOpenCV sample minarea.cpp .

 

@parampoints Input vector of 2D points, stored in std::vector\<\> or Mat

@paramcenter Output center of the circle.

@paramradius Output radius of the circle.

 */

 

源码：

//--------------------------------------【程序说明】-------------------------------------------

//      程序描述：  利用霍夫变换检测圆

//      开发测试所用操作系统： Windows 7 64bit

//      开发测试所用IDE版本：VisualStudio 2015

//      开发测试所用OpenCV版本：  3.1

//      2015年5月 Created by @姬波林

//      2015年5月 Revised by @姬波林

//------------------------------------------------------------------------------------------------

 

 

//---------------------------------【头文件、命名空间包含部分】----------------------------

//          描述：包含程序所使用的头文件和命名空间

//------------------------------------------------------------------------------------------------

#include <opencv2/opencv.hpp>

#include <opencv2/imgproc/imgproc.hpp>

 

using namespace cv;

using namespace std;

 

#include <iostream>

#include <sstream>

#include <fstream>

 

//-----------------------------------【宏定义部分】--------------------------------------------

//      描述：定义一些辅助宏

//------------------------------------------------------------------------------------------------

 

 

//--------------------------------【全局函数声明部分】-------------------------------------

//      描述：全局函数声明

//-----------------------------------------------------------------------------------------------

void read_csv(string&csvPath, vector<String>&CirclePath, vector<int>&Circle_x, vector<int>&Circle_y, vector<int>&Circle_r)

{

   

    string line, path, classLabel1, classLabel2, classLabel3;

    ifstream file(csvPath.c_str(),ifstream::in);

    while (getline(file, line))

    {

        stringstream lines(line);

        getline(lines, path, '(');

 

        getline(lines, classLabel1,',');

        if (!classLabel1.empty())

        {

            Circle_x.push_back(atoi(classLabel1.c_str()));

        }

        getline(lines, classLabel2, ')');

        if (!classLabel2.empty())

        {

            Circle_y.push_back(atoi(classLabel2.c_str()));

        }

        getline(lines, classLabel3, '.');

        if (!classLabel3.empty())

        {

            Circle_r.push_back(atoi(classLabel3.c_str()));

        }

 

        if (!path.empty())

        {

            CirclePath.push_back(path+"("+ classLabel1+","+ classLabel2+")" + classLabel3+".bmp");

        }

    }

}

 

 

//-----------------------------------【ShowpText( )函数】----------------------------------

//          描述：输出一些帮助信息

//----------------------------------------------------------------------------------------------

void ShowText()

{

    //输出程序说明

    printf("绘制100个半径随机、圆心随机的圆，并保存为500X500BMP文件\n");

    printf("存储位置：D:\\圆\n");

    printf("命名规则：序号+圆心坐标+半径\n");

    printf("当前使用的OpenCV版本为："CV_VERSION );

    printf("\n\n ----------------------------------------------------------------------------\n");

}

 

 

 

//---------------------------------------【main( )函数】--------------------------------------

//      描述：控制台应用程序的入口函数，我们的程序从这里开始执行

//-----------------------------------------------------------------------------------------------

 

const int kvalue = 15;//双边滤波邻域大小

 

int main()

{

    //批量读入圆路径

    string CircleCsvPath ="D:\\Circle\\at.txt";

    vector<String> vecCirclePath;

    vector<int> Circle_x, Circle_y, Circle_r;

    read_csv(CircleCsvPath, vecCirclePath,Circle_x, Circle_y, Circle_r);

 

    ofstream file("D:\\Circle\\圆拟合检测结果.txt",ios::out);

    if (file.is_open())

    {

        file << "序号" <<"\t" <<"圆心x"<< "\t"<< "圆心y"<< "\t"<< "半径r"<< endl;

    }

    cout << "圆心x" <<"\t" <<"圆心y"<< "\t"<< "半径r"<< endl;

 

    Mat threshold_output, srcImage;

    Mat midImage, dstImage;//临时变量和目标图的定义

    vector<vector<Point> > contours;

    vector<Vec4i> hierarchy;

 

    for (size_t i = 0; i < 100; i++)

    {

        srcImage = imread(vecCirclePath[i].c_str(), 1);//读取原彩色图                                             

        //【3】转为灰度图并进行图像平滑

        cvtColor(srcImage, midImage, COLOR_BGR2GRAY);//转化边缘检测后的图为灰度图

        //GaussianBlur(midImage, midImage, Size(9, 9), 2,2);

        //blur(midImage, midImage, Size(3, 3));

        //imshow("图像平滑", midImage);

        /// Detect edges using Threshold

        threshold(midImage, threshold_output,100, 255, THRESH_BINARY);

        /// Find contours

        findContours(threshold_output, contours,hierarchy, CV_RETR_TREE, CV_CHAIN_APPROX_SIMPLE,Point(0, 0));

        /// Approximate contours to polygons+ get bounding rects and circles

        vector<vector<Point> > contours_poly(contours.size());

        vector<Rect> boundRect(contours.size());

        vector<Point2f>center(contours.size());

        vector<float>radius(contours.size());

        for (int j = 0; j < 1; j++)

        {

            approxPolyDP(Mat(contours[j]), contours_poly[j], 3,true);

            boundRect[j]= boundingRect(Mat(contours_poly[j]));

            minEnclosingCircle((Mat)contours_poly[j], center[j], radius[j]);

 

            //绘制圆心

            circle(srcImage, center[j], 1,Scalar(0, 255, 0), -1, 8,0);

            //绘制圆轮廓

            circle(srcImage, center[j], radius[j],Scalar(155, 50, 255), 1,8, 0);

 

            if (file.is_open())

            {

                file << i << "\t" << Circle_x[i]<< "\t"<< Circle_y[i]<< "\t"

                    << Circle_r[i]<< "\t"<< "实际结果"<< endl;

                file << i << "\t" << center[j].x<< "\t"<< center[j].y<< "\t"

                    << radius[j]<< "\t"<< "检测结果"<< endl;//在控制台输出圆心坐标和半径  

            }

 

            std::cout << Circle_x[i]<< "\t"<< Circle_y[i]<< "\t"

                << Circle_r[i]<< "\t"<< "实际结果"<< endl;

            std::cout << center[j].x<< "\t"<< center[j].y<< "\t"

                << radius[j]<< "\t"<< "检测结果"<< endl;//在控制台输出圆心坐标和半径  

        }

    }

 

    /// Show in a window

    //namedWindow("Contours", CV_WINDOW_AUTOSIZE);

    //imshow("Contours", srcImage);

 

    file.close();

    waitKey();

    return 0;

}

 

效果图：