opencv实践项目-人脸检测

1. opencv CascadeClassifier人脸检测步骤

• 从文件加载级联分类器
• 读取图片并灰度化
• resize灰度图
• 直方图均衡化，得到对比度更强的输出图像
• detectMultiScale检测

2. CascadeClassifier分类器简介

分类器是判别某个事物是否属于某种分类的器件，其结果要么是，要么不是，级联分类器，可以理解为将 N 个单类的分类器串联起来，如果一个事物能属于这一系列串联起来的的所有分类器，则最终结果就成立，若有一项不符，则判定为不成立。比如人脸，它有很多属性，我们将每个属性做一成个分类器，如果一个模型符合了我们定义的人脸的所有属性，则我们人为这个模型就是一个人脸，比如人脸需要有两条眉毛，两只眼睛，一个鼻子，一张嘴，一个大概 U 形状的下巴或者是轮廓等等。

常用方法：

2.1 从文件中加载级联分类器

CV_WRAP bool load( const String& filename );

也可用于判断级联分类器是否加载成功

2.2 目标检测方法

CV_WRAP void detectMultiScale( InputArray image, 
CV_OUT std::vector<Rect>& objects, 
double scaleFactor = 1.1, 
int minNeighbors = 3, int flags = 0, 
Size minSize = Size(), 
Size maxSize = Size() ); 

功能：检测输入图像中不同大小的对象。检测到的对象以矩形列表的形式保存
参数：
image: 包含检测对象的图像的 CV_8U 类型矩阵
objects: 矩形的向量，其中每个矩形包含被检测的对象，矩形可以部分位于原始图像之外
scaleFactor: 指定在每个图像缩放时的缩放比例
minNeighbors：指定每个候选矩形需要保留多少个相邻矩形
flags：含义与函数 cvHaarDetectObjects 中的旧级联相同，一般是 0，它不用于新的级联
minSize：对象最小大小，小于该值的对象被忽略
maxSize：最大可能的对象大小，大于这个值的对象被忽略
返回值：无

3. 代码实现

#include "opencv2/objdetect.hpp"
#include "opencv2/highgui.hpp"
#include "opencv2/imgproc.hpp"
#include "opencv2/core/ocl.hpp"
#include 

using namespace std;
using namespace cv;

static void help()
{
    cout << "\nThis program demonstrates the cascade recognizer. Now you can use Haar or LBP features.\n"
            "This classifier can recognize many kinds of rigid objects, once the appropriate classifier is trained.\n"
            "It's most known use is for faces.\n"
            "Usage:\n"
            "./ufacedetect [--cascade= this is the primary trained classifier such as frontal face]\n"
               "   [--nested-cascade[=nested_cascade_path this an optional secondary classifier such as eyes]]\n"
               "   [--scale=]\n"
               "   [--try-flip]\n"
               "   [filename|camera_index]\n\n"
            "see facedetect.cmd for one call:\n"
            "./ufacedetect --cascade=\"../../data/haarcascades/haarcascade_frontalface_alt.xml\" --nested-cascade=\"../../data/haarcascades/haarcascade_eye_tree_eyeglasses.xml\" --scale=1.3\n\n"
            "During execution:\n\tHit any key to quit.\n"
            "\tUsing OpenCV version " << CV_VERSION << "\n" << endl;
}

void detectAndDraw( UMat& img, Mat& canvas, CascadeClassifier& cascade,
                    CascadeClassifier& nestedCascade,
                    double scale, bool tryflip );

int main( int argc, const char** argv )
{
    VideoCapture capture;
    UMat frame, image;
    Mat canvas;

    string inputName;
    bool tryflip;

    CascadeClassifier cascade, nestedCascade;
    double scale;

    cv::CommandLineParser parser(argc, argv,
        "{cascade|data/haarcascades/haarcascade_frontalface_alt.xml|}"
        "{nested-cascade|data/haarcascades/haarcascade_eye_tree_eyeglasses.xml|}"
        "{help h ||}{scale|1|}{try-flip||}{@filename||}"
    );
    if (parser.has("help"))
    {
        help();
        return 0;
    }
    string cascadeName = samples::findFile(parser.get<string>("cascade"));
    string nestedCascadeName = samples::findFileOrKeep(parser.get<string>("nested-cascade"));
    scale = parser.get<double>("scale");
    tryflip = parser.has("try-flip");
    inputName = parser.get<string>("@filename");
    if ( !parser.check())
    {
        parser.printErrors();
        help();
        return -1;
    }

    if ( !nestedCascade.load( nestedCascadeName ) )
        cerr << "WARNING: Could not load classifier cascade for nested objects: " << nestedCascadeName << endl;
    if( !cascade.load( cascadeName ) )
    {
        cerr << "ERROR: Could not load classifier cascade: " << cascadeName << endl;
        help();
        return -1;
    }

    cout << "old cascade: " << (cascade.isOldFormatCascade() ? "TRUE" : "FALSE") << endl;

    if( inputName.empty() || (isdigit(inputName[0]) && inputName.size() == 1) )
    {
        int camera = inputName.empty() ? 0 : inputName[0] - '0';
        if(!capture.open(camera))
            cout << "Capture from camera #" <<  camera << " didn't work" << endl;
    }
    else
    {
        inputName = samples::findFileOrKeep(inputName);
        imread(inputName, IMREAD_COLOR).copyTo(image);
        if( image.empty() )
        {
            if(!capture.open( inputName ))
                cout << "Could not read " << inputName << endl;
        }
    }

    if( capture.isOpened() )
    {
        cout << "Video capturing has been started ..." << endl;
        for(;;)
        {
            capture >> frame;
            if( frame.empty() )
                break;

            detectAndDraw( frame, canvas, cascade, nestedCascade, scale, tryflip );

            char c = (char)waitKey(10);
            if( c == 27 || c == 'q' || c == 'Q' )
                break;
        }
    }
    else
    {
        cout << "Detecting face(s) in " << inputName << endl;
        if( !image.empty() )
        {
            detectAndDraw( image, canvas, cascade, nestedCascade, scale, tryflip );
            waitKey(0);
        }
        else if( !inputName.empty() )
        {
            /* assume it is a text file containing the
            list of the image filenames to be processed - one per line */
            FILE* f = fopen( inputName.c_str(), "rt" );
            if( f )
            {
                char buf[1000+1];
                while( fgets( buf, 1000, f ) )
                {
                    int len = (int)strlen(buf);
                    while( len > 0 && isspace(buf[len-1]) )
                        len--;
                    buf[len] = '\0';
                    cout << "file " << buf << endl;
                    imread(samples::findFile(buf), IMREAD_COLOR).copyTo(image);
                    if( !image.empty() )
                    {
                        detectAndDraw( image, canvas, cascade, nestedCascade, scale, tryflip );
                        char c = (char)waitKey(0);
                        if( c == 27 || c == 'q' || c == 'Q' )
                            break;
                    }
                    else
                    {
                        cerr << "Aw snap, couldn't read image " << buf << endl;
                    }
                }
                fclose(f);
            }
        }
    }

    return 0;
}

void detectAndDraw( UMat& img, Mat& canvas, CascadeClassifier& cascade,
                    CascadeClassifier& nestedCascade,
                    double scale, bool tryflip )
{
    double t = 0;
    vector<Rect> faces, faces2;
    const static Scalar colors[] =
    {
        Scalar(255,0,0),
        Scalar(255,128,0),
        Scalar(255,255,0),
        Scalar(0,255,0),
        Scalar(0,128,255),
        Scalar(0,255,255),
        Scalar(0,0,255),
        Scalar(255,0,255)
    };
    static UMat gray, smallImg;

    t = (double)getTickCount();

    cvtColor( img, gray, COLOR_BGR2GRAY );
    double fx = 1 / scale;
    resize( gray, smallImg, Size(), fx, fx, INTER_LINEAR_EXACT );
    equalizeHist( smallImg, smallImg );

    cascade.detectMultiScale( smallImg, faces,
        1.1, 3, 0
        //|CASCADE_FIND_BIGGEST_OBJECT
        //|CASCADE_DO_ROUGH_SEARCH
        |CASCADE_SCALE_IMAGE,
        Size(30, 30) );
    if( tryflip )
    {
        flip(smallImg, smallImg, 1);
        cascade.detectMultiScale( smallImg, faces2,
                                 1.1, 2, 0
                                 //|CASCADE_FIND_BIGGEST_OBJECT
                                 //|CASCADE_DO_ROUGH_SEARCH
                                 |CASCADE_SCALE_IMAGE,
                                 Size(30, 30) );
        for( vector<Rect>::const_iterator r = faces2.begin(); r != faces2.end(); ++r )
        {
            faces.push_back(Rect(smallImg.cols - r->x - r->width, r->y, r->width, r->height));
        }
    }
    t = (double)getTickCount() - t;
    img.copyTo(canvas);

    double fps = getTickFrequency()/t;
    static double avgfps = 0;
    static int nframes = 0;
    nframes++;
    double alpha = nframes > 50 ? 0.01 : 1./nframes;
    avgfps = avgfps*(1-alpha) + fps*alpha;

    putText(canvas, format("OpenCL: %s, fps: %.1f", ocl::useOpenCL() ? "ON" : "OFF", avgfps), Point(50, 30),
            FONT_HERSHEY_SIMPLEX, 0.8, Scalar(0,255,0), 2);

    for ( size_t i = 0; i < faces.size(); i++ )
    {
        Rect r = faces[i];
        vector<Rect> nestedObjects;
        Point center;
        Scalar color = colors[i%8];
        int radius;

        double aspect_ratio = (double)r.width/r.height;
        if( 0.75 < aspect_ratio && aspect_ratio < 1.3 )
        {
            center.x = cvRound((r.x + r.width*0.5)*scale);
            center.y = cvRound((r.y + r.height*0.5)*scale);
            radius = cvRound((r.width + r.height)*0.25*scale);
            circle( canvas, center, radius, color, 3, 8, 0 );
        }
        else
            rectangle( canvas, Point(cvRound(r.x*scale), cvRound(r.y*scale)),
                       Point(cvRound((r.x + r.width-1)*scale), cvRound((r.y + r.height-1)*scale)),
                       color, 3, 8, 0);
        if( nestedCascade.empty() )
            continue;
        UMat smallImgROI = smallImg(r);
        nestedCascade.detectMultiScale( smallImgROI, nestedObjects,
            1.1, 2, 0
            //|CASCADE_FIND_BIGGEST_OBJECT
            //|CASCADE_DO_ROUGH_SEARCH
            //|CASCADE_DO_CANNY_PRUNING
            |CASCADE_SCALE_IMAGE,
            Size(30, 30) );

        for ( size_t j = 0; j < nestedObjects.size(); j++ )
        {
            Rect nr = nestedObjects[j];
            center.x = cvRound((r.x + nr.x + nr.width*0.5)*scale);
            center.y = cvRound((r.y + nr.y + nr.height*0.5)*scale);
            radius = cvRound((nr.width + nr.height)*0.25*scale);
            circle( canvas, center, radius, color, 3, 8, 0 );
        }
    }
    imshow( "result", canvas );
}

该代码为官方示例代码，基本调用方法如下：

example_tapi_ufacedetect --cascade=D:\wyw\opencv\opencv-4.5.0\data\haarcascades\haarcascade_frontalface_alt.xml --nested-cascade=D:\wyw\opencv\opencv-4.5.0\data\haarcascades\haarcascade_eye_tree_eyeglasses.xml --scale=1 --try-flip=true E:\code\other\SeetaFaceDetector\FaceDetector\x64\Release\6.jpg

本文部分引用：https://zhuanlan.zhihu.com/p/159013461