计算机视觉
图像处理

Kinect+OpenNI学习笔记之13(Kinect驱动类,OpenCV显示类和手部预分割类的设计)

前言

为了减小以后项目的开发效率,本次实验将OpenNI底层驱动KinectOpencv初步处理OpenNI获得的原始数据,以及手势识别中 的分割(因为本系统最后是开发手势识别的)这3个部分的功能单独做成类,以便以后移植和扩展。其实在前面已经有不少文章涉及到了这3部分的设计,比如说:Kinect+OpenNI学习笔记之3(获取kinect的数据并在Qt中显示的类的设计)Kinect+OpenNI学习笔记之11(OpenNI驱动kinect手势相关的类的设计)Kinect+OpenNI学习笔记之12(简单手势所表示的数字的识别) 。这次是综合前面几次的设计,优化了下这几个类。

开发环境:开发环境:QtCreator2.5.1+OpenNI1.5.4.0+Qt4.8.2+OpenCV2.4.3

 

实验基础

OPenNI/OPenCV知识点总结:

Kinect驱动类,OpenCV显示类手部预分割类这3个类,单独来设计其实参考了前面的博文还是很简单的,但是由于这3个类之间有相互联 系,设计不好就会出现图像显示非常卡。这时候,需要注意下面几点问题(在本程序代码中,Kinect驱动类为COpenniHand,OpenCV显示类 为CKinectOpenCV, 手部预分割类为CKinectHandSegment):

因为在kinect驱动类中有完整的kinect驱动程序(这个驱动会占用一部分时间的),而OpenCV显示类调用了Kinect驱动类中的 内容,相当于完成了一次Kinect驱动完整过程,这时候,因为在手部预分割过程中,要获得手部的中心点,如果在该类中再次执行kinect的驱动来获得 该中心点,那么整个系统中一个流程的时间其kinect需要驱动两次,这会浪费很多系统资源,导致图像显示不流畅等。因此我们应该在OpenCV显示类中 就返回Kinect驱动类中能够返回的值,比如说手部中心点的位置

在CKinectOpenCV类中由于要返回手部的中心点位置,本打算在类内部公共部分设置一个获取手部中心点位置的函数的,但是发现如果这个 函数的返回值是map类型时,运行时老出错误(理论上应该是不会出错的),所以后面该为直接返回手部中心点的变量(map类型),但是在这个变量返回前要 保证它的值是实时更新的,所以应该在返回前加入kinect驱动程序中的Updata函数,我这里将其设计成了一个开关函数,即如果允许获取手部中心点, 就将开关函数中的参数设置为ture,具体参见代码部分。

C/C++知识点总结:

定义类的对象并使用该对象后,一般会先调用该类的初始化函数,该函数的作用一般是为类成员变量进行一些初始设置,但如果类中其它函数的调用前每 次都初始化某些变量时,这些变量的初始化不宜放在类的初始化函数中,而应该单独给个私有函数,在那些需要调用它的函数前面进行被调用,达到初始化某些变量 的目的。

类的设计的目的一是为了方便,而是为了提高效率,有时候不能够光为了方便而去设计,比如说在本次类设计中要获得分割好了的图像,或者原始图像, 或者深度图像等等,确实是可以直接使用一个函数每一幅图像,不过每次获得图像就要更新一个下kinect驱动中的数据,因此这样的效率就非常低了,在实际 设计中,我把那些kinect驱动设备的程序写在了一个函数中,但是这个函数又不能被获取图像的每个函数去调用,否则还是相当于驱动了多次,因此只能由类 所定义的对象来调用了。结果是每一个主函数循环中,我们在定义了类的对象后多调用一个函数,再去获得所需的图像,这样只是多了一句代码,却节省了不少时间 消耗。

 

实验结果

本次实验完成的功能依旧是获取kinect的深度图,颜色图,手势分割图,手势轮廓图等。下面是手势分割图和轮廓处理图的结果:

 

实验代码及注释:

copennihand.h:

#ifndef COpenniHand_H
#define COpenniHand_H

#include <XnCppWrapper.h>
#include <iostream>
#include <vector>
#include <map>

using namespace xn;
using namespace std;

class COpenniHand
{
public:
    COpenniHand();
    ~COpenniHand();

    /*OpenNI的内部初始化,属性设置*/
    bool Initial();

    /*启动OpenNI读取Kinect数据*/
    bool Start();

    /*更新OpenNI读取到的数据*/
    bool UpdateData();

    /*得到色彩图像的node*/
    ImageGenerator& getImageGenerator();

    /*得到深度图像的node*/
    DepthGenerator& getDepthGenerator();

    /*得到手势姿势的node*/
    GestureGenerator& getGestureGenerator();

    /*得到手部的node*/
    HandsGenerator& getHandGenerator();
    DepthMetaData depth_metadata_;   //返回深度图像数据
    ImageMetaData image_metadata_;   //返回彩色图像数据
    std::map<XnUserID, XnPoint3D> hand_points_;  //为了存储不同手的实时点而设置的
    std::map< XnUserID, vector<XnPoint3D> > hands_track_points_; //为了绘画后面不同手部的跟踪轨迹而设定的

private:
    /*该函数返回真代表出现了错误,返回假代表正确*/
    bool CheckError(const char* error);

    /*表示某个手势动作已经完成检测的回调函数*/
    static void XN_CALLBACK_TYPE  CBGestureRecognized(xn::GestureGenerator &generator, const XnChar *strGesture,
                                                      const XnPoint3D *pIDPosition, const XnPoint3D *pEndPosition,
                                                      void *pCookie);

    /*表示检测到某个手势开始的回调函数*/
    static void XN_CALLBACK_TYPE CBGestureProgress(xn::GestureGenerator &generator, const XnChar *strGesture,
                                                   const XnPoint3D *pPosition, XnFloat fProgress, void *pCookie);

    /*手部开始建立的回调函数*/
    static void XN_CALLBACK_TYPE HandCreate(HandsGenerator& rHands, XnUserID xUID, const XnPoint3D* pPosition,
                                            XnFloat fTime, void* pCookie);

    /*手部开始更新的回调函数*/
    static void XN_CALLBACK_TYPE HandUpdate(HandsGenerator& rHands, XnUserID xUID, const XnPoint3D* pPosition, XnFloat fTime,
                                            void* pCookie);

    /*手部销毁的回调函数*/
    static void XN_CALLBACK_TYPE HandDestroy(HandsGenerator& rHands, XnUserID xUID, XnFloat fTime, void* pCookie);

    XnStatus status_;
    Context context_;
    XnMapOutputMode xmode_;
    ImageGenerator  image_generator_;
    DepthGenerator  depth_generator_;
    GestureGenerator gesture_generator_;
    HandsGenerator  hand_generator_;
};

#endif // COpenniHand_H

 

  copennihand.cpp:

#include "copennihand.h"
#include <XnCppWrapper.h>
#include <iostream>
#include <map>

using namespace xn;
using namespace std;

COpenniHand::COpenniHand()
{
}

COpenniHand::~COpenniHand()
{
}

bool COpenniHand::Initial()
{
    status_ = context_.Init();
    if(CheckError("Context initial failed!")) {
        return false;
    }

    context_.SetGlobalMirror(true);//设置镜像
    xmode_.nXRes = 640;
    xmode_.nYRes = 480;
    xmode_.nFPS = 30;

    //产生颜色node
    status_ = image_generator_.Create(context_);
    if(CheckError("Create image generator  error!")) {
        return false;
    }

    //设置颜色图片输出模式
    status_ = image_generator_.SetMapOutputMode(xmode_);
    if(CheckError("SetMapOutputMdoe error!")) {
        return false;
    }

    //产生深度node
    status_ = depth_generator_.Create(context_);
    if(CheckError("Create depth generator  error!")) {
        return false;
    }

    //设置深度图片输出模式
    status_ = depth_generator_.SetMapOutputMode(xmode_);
    if(CheckError("SetMapOutputMdoe error!")) {
        return false;
    }

    //产生手势node
    status_ = gesture_generator_.Create(context_);
    if(CheckError("Create gesture generator error!")) {
        return false;
    }

    /*添加手势识别的种类*/
    gesture_generator_.AddGesture("Wave", NULL);
    gesture_generator_.AddGesture("click", NULL);
    gesture_generator_.AddGesture("RaiseHand", NULL);
    gesture_generator_.AddGesture("MovingHand", NULL);

    //产生手部的node
    status_ = hand_generator_.Create(context_);
    if(CheckError("Create hand generaotr error!")) {
        return false;
    }

    //视角校正
    status_ = depth_generator_.GetAlternativeViewPointCap().SetViewPoint(image_generator_);
    if(CheckError("Can't set the alternative view point on depth generator!")) {
        return false;
    }

    //设置与手势有关的回调函数
    XnCallbackHandle gesture_cb;
    gesture_generator_.RegisterGestureCallbacks(CBGestureRecognized, CBGestureProgress, this, gesture_cb);

    //设置于手部有关的回调函数
    XnCallbackHandle hands_cb;
    hand_generator_.RegisterHandCallbacks(HandCreate, HandUpdate, HandDestroy, this, hands_cb);

    return true;
}

bool COpenniHand::Start()
{
    status_ = context_.StartGeneratingAll();
    if(CheckError("Start generating error!")) {
        return false;
    }
    return true;
}

bool COpenniHand::UpdateData()
{
    status_ = context_.WaitNoneUpdateAll();
    if(CheckError("Update date error!")) {
        return false;
    }
    //获取数据
    image_generator_.GetMetaData(image_metadata_);
    depth_generator_.GetMetaData(depth_metadata_);

    return true;
}

ImageGenerator &COpenniHand::getImageGenerator()
{
    return image_generator_;
}

DepthGenerator &COpenniHand::getDepthGenerator()
{
    return depth_generator_;
}

GestureGenerator &COpenniHand::getGestureGenerator()
{
    return gesture_generator_;
}

HandsGenerator &COpenniHand::getHandGenerator()
{
    return hand_generator_;
}

bool COpenniHand::CheckError(const char *error)
{
    if(status_ != XN_STATUS_OK) {
        cerr << error << ": " << xnGetStatusString( status_ ) << endl;
        return true;
    }
    return false;
}

void COpenniHand::CBGestureRecognized(GestureGenerator &generator, const XnChar *strGesture, const XnPoint3D *pIDPosition, const XnPoint3D *pEndPosition, void *pCookie)
{
    COpenniHand *openni = (COpenniHand*)pCookie;
    openni->hand_generator_.StartTracking(*pEndPosition);
}

void COpenniHand::CBGestureProgress(GestureGenerator &generator, const XnChar *strGesture, const XnPoint3D *pPosition, XnFloat fProgress, void *pCookie)
{
}

void COpenniHand::HandCreate(HandsGenerator &rHands, XnUserID xUID, const XnPoint3D *pPosition, XnFloat fTime, void *pCookie)
{
    COpenniHand *openni = (COpenniHand*)pCookie;
    XnPoint3D project_pos;
    openni->depth_generator_.ConvertRealWorldToProjective(1, pPosition, &project_pos);
    pair<XnUserID, XnPoint3D> hand_point_pair(xUID, XnPoint3D());//在进行pair类型的定义时,可以将第2个设置为空
    hand_point_pair.second = project_pos;
    openni->hand_points_.insert(hand_point_pair);//将检测到的手部存入map类型的hand_points_中。

    pair<XnUserID, vector<XnPoint3D>> hand_track_point(xUID, vector<XnPoint3D>());
    hand_track_point.second.push_back(project_pos);
    openni->hands_track_points_.insert(hand_track_point);
}

void COpenniHand::HandUpdate(HandsGenerator &rHands, XnUserID xUID, const XnPoint3D *pPosition, XnFloat fTime, void *pCookie)
{
    COpenniHand *openni = (COpenniHand*)pCookie;
    XnPoint3D project_pos;
    openni->depth_generator_.ConvertRealWorldToProjective(1, pPosition, &project_pos);
    openni->hand_points_.find(xUID)->second = project_pos;
    openni->hands_track_points_.find(xUID)->second.push_back(project_pos);
}

void COpenniHand::HandDestroy(HandsGenerator &rHands, XnUserID xUID, XnFloat fTime, void *pCookie)
{
    COpenniHand *openni = (COpenniHand*)pCookie;
    openni->hand_points_.erase(openni->hand_points_.find(xUID));
    openni->hands_track_points_.erase(openni->hands_track_points_.find(xUID ));
}

 

  ckinectopencv.h:

#ifndef CKINECTOPENCV_H
#define CKINECTOPENCV_H

#include <opencv2/core/core.hpp>
#include "copennihand.h"

using namespace cv;

class CKinectOpenCV
{
public:
    CKinectOpenCV();
    ~CKinectOpenCV();
    void GetAllInformation();   //在返回有用信息前调用该函数,因为openni的数据在不断更新,信息的处理最好放在一个函数中
    Mat GetColorImage() ;
    Mat GetDepthImage() ;
    std::map<XnUserID, XnPoint3D> GetHandPoints();

private:
    COpenniHand openni_hand_;
    std::map<XnUserID, XnPoint3D> hand_points_;  //为了存储不同手的实时点而设置的
    Mat color_image_;    //颜色图像
    Mat depth_image_;    //深度图像


};

#endif // CKINECTOPENCV_H

  ckinectopencv.cpp:

#include "ckinectopencv.h"
#include <opencv2/imgproc/imgproc.hpp>
#include <map>

using namespace cv;
using namespace std;

#define DEPTH_SCALE_FACTOR 255./4096.

CKinectOpenCV::CKinectOpenCV()
{   
    /*初始化openni对应的设备*/
     CV_Assert(openni_hand_.Initial());

    /*启动openni对应的设备*/
    CV_Assert(openni_hand_.Start());
}

CKinectOpenCV::~CKinectOpenCV()
{
}

void CKinectOpenCV::GetAllInformation()
{
    CV_Assert(openni_hand_.UpdateData());
    /*获取色彩图像*/
    Mat color_image_src(openni_hand_.image_metadata_.YRes(), openni_hand_.image_metadata_.XRes(),
                        CV_8UC3, (char *)openni_hand_.image_metadata_.Data());
    cvtColor(color_image_src, color_image_, CV_RGB2BGR);

    /*获取深度图像*/
    Mat depth_image_src(openni_hand_.depth_metadata_.YRes(), openni_hand_.depth_metadata_.XRes(),
                        CV_16UC1, (char *)openni_hand_.depth_metadata_.Data());//因为kinect获取到的深度图像实际上是无符号的16位数据
    depth_image_src.convertTo(depth_image_, CV_8U, DEPTH_SCALE_FACTOR);

    hand_points_ = openni_hand_.hand_points_;   //返回手部点的位置

    return;
}

Mat CKinectOpenCV::GetColorImage()
{
    return color_image_;
}

Mat CKinectOpenCV::GetDepthImage()
{
    return depth_image_;
}

std::map<XnUserID, XnPoint3D> CKinectOpenCV::GetHandPoints()
{
    return hand_points_;
}

 

  ckinecthandsegment.h:

#ifndef KINECTHAND_H
#define KINECTHAND_H

#include "ckinectopencv.h"

using namespace cv;

#define MAX_HANDS_COLOR 10
#define MAX_HANDS_NUMBER  10

class CKinectHandSegment
{
public:
    CKinectHandSegment();
    ~CKinectHandSegment();
    void Initial();
    void StartKinectHand(); //启动kinect手部设备驱动
    Mat GetColorImageWithHandsPoint();
    Mat GetHandSegmentImage();
    Mat GetHandHandlingImage();
    Mat GetColorImage();
    Mat GetDepthImage();


private:
    CKinectOpenCV kinect_opencv_;
    vector<Scalar> hand_center_color_array_;//采用默认的10种颜色
    std::map<XnUserID, XnPoint3D> hand_points_;
    vector<unsigned int> hand_depth_;
    vector<Rect> hands_roi_;
    bool hand_segment_flag_;
    Mat color_image_with_handspoint_;   //带有手部中心位置的色彩图
    Mat color_image_;   //色彩图
    Mat depth_image_;
    Mat hand_segment_image_;
    Mat hand_handling_image_;
    Mat hand_segment_mask_;
};

#endif // KINECTHAND_H

 

  ckinecthandsegment.cpp:

#include <opencv2/core/core.hpp>
#include <opencv2/imgproc/imgproc.hpp>
#include "ckinecthandsegment.h"
#include "copennihand.h"
#include "ckinectopencv.h"

using namespace cv;
using namespace std;

#define DEPTH_SCALE_FACTOR 255./4096.
#define ROI_HAND_WIDTH 140
#define ROI_HAND_HEIGHT 140
#define MEDIAN_BLUR_K 5
#define XRES  640
#define YRES  480
#define DEPTH_SEGMENT_THRESH 5
#define HAND_LIKELY_AREA 2000


CKinectHandSegment::CKinectHandSegment()
{
}


CKinectHandSegment::~CKinectHandSegment()
{
}


void CKinectHandSegment::Initial()
{
    color_image_with_handspoint_ = kinect_opencv_.GetColorImage();
    depth_image_ = kinect_opencv_.GetDepthImage();
    {
        hand_center_color_array_.push_back(Scalar(255, 0, 0));
        hand_center_color_array_.push_back(Scalar(0, 255, 0));
        hand_center_color_array_.push_back(Scalar(0, 0, 255));
        hand_center_color_array_.push_back(Scalar(255, 0, 255));
        hand_center_color_array_.push_back(Scalar(255, 255, 0));
        hand_center_color_array_.push_back(Scalar(0, 255, 255));
        hand_center_color_array_.push_back(Scalar(128, 255, 0));
        hand_center_color_array_.push_back(Scalar(0, 128, 255));
        hand_center_color_array_.push_back(Scalar(255, 0, 128));
        hand_center_color_array_.push_back(Scalar(255, 128, 255));
    }
    vector<unsigned int> hand_depth_temp(MAX_HANDS_NUMBER, 0);
    hand_depth_ = hand_depth_temp;
    vector<Rect> hands_roi_temp(MAX_HANDS_NUMBER, Rect(XRES/2, YRES/2, ROI_HAND_WIDTH, ROI_HAND_HEIGHT));
    hands_roi_ = hands_roi_temp;
}


void CKinectHandSegment::StartKinectHand()
{
    kinect_opencv_.GetAllInformation();
}


Mat CKinectHandSegment::GetColorImage()
{
    return kinect_opencv_.GetColorImage();
}


Mat CKinectHandSegment::GetDepthImage()
{
    return kinect_opencv_.GetDepthImage();
}


/*该函数只是在Kinect获取的色彩图片上将手的中心点位置画出来而已,图片的其它地方不变*/
Mat CKinectHandSegment::GetColorImageWithHandsPoint()
{
    color_image_with_handspoint_ = kinect_opencv_.GetColorImage();
    hand_points_ = kinect_opencv_.GetHandPoints();
    for(auto itUser = hand_points_.cbegin(); itUser != hand_points_.cend(); ++itUser) {
        circle(color_image_with_handspoint_, Point(itUser->second.X, itUser->second.Y),
               5, hand_center_color_array_.at(itUser->first % hand_center_color_array_.size()), 3, 8);
    }

    return color_image_with_handspoint_;
}


Mat CKinectHandSegment::GetHandSegmentImage()
{
    hand_segment_flag_ = false;
    color_image_ = kinect_opencv_.GetColorImage();
    depth_image_ = kinect_opencv_.GetDepthImage();
    hand_points_ = kinect_opencv_.GetHandPoints();
    hand_segment_mask_ = Mat::zeros(color_image_.size(), CV_8UC1);  //  因为zeros是一个静态函数,所以不能直接用具体的对象去调用,而需要用类来调用

    for(auto itUser = hand_points_.cbegin(); itUser != hand_points_.cend(); ++itUser) {

        /*设置不同手部的深度*/
        hand_depth_.at(itUser->first % MAX_HANDS_COLOR) = (unsigned int)(itUser->second.Z* DEPTH_SCALE_FACTOR);//itUser->first会导致程序出现bug

        /*设置不同手部的不同感兴趣区域*/
        hands_roi_.at(itUser->first % MAX_HANDS_NUMBER) = Rect(itUser->second.X - ROI_HAND_WIDTH/2, itUser->second.Y - ROI_HAND_HEIGHT/2,
                                           ROI_HAND_WIDTH, ROI_HAND_HEIGHT);
        hands_roi_.at(itUser->first % MAX_HANDS_NUMBER).x =  itUser->second.X - ROI_HAND_WIDTH/2;
        hands_roi_.at(itUser->first % MAX_HANDS_NUMBER).y =  itUser->second.Y - ROI_HAND_HEIGHT/2;
        hands_roi_.at(itUser->first % MAX_HANDS_NUMBER).width = ROI_HAND_WIDTH;
        hands_roi_.at(itUser->first % MAX_HANDS_NUMBER).height = ROI_HAND_HEIGHT;
        if(hands_roi_.at(itUser->first % MAX_HANDS_NUMBER).x <= 0)
            hands_roi_.at(itUser->first % MAX_HANDS_NUMBER).x  = 0;
        if(hands_roi_.at(itUser->first % MAX_HANDS_NUMBER).x > XRES)
            hands_roi_.at(itUser->first % MAX_HANDS_NUMBER).x =  XRES;
        if(hands_roi_.at(itUser->first % MAX_HANDS_NUMBER).y <= 0)
            hands_roi_.at(itUser->first % MAX_HANDS_NUMBER).y = 0;
        if(hands_roi_.at(itUser->first % MAX_HANDS_NUMBER).y > YRES)
            hands_roi_.at(itUser->first % MAX_HANDS_NUMBER).y =  YRES;
    }

    //取出手的mask部分,不管原图像时多少通道的,mask矩阵声明为单通道就ok
    for(auto itUser = hand_points_.cbegin(); itUser != hand_points_.cend(); ++itUser) {
        for(int i = hands_roi_.at(itUser->first % MAX_HANDS_NUMBER).x; i < min(hands_roi_.at(itUser->first % MAX_HANDS_NUMBER).x+hands_roi_.at(itUser->first % MAX_HANDS_NUMBER).width, XRES); i++)
            for(int j = hands_roi_.at(itUser->first % MAX_HANDS_NUMBER).y; j < min(hands_roi_.at(itUser->first % MAX_HANDS_NUMBER).y+hands_roi_.at(itUser->first % MAX_HANDS_NUMBER).height, YRES); j++) {
                hand_segment_mask_.at<unsigned char>(j, i) = ((hand_depth_.at(itUser->first % MAX_HANDS_NUMBER)-DEPTH_SEGMENT_THRESH) < depth_image_.at<unsigned char>(j, i))
                                                            & ((hand_depth_.at(itUser->first % MAX_HANDS_NUMBER)+DEPTH_SEGMENT_THRESH) > depth_image_.at<unsigned char>(j,i));
                hand_segment_mask_.at<unsigned char>(j, i) = 255*hand_segment_mask_.at<unsigned char>(j, i);
            }
        }

    medianBlur(hand_segment_mask_, hand_segment_mask_, MEDIAN_BLUR_K);
    hand_segment_image_.convertTo(hand_segment_image_, CV_8UC3, 0, 0 ); //  需要清零
    color_image_.copyTo(hand_segment_image_, hand_segment_mask_);
    hand_segment_flag_ = true;  //返回之前将分割标志置位为1,表示已经完成分割函数

    return hand_segment_image_;
}


Mat CKinectHandSegment::GetHandHandlingImage()
{
    /*对mask图像进行轮廓提取,并在手势识别图像中画出来*/
    std::vector< std::vector<Point> > contours;
    CV_Assert(hand_segment_flag_);  //  因为后面要用到分割函数的mask图,所以这里先要保证调用过分割函数
    findContours(hand_segment_mask_, contours, CV_RETR_LIST, CV_CHAIN_APPROX_SIMPLE);//找出mask图像的轮廓
    hand_handling_image_ = Mat::zeros(color_image_.rows, color_image_.cols, CV_8UC3);

    for(int i = 0; i < contours.size(); i++) {  //只有在检测到轮廓时才会去求它的多边形,凸包集,凹陷集
        /*找出轮廓图像多边形拟合曲线*/
        Mat contour_mat = Mat(contours[i]);
        if(contourArea(contour_mat) > HAND_LIKELY_AREA) {   //比较有可能像手的区域
            std::vector<Point> approx_poly_curve;
            approxPolyDP(contour_mat, approx_poly_curve, 10, true);//找出轮廓的多边形拟合曲线
            std::vector< std::vector<Point> > approx_poly_curve_debug;
            approx_poly_curve_debug.push_back(approx_poly_curve);

             drawContours(hand_handling_image_, contours, i, Scalar(255, 0, 0), 1, 8); //画出轮廓
//            drawContours(hand_handling_image_, approx_poly_curve_debug, 0, Scalar(256, 128, 128), 1, 8); //画出多边形拟合曲线

            /*对求出的多边形拟合曲线求出其凸包集*/
            vector<int> hull;
            convexHull(Mat(approx_poly_curve), hull, true);
            for(int i = 0; i < hull.size(); i++) {
                circle(hand_handling_image_, approx_poly_curve[hull[i]], 2, Scalar(0, 255, 0), 2, 8);
            }

            /*对求出的多边形拟合曲线求出凹陷集*/
            std::vector<Vec4i> convexity_defects;
            if(Mat(approx_poly_curve).checkVector(2, CV_32S) > 3)
                convexityDefects(approx_poly_curve, Mat(hull), convexity_defects);
            for(int i = 0; i < convexity_defects.size(); i++) {
                circle(hand_handling_image_, approx_poly_curve[convexity_defects[i][2]] , 2, Scalar(0, 0, 255), 2, 8);

            }
        }
    }

    /**画出手势的中心点**/
    for(auto itUser = hand_points_.cbegin(); itUser != hand_points_.cend(); ++itUser) {
        circle(hand_handling_image_, Point(itUser->second.X, itUser->second.Y), 3, Scalar(0, 255, 255), 3, 8);
    }

    return hand_handling_image_;
}

 

  main.cpp:

#include <opencv2/core/core.hpp>
#include <opencv2/highgui/highgui.hpp>
#include "ckinectopencv.h"
#include "ckinecthandsegment.h"

using namespace std;
using namespace cv;

int main()
{
    CKinectHandSegment kinect_hand_segment;
    Mat color_image ;
    Mat depth_image ;
    Mat hand_segment;
    Mat hand_handling_image;

    kinect_hand_segment.Initial();
    while(1)
    {
        kinect_hand_segment.StartKinectHand();
        color_image = kinect_hand_segment.GetColorImageWithHandsPoint();
        hand_segment = kinect_hand_segment.GetHandSegmentImage();
        hand_handling_image = kinect_hand_segment.GetHandHandlingImage();
        depth_image = kinect_hand_segment.GetDepthImage();

        imshow("color_image", color_image);
        imshow("depth_image", depth_image);
        imshow("hand_segment", hand_segment);
        imshow("hand_handling", hand_handling_image);
        waitKey(30);
    }

    return 0;
}

 

实验总结:把这些基本功能类设计好了后,就可以更方面测试我后面的手势识别算法了,加油!

参考资料:

Kinect+OpenNI学习笔记之3(获取kinect的数据并在Qt中显示的类的设计)

Kinect+OpenNI学习笔记之11(OpenNI驱动kinect手势相关的类的设计)

Kinect+OpenNI学习笔记之12(简单手势所表示的数字的识别) 

转载注明来源:CV视觉网 » Kinect+OpenNI学习笔记之13(Kinect驱动类,OpenCV显示类和手部预分割类的设计)

分享到:更多 ()
扫描二维码,给作者 打赏
pay_weixinpay_weixin

请选择你看完该文章的感受:

0不错 0超赞 0无聊 0扯淡 0不解 0路过

评论 3

评论前必须登录!