1、Kinect+OpenNI学习笔记之9不需要骨骼跟踪的人体手部分割Kinect+OpenNI学习笔记之9(不需要骨骼跟踪的人体手部分割)前言手势识别非常重要的一个特点是要体验要好,即需要以用户为核心。而手势的定位一般在手势识别过程的前面,在上一篇博文Kinect+OpenNI学习笔记之8(Robert Walter手部提取代码的分析)中已经介绍过怎样获取手势区域,且取得了不错的效果,但是那个手势部位的提取有一个大的缺点,即需要人站立起来,当站立起来后才能够分隔出手。而手势在人之间的交流时,并不一定要处于站立状态,所以这不是一个好的HCI。因此本文介绍的手势部位的提取并不需要人处于站立状态,同样
2、取得了不错的效果。实验说明其实,本实验实现的过程非常简单。首先通过手部的跟踪来获取手所在的坐标,手部跟踪可以参考本人前面的博文:Kinect+OpenNI学习笔记之7(OpenNI自带的类实现手部跟踪)。当定位到手所在的坐标后,因为该坐标是3D的,因此在该坐标领域的3维空间领域内提取出手的部位即可,整个过程的大概流程图如下:OpenCV知识点总结:调用Mat:copyTo()函数时,如果需要有mask操作,则不管源图像是多少通道的,其mask矩阵都要定义为单通道,另外可以对一个mask矩阵画一个填充的矩形来达到使mask矩阵中对应ROI的位置的值为设定值,这样就不需要去一一扫描赋值了。在使用O
3、penCV的Mat矩阵且需要对该矩阵进行扫描时,一定要注意其取值顺序,比如说列和行的顺序,如果弄反了,则经常会报内存错误。实验结果本实验并不要求人的手一定要放在人体的前面,且也不需要人一定是处在比较简单的背景环境中,本实验结果允许人处在复杂的背景环境下,且手可以到处随便移动。当然了,环境差时有时候效果就不太好。下面是3张实验结果的截图,手势分隔图1:手势分隔图2:手势分隔图3:实验主要部分代码即注释(附录有工程code下载链接):main.cpp:#include #include opencv2/highgui/highgui.hpp#include opencv2/imgproc/imgp
4、roc.hpp#include #include copenni.cpp#include #define DEPTH_SCALE_FACTOR 255./4096.#define ROI_HAND_WIDTH 140#define ROI_HAND_HEIGHT 140#define MEDIAN_BLUR_K 5int XRES = 640;int YRES = 480;#define DEPTH_SEGMENT_THRESH 5using namespace cv;using namespace xn;using namespace std;int main (int argc, char
5、 *argv) COpenNI openni; int hand_depth; Rect roi; roi.x = XRES/2; roi.y = YRES/2; roi.width = ROI_HAND_WIDTH; roi.height = ROI_HAND_HEIGHT; if(!openni.Initial() return 1; namedWindow(color image, CV_WINDOW_AUTOSIZE); namedWindow(depth image, CV_WINDOW_AUTOSIZE); namedWindow(hand_segment, CV_WINDOW_A
6、UTOSIZE);/显示分割出来的手的区域 if(!openni.Start() return 1; while(1) if(!openni.UpdateData() return 1; /*获取并显示色彩图像*/ Mat color_image_src(openni.image_metadata.YRes(), openni.image_metadata.XRes(), CV_8UC3, (char *)openni.image_metadata.Data(); Mat color_image; cvtColor(color_image_src, color_image, CV_RGB2BG
7、R); circle(color_image, Point(hand_point.X, hand_point.Y), 5, Scalar(255, 0, 0), 3, 8); imshow(color image, color_image); /*获取并显示深度图像*/ Mat depth_image_src(openni.depth_metadata.YRes(), openni.depth_metadata.XRes(), CV_16UC1, (char *)openni.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 Mat de
8、pth_image; depth_image_src.convertTo(depth_image, CV_8U, DEPTH_SCALE_FACTOR); imshow(depth image, depth_image); /*下面的代码是提取手的轮廓部分*/ hand_depth = hand_point.Z * DEPTH_SCALE_FACTOR; roi.x = hand_point.X - ROI_HAND_WIDTH/2; roi.y = hand_point.Y - ROI_HAND_HEIGHT/2; if(roi.x = XRES) roi.x = XRES; if(roi.
9、y = YRES) roi.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 Mat hand_segment_mask(depth_image.size(), CV_8UC1, Scalar:all(0); for(int i = roi.x; i std:min(roi.x+roi.width, XRES); i+) for(int j = roi.y; j std:min(roi.y+roi.height, YRES); j+) hand_segment_mask.at(j, i) = (hand_depth-DEPTH_SEGMENT
10、_THRESH) depth_image.at(j, i) & (hand_depth+DEPTH_SEGMENT_THRESH) depth_image.at(j,i); medianBlur(hand_segment_mask, hand_segment_mask, MEDIAN_BLUR_K); Mat hand_segment(color_image.size(), CV_8UC3); color_image.copyTo(hand_segment, hand_segment_mask); imshow(hand_segment, hand_segment); waitKey(20);
11、 copenni,cpp:#ifndef COPENNI_CLASS#define COPENNI_CLASS#include #include #include using namespace xn;using namespace std;static DepthGenerator depth_generator;static HandsGenerator hands_generator;static XnPoint3D hand_point;static std:mapXnUserID, vector hands_track_points;class COpenNIpublic: COpe
12、nNI() context.Release();/释放空间 bool Initial() /初始化 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 ge
13、nerator 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.SetMapO
14、utputMode(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_generat
15、or.AddGesture(RaiseHand, NULL); gesture_generator.AddGesture(MovingHand, NULL); /产生手部的node status = hands_generator.Create(context); if(CheckError(Create hand generaotr error!) return false; /产生人体node status = user_generator.Create(context); if(CheckError(Create gesturen generator error!) return fal
16、se; /视角校正 status = depth_generator.GetAlternativeViewPointCap().SetViewPoint(image_generator); if(CheckError(Cant 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, CBGestureProg
17、ress, NULL, gesture_cb); /设置于手部有关的回调函数 XnCallbackHandle hands_cb; hands_generator.RegisterHandCallbacks(HandCreate, HandUpdate, HandDestroy, NULL, hands_cb); /设置有人进入视野的回调函数 XnCallbackHandle new_user_handle; user_generator.RegisterUserCallbacks(CBNewUser, NULL, NULL, new_user_handle); user_generator.
18、GetSkeletonCap().SetSkeletonProfile(XN_SKEL_PROFILE_ALL);/设定使用所有关节(共15个) /设置骨骼校正完成的回调函数 XnCallbackHandle calibration_complete; user_generator.GetSkeletonCap().RegisterToCalibrationComplete(CBCalibrationComplete, NULL, calibration_complete); return true; bool Start() status = context.StartGeneratingA
19、ll(); if(CheckError(Start generating error!) return false; return true; bool 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; /得到色彩图像的node Im
20、ageGenerator& getImageGenerator() return image_generator; /得到深度图像的node DepthGenerator& getDepthGenerator() return depth_generator; /得到人体的node UserGenerator& getUserGenerator() return user_generator; /得到手势姿势node GestureGenerator& getGestureGenerator() return gesture_generator; public: DepthMetaData d
21、epth_metadata; ImageMetaData image_metadata;/ static std:mapXnUserID, vector hands_track_points;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) if(status != XN_STATUS_OK ) /QMessageBox:critical(NULL, error, xnGetStatusString(status); cerr error : xnGetStatusString( status ) hands_
22、generator.StartTracking(*pIDPosition); hands_generator.StartTracking(*pIDPosition); /手势开始检测的回调函数 static void XN_CALLBACK_TYPE CBGestureProgress(xn:GestureGenerator &generator, const XnChar *strGesture, const XnPoint3D *pPosition, XnFloat fProgress, void *pCookie) / COpenNI *openni = (COpenNI*)pCooki
23、e; / openni-hands_generator.StartTracking(*pPosition); hands_generator.StartTracking(*pPosition); /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) / COpenNI *openni = (COpenNI*)pCookie; XnPoint3D pro
24、ject_pos; depth_generator.ConvertRealWorldToProjective(1, pPosition, &project_pos); / openni-hand_point = project_pos; /返回手部所在点的位置 hand_point = project_pos; pairXnUserID, vector hand_track_point(xUID, vector(); hand_track_point.second.push_back(project_pos); hands_track_points.insert(hand_track_poin
25、t); /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) / COpenNI *openni = (COpenNI*)pCookie; XnPoint3D project_pos; depth_generator.ConvertRealWorldToProjective(1, pPosition, &project_pos); / openni-h
26、and_point = project_pos; /返回手部所在点的位置 hand_point = project_pos; hands_track_points.find(xUID)-second.push_back(project_pos); /销毁手部的回调函数 static void XN_CALLBACK_TYPE HandDestroy(HandsGenerator& rHands, XnUserID xUID, XnFloat fTime, void* pCookie) / COpenNI *openni = (COpenNI*)pCookie; /openni-hand_poi
27、nt.clear(); /返回手部所在点的位置 hands_track_points.erase(hands_track_points.find(xUID); /有人进入视野时的回调函数 static void XN_CALLBACK_TYPE CBNewUser(UserGenerator &generator, XnUserID user, void *p_cookie) /得到skeleton的capability,并调用RequestCalibration函数设置对新检测到的人进行骨骼校正 generator.GetSkeletonCap().RequestCalibration(us
28、er, true); /完成骨骼校正的回调函数 static void XN_CALLBACK_TYPE CBCalibrationComplete(SkeletonCapability &skeleton, XnUserID user, XnCalibrationStatus calibration_error, void *p_cookie) if(calibration_error = XN_CALIBRATION_STATUS_OK) skeleton.StartTracking(user);/骨骼校正完成后就开始进行人体跟踪了 else UserGenerator *p_user = (UserGenerator*)p_cookie; skeleton.RequestCalibration(user, true);/骨骼校正失败时重新设置对人体骨骼继续进行校正 private: XnStatus status; Context context;
copyright@ 2008-2022 冰豆网网站版权所有
经营许可证编号:鄂ICP备2022015515号-1