OpenCV_学习基础.ppt

运动对象跟踪：模块(6) 轨迹分析模块 /* list of BLOB TRAJECTORY ANALYSIS modules */ CvBlobTrackAnalysis* cvCreateModuleBlobTrackAnalysisDetector(); typedef struct DefModule_BlobTrackAnalysis{ …… } DefModule_BlobTrackAnalysis; DefModule_BlobTrackAnalysis BlobTrackAnalysis_Modules[] ={ {cvCreateModuleBlobTrackAnalysisHistPVS,HistPVS,Histogram of 5D feature vector analysis (x,y,vx,vy,state)}, {NULL,None,No trajectory analiser}, {cvCreateModuleBlobTrackAnalysisHistP,HistP,Histogram of 2D feature vector analysis (x,y)}, {cvCreateModuleBlobTrackAnalysisHistPV,HistPV,Histogram of 4D feature vector analysis (x,y,vx,vy)}, {cvCreateModuleBlobTrackAnalysisHistSS,HistSS,Histogram of 4D feature vector analysis (startpos,endpos)}, {cvCreateModuleBlobTrackAnalysisTrackDist,TrackDist,Compare tracks directly}, {cvCreateModuleBlobTrackAnalysisIOR,IOR,Integrator (by OR operation) of several analysers }, {NULL,NULL,NULL} }; * DEMO 程序源码参见”OpenCV样例”blobtrack 可以在blobtrack基础上添加自己的算法以实现前景检测、团块检测与跟踪、轨迹生成与后处理等功能 留待课后思考 * 基于OpenCV的研究开发：DEMO 视频文字/人脸检测、分割与识别 视频指纹提取与比对 * * 视频文字/人脸检测与识别DEMO 文字和人脸是图像/视频检索中最重要的两种语义线索 文字：文字检测、分割与识别（OCR） 人脸：人脸检测、标注与识别 视纹提取与应用系统 * * Thanks! * * * * * * * * * * * * 程序示例2：Canny边缘检测 Edges are the boundaries separating regions with different brightness or color. J.Canny suggested in [Canny86] an efficient method for detecting edges. It takes grayscale image on input and returns bi-level image where non-zero pixels mark detected edges. cvCanny：采用Canny算子寻找并标志输入图像中的边缘void cvCanny(const CvArr* image, CvArr*edges, double threshold1, double threshold2, int aperture_size=3); Threshold1和threshold2：两个阈值，其中小的阈值用于控制边缘连接，打阈值用于控制强边缘的初始分割 aperture_size：Sobel算子内核大小 Canny算子： 用高斯滤波器平滑图像； 用一阶偏导的有限差分计算梯度的幅值和方向 对梯度幅值进行非极大值抑制 用双阈值算法检测和链接边缘 采用高斯平滑函数 * 程序示例2：Canny边缘检测(1) #include cv.h #include highgui.h char wndname[] = Edge; char tbarname[] = Threshold; int edge_thresh = 1; IplImage *image = 0, *cedge = 0, *gray = 0, *edge = 0; void on_trackbar(int h){// 定义跟踪条的 c