0x00. 特征识别

这里主要用到两个函数:

GoodFeaturesToTrackextractSURF

  • GoodFeaturesToTrack: 在图像中寻找具有大特征值的角点。
  • SURF算法: 是一个稳健的图像识别和描述算法。

总之这俩个我目前也不清楚能用来干嘛,以后用到了在更新吧。

1234567891011121314151617181920212223242526272829303132333435363738 import cv2.cv as cvimport math im = cv.LoadImage(\”img/church.png\”, cv.CV_LOAD_IMAGE_GRAYSCALE)im2 = cv.CloneImage(im) # Goodfeatureto track algorithmeigImage = cv.CreateMat(im.height, im.width, cv.IPL_DEPTH_32F)tempImage = cv.CloneMat(eigImage)cornerCount = 500quality = 0.01minDistance = 10 corners = cv.GoodFeaturesToTrack(im, eigImage, tempImage, cornerCount, quality, minDistance) radius = 3thickness = 2 for (x,y) in corners:    cv.Circle(im, (int(x),int(y)), radius, (255,255,255), thickness) cv.ShowImage(\”GoodfeaturesToTrack\”, im) #SURF algorithmhessthresh = 1500 # 400 500dsize = 0 # 1layers = 1 # 3 10 keypoints, descriptors = cv.ExtractSURF(im2, None, cv.CreateMemStorage(), (dsize, hessthresh, 3, layers))for ((x, y), laplacian, size, dir, hessian) in keypoints:    cv.Circle(im2, (int(x),int(y)), cv.Round(size/2), (255,255,255), 1)    x2 = x+((size/2)*math.cos(dir))    y2 = y+((size/2)*math.sin(dir))    cv.Line(im2, (int(x),int(y)), (int(x2),int(y2)), (255,255,255), 1) cv.ShowImage(\”SURF \”, im2) cv.WaitKey(0)

0x01. 人脸识别

可以使用 OpenCV 训练好的级联分类器来识别图像中的人脸,当然还有很多其他的分类器:例如表情识别,鼻子等,具体可在这里下载:

OpenCV分类器

具体使用代码:

12345678910111213141516171819202122232425262728293031323334353637383940 #import library – MUST use cv2 if using opencv_traincascadeimport cv2 # rectangle color and strokecolor = (0,0,255)       # reverse of RGB (B,G,R) – weirdstrokeWeight = 1        # thickness of outline # set window namewindowName = \”Object Detection\” # load an image to search for facesimg = cv2.imread(\”mao.jpg\”) # load detection file (various files for different views and uses)cascade = cv2.CascadeClassifier(\”haarcascade_frontalface_alt.xml\”) # preprocessing, as suggested by: http://www.bytefish.de/wiki/opencv/object_detection# img_copy = cv2.resize(img, (img.shape[1]/2, img.shape[0]/2))# gray = cv2.cvtColor(img_copy, cv2.COLOR_BGR2GRAY)# gray = cv2.equalizeHist(gray) # detect objects, return as listrects = cascade.detectMultiScale(img) # display until escape key is hitwhile True:     # get a list of rectangles    for x,y, width,height in rects:        cv2.rectangle(img, (x,y), (x+width, y+height), color, strokeWeight)     # display!    cv2.imshow(windowName, img)     # escape key (ASCII 27) closes window    if cv2.waitKey(20) == 27:        break # if esc key is hit, quit!exit()

效果: