1 原理(略)
2.1 OpenCV中的Harris角点检测
OpenCV中的函数cv2.cornerHarris()可以用来进行角点检测。参数如下：
img：数据类型为float32的输入图像。blockSize：角点检测中要考虑的邻域大小。ksize：Sobel求导中使用的窗口大小。k：Harris角点检测方程中的自由参数，取值参数为[0.04,0.06]
代码演示：
import cv2import numpy as npimg = cv2.imread('xiangqi.jpg')gray = cv2.cvtColor(img,cv2.COLOR_BGR2GRAY)gray = np.float32(gray)#输入图像必须是float32，最后一个参数在0.04到0.05dst = cv2.cornerHarris(gray,2,3,0.04)dst = cv2.dilate(dst,None)# Threshold for an optimal value, it may vary depending on the image.img [dst>0.01*dst.max()] = [0,0,255]cv2.imshow('dst',img)if cv2.waitKey(0) == 27:cv2.destroyAllWindows()
2.2 亚像素级精确度的角点
有时我们需要最大精度的角点检测。OpenCV 为我们提供了函数 cv2.cornerSubPix()，它可以提供亚像素级别的角点检测。下面是一个例子。首先我们要找到 Harris角点，然后将角点的重心传给这个函数进行修正。Harris 角点用红色像素标出，绿色像素是修正后的像素。在使用这个函数是我们要定义一个迭代停止条件。当迭代次数达到或者精度条件满足后迭代就会停止。我们同样需要定义进行角点搜索的邻域大小。
import cv2import numpy as np# find Harris cornersimg = cv2.imread('xiangqi.jpg')gray = cv2.cvtColor(img,cv2.COLOR_BGR2GRAY)gray = np.float32(gray)dst = cv2.cornerHarris(gray,2,3,0.04)dst = cv2.dilate(dst,None)ret,dst = cv2.threshold(dst,0.01*dst.max(),255,0)dst = np.uint8(dst)# find centroids#connectedComponentsWithStats(InputArray image, OutputArray labels, OutputArray stats,#OutputArray centroids, int connectivity=8, int ltype=CV_32S)ret,labels,stats,centroids = cv2.connectedComponentsWithStats(dst)# define the criteria to stop and refine the cornerscriteria = (cv2.TERM_CRITERIA_EPS + cv2.TERM_CRITERIA_MAX_ITER,100,0.001)#Python: cv2.cornerSubPix(image, corners, winSize, zeroZone, criteria)#zeroZone – Half of the size of the dead region in the middle of the search zone#over which the summation in the formula below is not done. It is used sometimes# to avoid possible singularities of the autocorrelation matrix. The value of (-1,-1)# indicates that there is no such a size.# 返回值由角点坐标组成的一个数组（而非图像）corners = cv2.cornerSubPix(gray,np.float32(centroids),(5,5),(-1,-1),criteria)res = np.hstack((centroids,corners))#np.int0 可以用来省略小数点后面的数字（非四㮼五入）。res = np.int0(res)img[res[:,1],res[:,0]] = [0,0,255]img[res[:,3],res[:,2]] = [0,2555,0]cv2.imshow('111',img)cv2.waitKey(0)
结果如下，为了方便查看我们对角点的部分进行了放大：
