opencv_角点算法

harris检测原理

import numpy as np
import  cv2 as cv
import matplotlib.pyplot as plt
img =cv.imread("opencv_3.png")
gray =cv.cvtColor(img,cv.COLOR_RGB2GRAY)
gray=np.float32(gray)
dst=cv.cornerHarris(gray,2,3,0.04)
img [dst>0.001*dst.max()]=[0,0,255]
plt.imshow(img[:,:,::-1])
cv.waitKey(0)

Shi_Tomasi算法

Shi_Tomasi算法是对Harris角点算法的改进，一般会比Harris算法得到更好的角点。Harris
算法的角点响应函数是将矩阵M的行列式值与M的迹相减，利用差值判断是否为角点。若
矩阵M的两个特征值中的较小的一个大于阈值，则认为是角点。

import numpy as np
import  cv2 as cv
import matplotlib.pyplot as plt

img=cv.imread("opencv_4.png")
gray=cv.cvtColor(img,cv.COLOR_RGB2GRAY)
# 角点检测
corners=cv.goodFeaturesToTrack(gray,1000,0.01,10)

for i in corners:
    x,y=i.ravel()
    cv.circle(img,(x,y),2,(0,0,255),-1)
plt.figure(figsize=(10,8),dpi=100)
plt.imshow(img[:,:,::-1])
plt.xticks([]),plt.yticks([])
plt.show()

SIFT

SIFT算法的实质是在不同尺度空间上查找关键点（特征点），并计算出关键点的方向。
SIFT所查找的关键点是一些十分突出的，不会因为光照，放射变换和噪音的因素影响的点
如角点、边缘点、暗区的亮点及亮区的暗点。

import cv2 as cv
import numpy as np
import matplotlib.pyplot as plt

img=cv.imread("opencv_4.png")
gray=cv.cvtColor(img,cv.COLOR_RGB2GRAY)
sift=cv.xfeatures2d.SIFT_create()

kp,des =sift.detectAndCompute(gray,None)

cv.drawKeypoints(img,kp,img,flags=cv.DRAW_MATCHES_FLAGS_DRAW_RICH_KEYPOINTS)
plt.figure(figsize=(8,6),dpi=100)
plt.imshow(img[:,:,::-1])
plt.xticks([]),plt.yticks([])
plt.show()

Fast

import cv2 as cv
import numpy as np
import matplotlib.pyplot as plt

img=cv.imread("opencv_4.png")
gray=cv.cvtColor(img,cv.COLOR_RGB2GRAY)
fast=cv.FastFeatureDetector_create(threshold=30)
kp=fast.delete(img,None)
img2=cv.drawKeypoints(img,kp,None,color=(0,0,255))

plt.imshow(img2[:,:,::-1])
fast.setNonmaxSuppression(0)