基于ORB特征提取算法图像匹配 python实现

简单的ORB匹配算法：

# coding=utf-8import cv2import numpy as npimg1 = cv2.imread('test1.png')img2 = cv2.imread('test12.png')#最大特征点数,需要修改，5000太大。orb = cv2.ORB_create(5000)kp1, des1 = orb.detectAndCompute(img1,None)kp2, des2 = orb.detectAndCompute(img2,None)#提取并计算特征点bf = cv2.BFMatcher(cv2.NORM_HAMMING)#knn筛选结果matches = bf.knnMatch(des1, trainDescriptors = des2, k = 2)good = [m for (m,n) in matches if m.distance < 0.75*n.distance]#查看最大匹配点数目print len(good)

基于匹配算法，做了一个基于python的图像搜索引擎（github）：

基于python的图像搜索引擎