SIFT圖像配准 python3.6 + opencv3.3代碼

opencv3.x 中部分函數有改變：

1. SIFT：可以采用help(cv2.xfeatures2d)查詢

2.drawKeypoints： 同樣采用help()方法查詢

opencv3 版本sift,surf 及其他不穩定的算法函數都放在opencv3.x的contrib版里。該模塊下載地址 https://www.lfd.uci.edu/~gohlke/pythonlibs/

import cv2 
import numpy as np

def sift_kp(image):
    gray_image = cv2.cvtColor(image,cv2.COLOR_BGR2GRAY)
    sift=cv2.xfeatures2d.SIFT_create()
    kp,des = sift.detectAndCompute(image,None)
    kp_image = cv2.drawKeypoints(gray_image,kp,None)
    return kp_image,kp,des


def get_good_match(des1,des2):
    bf = cv2.BFMatcher()
    matches = bf.knnMatch(des1, des2, k=2) #des1為模板圖，des2為匹配圖
    matches = sorted(matches,key=lambda x:x[0].distance/x[1].distance)
    good = []
    for m, n in matches:
        if m.distance < 0.75 * n.distance:
            good.append(m)
    return good


def siftImageAlignment(img1,img2):
   _,kp1,des1 = sift_kp(img1)
   _,kp2,des2 = sift_kp(img2)
   goodMatch = get_good_match(des1,des2) 
   if len(goodMatch) > 4:
       ptsA= np.float32([kp1[m.queryIdx].pt for m in goodMatch]).reshape(-1, 1, 2)
       ptsB = np.float32([kp2[m.trainIdx].pt for m in goodMatch]).reshape(-1, 1, 2)
       ransacReprojThreshold = 4
       H, status =cv2.findHomography(ptsA,ptsB,cv2.RANSAC,ransacReprojThreshold);
       imgOut = cv2.warpPerspective(img2, H, (img1.shape[1],img1.shape[0]),flags=cv2.INTER_LINEAR + cv2.WARP_INVERSE_MAP)
   return imgOut,H,status

img1 = cv2.imread(r'sift_img/8.png') 
img2 = cv2.imread(r'sift_img/7.png') 

_,kp1,des1 = sift_kp(img1)
_,kp2,des2 = sift_kp(img2)
goodMatch = get_good_match(des1,des2)

img3 = cv2.drawMatches(img1, kp1, img2, kp2, goodMatch[:5], None, flags=2)
#----or----
#goodMatch = np.expand_dims(goodMatch,1)
#img3 = cv2.drawMatchesKnn(img1, kp1, img2, kp2, goodMatch[:5], None, flags=2)

cv2.imshow('img',img3) 
cv2.waitKey(0) 
cv2.destroyAllWindows()

 

SIFT特征詳解：

http://www.cnblogs.com/wangguchangqing/p/4853263.html

http://blog.csdn.net/abcjennifer/article/details/7639681