1

我正在使用OpenCV,SIFT并且Homography为了检测图片中的所有对象。

我的全局图片如下所示:

在此处输入图像描述

而且我想检测图片上的所有灯,即使每个灯之间的方向不完全相同。

我的模型图片如下所示:

在此处输入图像描述

我用 Python 编写了这个脚本:

#-*- coding: utf-8! -*-

import os, shutil
import numpy as np
import cv2

#########################
# SIFT descriptors part #
#########################

img1 = cv2.imread('/Users/test/Desktop/SIFT/Ville/ville.jpg',0)
img2 = cv2.imread('/Users/test/Desktop/SIFT/Ville/lampe.jpg',0)

##########################
# Initiate SIFT detector #
##########################

sift = cv2.xfeatures2d.SIFT_create()

kp1, des1 = sift.detectAndCompute(img1,None)
kp2, des2 = sift.detectAndCompute(img2,None)

FLANN_INDEX_KDTREE = 0
index_params = dict(algorithm = FLANN_INDEX_KDTREE, trees = 5)
search_params = dict(checks = 50)

flann = cv2.FlannBasedMatcher(index_params, search_params)
#bf = cv2.BFMatcher()
matches = flann.knnMatch(des1,des2,k=2)

good = []
for m,n in matches :
    if m.distance < 0.7*n.distance :
        good.append([m])

MIN_MATCH_COUNT = 3

if len(good)>MIN_MATCH_COUNT:
    src_pts = np.float32([ kp1[m.queryIdx].pt for m in good ]).reshape(-1,1,2)
    dst_pts = np.float32([ kp2[m.trainIdx].pt for m in good ]).reshape(-1,1,2)

    M, mask = cv2.findHomography(src_pts, dst_pts, cv2.RANSAC,5.0)
    matchesMask = mask.ravel().tolist()

    h,w = img1.shape
    pts = np.float32([ [0,0],[0,h-1],[w-1,h-1],[w-1,0] ]).reshape(-1,1,2)
    dst = cv2.perspectiveTransform(pts,M)

    img2 = cv2.polylines(img2,[np.int32(dst)],True,255,3, cv2.LINE_AA)

else:
    print "Not enough matches are found - %d/%d" % (len(good),MIN_MATCH_COUNT)
    matchesMask = None

draw_params = dict(matchColor = (0,255,0), # draw matches in green color
                   singlePointColor = None,
                   matchesMask = matchesMask, # draw only inliers
                   flags = 2)

img3 = cv2.drawMatches(img1,kp1,img2,kp2,good,None,**draw_params)

plt.imshow(img3, 'grayfinal.jpg'),plt.show()
cv2.imwrite('matches.jpg',img3)

我收到此错误:

Traceback (most recent call last):
  File "image.py", line 34, in <module>
    src_pts = np.float32([ kp1[m.queryIdx].pt for m in good ]).reshape(-1,1,2)
AttributeError: 'list' object has no attribute 'queryIdx'

你有想法吗 ?

编辑 :

通过假设的解决方案,我得到了一些看起来正确的东西。但是我怎么能检测到图片上的其他灯呢?

在此处输入图像描述

4

2 回答 2

6

你应该只追加mgood,而不是做good.append([m]).

因为现在其中的每个元素good都是一个包含一个元素 ( ) 的列表m,并且您尝试访问它的queryIdx. 这就是你收到这个的原因AttributeError: 'list' object has no attribute 'queryIdx'

于 2017-01-06T11:33:01.237 回答
0

您可以维护两个列表,例如

good = []
good_without_list = []

for m, n in matches:
    if m.distance < 0.75 * n.distance:
        good.append([m])
        good_without_list.append(m)

然后你可以使用

  • “cv2.drawMatchesKnn”的“好”列表

    knn_image = cv2.drawMatchesKnn(img1, kp1, img2, kp2, good, None, flags=2)
    
  • “cv2.drawMatches”的“good_without_list”列表(在您的情况下)

    img3 = cv2.drawMatches(img1, kp1, img2, kp2, good_without_list, None, **draw_params)
    
于 2018-06-25T09:42:52.427 回答