-
Notifications
You must be signed in to change notification settings - Fork 88
/
pic_feature.py
199 lines (165 loc) · 5.91 KB
/
pic_feature.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
#!/usr/bin/env python
# encoding:utf-8
"""
图像特征检测算法与相似图片搜索
基于SIFT,SURF,ORB特征结合opencv实现图片特征与相似图片搜索
author : @h-j-13
time : 2018-8-8
ref : https://blog.csdn.net/sinat_26917383/article/details/63306206?locationNum=6&fps=1
ref : https://blog.csdn.net/samkieth/article/details/49615669
ref : https://blog.csdn.net/eds95/article/details/70146689
>>>image = cv2.imread('test.jpg') # 通过OPENCV读取文件
>>>sfit(image) # 进行sfit特征变换
[[31. 1. 0. ... 92. 1. 0.]
...
[10. 30. 5. ... 0. 0. 0.]]
>>>orb(image) # 进行ORB特征变换
[[ 88 185 59 ... 0 230 186]
...
[116 221 63 ... 123 37 122]]
>>>surf(image,'surf_test.jpg') # 进行SURF特征变化,并保存提取特征后的图片
[[-3.3737912e-03 -1.9819981e-03 3.3737912e-03 ... 4.7051071e-04
4.5059556e-03 1.3064195e-03]
....
>>>i1 = cv2.imread('test.jpg')
>>>i2 = cv2.imread('test2.jpg')
>>>print compare_pic(i1,i2,show=True) # 进行特征比对
165
"""
import cv2
import numpy as np
# 特征变换全局对象
D = 200
SFIT = None
ORB = None
SURF = None
"""
SIFT
尺度不变特征转换(Scale-invariant feature transform或SIFT)
是一种电脑视觉的算法用来侦测与描述影像中的局部性特征，
它在空间尺度中寻找极值点，并提取出其位置、尺度、旋转不变量，
"""
def sfit(image, image_save_path=''):
"""
图片进行sfit变换,生成特征点返回特征矩阵
:param image: cv2.imread
:param image_save_path: 如果指定了保存地址,则会将特征提取过的图片保存到制定的地址中
:return:sfit变换后的特征向量(128维度xD)矩阵
"""
global SFIT, D
if SFIT is None:
SFIT = cv2.xfeatures2d.SIFT_create(D)
# 灰度化/sfit特征
gray = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)
kp, des = SFIT.detectAndCompute(gray, None)
# 保存特征图片
if image_save_path:
image = cv2.drawKeypoints(gray, kp, image)
cv2.imwrite(image_save_path, image)
return des
"""
一种新的具有局部不变性的特征 —— ORB特征，
从它的名字中可以看出它是对FAST特征点与BREIF特征描述子的一种结合与改进
"""
def orb(image, image_save_path=''):
"""
图片进行orb变换,生成特征点返回特征矩阵
:param image: cv2.imread
:param image_save_path: 如果指定了保存地址,则会将特征提取过的图片保存到制定的地址中
:return:orb变换后的特征向量(32维度xD)矩阵
"""
global ORB, D
if ORB is None:
ORB = cv2.ORB_create(D)
# 灰度化/sfit特征
gray = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)
kp, des = ORB.detectAndCompute(gray, None)
# 保存特征图片
if image_save_path:
image = cv2.drawKeypoints(gray, kp, image)
cv2.imwrite(image_save_path, image)
return des
"""
SURF
采用快速Hessian算法检测关键点+提取特征。
Surf在速度上比sift要快许多，这主要得益于它的积分图技术，
已经Hessian矩阵的利用减少了降采样过程，另外它得到的特征向量维数也比较少，
有利于更快的进行特征点匹配。
"""
def surf(image, image_save_path=''):
"""
图片进行surf变换,生成特征点返回特征矩阵
:param image: cv2.imread
:param image_save_path: 如果指定了保存地址,则会将特征提取过的图片保存到制定的地址中
:return:surf变换后的特征向量(64维度xn)矩阵
"""
global SURF
if SURF is None:
SURF = cv2.xfeatures2d.SURF_create()
# 灰度化/sfit特征
gray = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)
kp, des = SURF.detectAndCompute(gray, None)
# 保存特征图片
if image_save_path:
image = cv2.drawKeypoints(gray, kp, image)
cv2.imwrite(image_save_path, image)
return des
def drawMatchesKnn_cv2(img1_gray, kp1, img2_gray, kp2, goodMatch):
h1, w1 = img1_gray.shape[:2]
h2, w2 = img2_gray.shape[:2]
vis = np.zeros((max(h1, h2), w1 + w2, 3), np.uint8)
vis[:h1, :w1] = img1_gray
vis[:h2, w1:w1 + w2] = img2_gray
p1 = [kpp.queryIdx for kpp in goodMatch]
p2 = [kpp.trainIdx for kpp in goodMatch]
post1 = np.int32([kp1[pp].pt for pp in p1])
post2 = np.int32([kp2[pp].pt for pp in p2]) + (w1, 0)
for (x1, y1), (x2, y2) in zip(post1, post2):
cv2.line(vis, (x1, y1), (x2, y2), (0, 0, 255))
cv2.namedWindow("match", cv2.WINDOW_NORMAL)
cv2.imshow("match", vis)
def compare_pic(image1, image2, show=False):
"""
比较两张图片的特征
:param image1: cv2.imread
:param image2: cv2.imread
:param draw: 是否要显示匹配的结果
:return:匹配的特征点的个数
"""
global SFIT, D
if SFIT is None:
SFIT = cv2.xfeatures2d.SIFT_create(D)
# sift
kp1, des1 = SFIT.detectAndCompute(image1, None)
kp2, des2 = SFIT.detectAndCompute(image2, None)
# BFmatcher with default parms
bf = cv2.BFMatcher(cv2.NORM_L2)
matches = bf.knnMatch(des1, des2, k=2)
# 匹配点
goodMatch = []
for m, n in matches:
if m.distance < 0.50 * n.distance:
goodMatch.append(m)
# show
if show:
drawMatchesKnn_cv2(image1, kp1, image2, kp2, goodMatch)
cv2.waitKey(0)
cv2.destroyAllWindows()
return len(goodMatch)
if __name__ == '__main__':
import os
import time
class_name = 'data/normal/'
image_list = []
for file_name in os.listdir(class_name):
image_list.append(cv2.imread(class_name + file_name))
for i in xrange(len(image_list) - 1):
start = time.time()
print compare_pic(image_list[i], image_list[i + 1]),
print time.time() - start
# image = cv2.imread('test.jpg')
# print sfit(image)
# image = cv2.imread('test.jpg')
# print orb(image)
# image = cv2.imread('test.jpg')
# print surf(image)