-
Notifications
You must be signed in to change notification settings - Fork 0
/
main.py
110 lines (82 loc) · 3.64 KB
/
main.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
# 处理光照不平衡问题
import numpy as np
import cv2 as cv
import sys
def localStd(img):
# 归一化
# img = img / 255.0
# 计算均值图像和均值图像的平方图像
img_blur = cv.blur(img, (21, 21))
reslut_1 = img_blur ** 2
# 计算图像的平方和平方后的均值
img_2 = img ** 2
reslut_2 = cv.blur(img_2, (21, 21))
reslut = np.sqrt(np.maximum(reslut_2 - reslut_1, 0))
return reslut
def get_reflect(img, img_illumination):
# get_img_illumination = get_illumination(img)
get_img_reflect = (img + 0.001) / (img_illumination + 0.001)
return get_img_reflect
def enhancement_reflect(img):
# 通过高斯滤波器
gaussian_blur_img = cv.GaussianBlur(img, (21, 21), 0)
enhancement_reflect_img = img * gaussian_blur_img
return enhancement_reflect_img
def get_enhancment_img(img_enhance_illumination, img_enahnce_reflect):
img = img_enhance_illumination * img_enahnce_reflect
img = img.astype('uint8')
return img
def read_img_from_disk(file_path):
# 0. 读取图像
img = cv.imread(file_path, cv.IMREAD_COLOR)
return img
def get_illumination(img):
return cv.GaussianBlur(img, (15, 15), 0)
"""
enhancment_illumination 增强反射分量，传入反射分量，返回增强后的反射分量
"""
def rgb2hsv(img_illumination):
img_hsv = cv.cvtColor(img_illumination, cv.COLOR_BGR2HSV)
return img_hsv
def v_of_hsv(img_hsv):
img_hsv = (img_hsv - np.min(img_hsv)) / (np.max(img_hsv) - np.min(img_hsv))
h, s, v = cv.split(img_hsv)
return v
def enhancment_illumination(img_illumination):
img_hsv = cv.cvtColor(img_illumination, cv.COLOR_BGR2HSV)
img_hsv = (img_hsv - np.min(img_hsv)) / (np.max(img_hsv) - np.min(img_hsv))
h, s, v = cv.split(img_hsv)
wsd = 5
gm = np.mean(v) / (1 + wsd * np.std(v)) # 一个数字
cst = localStd(v) # 300 * 400 的矩阵
lm = gm * v /(1 + wsd * cst) # 300 * 400 的矩阵
c = np.exp(gm) # 一个常数
wg = v ** 0.2 # 300 *400
wl = 1- wg
outM = v**c / (v**c +(wl * lm)**c + (wg * gm)**c + 0.001)
outM = 1.5 * outM - 0.5 * cv.GaussianBlur(outM, (21, 21), 0)
outM = (outM - np.min(outM))/(np.max(outM) - np.min(outM))
paramerter = 0.9
img_illumination[:, :, 0] = outM * (img_illumination[:, :, 0] / (v + 0.01))**paramerter
img_illumination[:, :, 1] = outM * (img_illumination[:, :, 1] / (v + 0.01))**paramerter
img_illumination[:, :, 2] = outM * (img_illumination[:, :, 2] / (v + 0.01))**paramerter
return img_illumination
if __name__ == '__main__':
file_name = "1.jpg"
img = read_img_from_disk(file_name) # 读取图像
img = cv.resize(img, (0, 0), fx=0.2, fy=0.2, interpolation=cv.INTER_NEAREST)
img_illumination = get_illumination(img) # 获得高频分量
img_reflect = get_reflect(img, img_illumination) # 获得反射分量
img_enhancement_reflect = enhancement_reflect(img_reflect) # 增强高频分量
img_enhancement_illumination = enhancment_illumination(img_illumination) # 增强低频分量
img_done = get_enhancment_img(img_enhancement_illumination, img_enhancement_reflect) # 照射分量与反射分量融合
cv.imshow("src", img)
cv.imshow("img_reflect", img_reflect)
cv.imshow("img_illumination", img_illumination)
cv.imshow("img_enhancement_reflect",img_enhancement_reflect)
cv.imshow("img_enhancement_illumination",img_enhancement_illumination)
cv.imshow("done", img_done)
cv.waitKey(0)
print("done")
cv.destroyAllWindows()
print("done for everthong")