python_script/6.py at main · Robot-Kevin/python_script · GitHub

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
#coding=utf-8
################################################
# 数据增广，包括
# 2018.09.02 add
################################################
import numpy as np
import os
import cv2
import copy
import matplotlib as plt


class DataAugment:
    def __init__(self,debug=False):
        self.debug=debug
        print("Data augment...")

    def basic_matrix(self,translation):
        """基础变换矩阵"""
        return np.array([[1,0,translation[0]],[0,1,translation[1]],[0,0,1]])

    def adjust_transform_for_image(self,img,trans_matrix):
        """根据图像调整当前变换矩阵"""
        transform_matrix=copy.deepcopy(trans_matrix)
        height, width, channels = img.shape
        transform_matrix[0:2, 2] *= [width, height]
        center = np.array((0.5 * width, 0.5 * height))
        transform_matrix = np.linalg.multi_dot([self.basic_matrix(center), transform_matrix, self.basic_matrix(-center)])
        return transform_matrix

    def apply_transform(self,img,transform):
        """仿射变换"""
        # 输入图片、转换矩阵、输出图片大小、插值方式、边界像素模式、边界填充
        output = cv2.warpAffine(img, transform[:2, :], dsize=(img.shape[1], img.shape[0]),
                                flags=cv2.INTER_LINEAR, borderMode=cv2.BORDER_REFLECT, borderValue=0,)   #cv2.BORDER_REPLICATE,cv2.BORDER_TRANSPARENT
        return output

    def apply(self,img,trans_matrix):
        """应用变换"""
        tmp_matrix=self.adjust_transform_for_image(img, trans_matrix)
        out_img=self.apply_transform(img, tmp_matrix)
        if self.debug:
            self.show(out_img)
        return out_img

    def random_vector(self,min,max):
        """生成范围矩阵"""
        min=np.array(min)
        max=np.array(max)
        print(min.shape,max.shape)
        assert min.shape==max.shape
        assert len(min.shape) == 1
        return np.random.uniform(min, max)

    def show(self,img):
        """可视化"""
        cv2.imshow("outimg",img)
        cv2.waitKey()

    def random_transform(self,img,min_translation,max_translation):
        """平移变换"""
        factor=self.random_vector(min_translation,max_translation)
        trans_matrix=np.array([[1, 0, factor[0]],[0, 1, factor[1]],[0, 0, 1]])
        out_img=self.apply(img,trans_matrix)
        return trans_matrix, out_img

    def random_flip(self,img,factor):
        """水平或垂直翻转"""
        flip_matrix = np.array([[factor[0], 0, 0],[0, factor[1], 0],[0, 0, 1]])
        out_img=self.apply(img,flip_matrix)
        return flip_matrix, out_img

    def random_rotate(self,img,factor):
        """随机旋转"""
        angle=np.random.uniform(factor[0],factor[1])
        print("angle:{}".format(angle))
        rotate_matrix=np.array([[np.cos(angle), -np.sin(angle), 0],[np.sin(angle), np.cos(angle), 0],[0, 0, 1]])
        out_img=self.apply(img,rotate_matrix)
        return rotate_matrix, out_img

    def random_scale(self,img,min_translation,max_translation):
        """随机缩放"""
        factor=self.random_vector(min_translation, max_translation)
        scale_matrix = np.array([[factor[0], 0, 0],[0, factor[1], 0],[0, 0, 1]])
        out_img=self.apply(img,scale_matrix)
        return scale_matrix, out_img

    def random_shear(self,img,factor):
        """随机剪切，包括横向和众向剪切"""
        angle = np.random.uniform(factor[0], factor[1])
        print("fc:{}".format(angle))
        crop_matrix = np.array([[1, factor[0], 0], [factor[1], 1, 0], [0, 0, 1]])
        out_img=self.apply(img,crop_matrix)
        return crop_matrix, out_img


if __name__=="__main__":
    demo=DataAugment(debug=False)
    img=cv2.imread("/home/kerwin/python_script/python_code_template/Data_0226 (1)/Data/Final_State/fs_1.png")

    # 平移测试
    _,outimg=demo.random_transform(img,(0.1,0.1),(0.2,0.2))  #(-0.3,-0.3),(0.3,0.3)
    cv2.imwrite("/home/kerwin/python_script/python_code_template/transform.jpg",outimg)

#    # 垂直变换测试
    _, outimg =demo.random_flip(img,(1.0,-1.0))
    cv2.imwrite("/home/kerwin/python_script/python_code_template/random_flip.png",outimg)

    # 水平变换测试
    _, outimg =demo.random_flip(img, (-1.0, 1.0))
    cv2.imwrite("/home/kerwin/python_script/python_code_template/random_flip1.jpg",outimg)


#     # 旋转变换测试
    _, outimg =demo.random_rotate(img,(0.5,0.8))
    cv2.imwrite("/home/kerwin/python_script/python_code_template/random_rotate.jpg",outimg)

#     # # 缩放变换测试
    _, outimg =demo.random_scale(img,(1.2, 1.2),(1.3,1.3))
    cv2.imwrite("/home/kerwin/python_script/python_code_template/random_scale.jpg",outimg)

#     # 随机裁剪测试
    _, outimg =demo.random_shear(img,(0.2,0.3))
    cv2.imwrite("/home/kerwin/python_script/python_code_template/random_shear.jpg",outimg)

    # 组合变换
    t1,_=demo.random_transform(img,(-0.3,-0.3),(0.3,0.3))
    t2,_=demo.random_rotate(img,(0.5,0.8))
    t3,_=demo.random_scale(img,(1.5,1.5),(1.7,1.7))
    tmp=np.linalg.multi_dot([t1,t2,t3])
    print("tmp:{}".format(tmp))
    out=demo.apply(img,tmp)