服务器之家

服务器之家 > 正文

Python实现遗传算法(虚拟机中运行)

时间:2022-03-07 00:24     来源/作者:灼灼华

(一)问题

遗传算法求解正方形拼图游戏

(二)代码

?
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
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
#!/usr/bin/env python
# -*- coding: utf-8 -*-
 
from PIL import Image, ImageDraw
import os
import gc
import random as r
import minpy.numpy as np
 
class Color(object):
    '''
    定义颜色的类,这个类包含r,g,b,a表示颜色属性
    '''
    def __init__(self):
        self.r = r.randint(0, 255)
        self.g = r.randint(0, 255)
        self.b = r.randint(0, 255)
        self.a = r.randint(95, 115)
 
 
def mutate_or_not(rate):
    '''
    生成随机数,判断是否需要变异
    '''
    return True if rate > r.random() else False
 
 
class Triangle(object):
    '''
    定义三角形的类
    属性:
            ax,ay,bx,by,cx,cy:表示每个三角形三个顶点的坐标
            color            : 表示三角形的颜色
            img_t            : 三角形绘制成的图,用于合成图片
    方法:
            mutate_from(self, parent):      从父代三角形变异
            draw_it(self, size=(256, 256)): 绘制三角形
    '''
 
 
    max_mutate_rate = 0.08
    mid_mutate_rate = 0.3
    min_mutate_rate = 0.8
 
 
    def __init__(self, size=(255, 255)):
        t = r.randint(0, size[0])
        self.ax = r.randint(0, size[0])
        self.ay = r.randint(0, size[1])
        self.bx = self.ax+t
        self.by = self.ay
        self.cx = self.ax+t
        self.cy = self.ay-t
        self.dx = self.ax
        self.dy = self.ay-t
        self.color = Color()
        self.img_t = None
 
 
    def mutate_from(self, parent):
        if mutate_or_not(self.max_mutate_rate):
            t = r.randint(0, 255)
            self.ax = r.randint(0, 255)
            self.ay = r.randint(0, 255)
            self.bx = self.ax + t
            self.by = self.ay
            self.dx = self.ax
            self.dy = self.ay - t
            self.cx = self.ax + t
            self.cy = self.ay - t
        if mutate_or_not(self.mid_mutate_rate):
            t = min(max(0, parent.ax + r.randint(-15, 15)), 255)
            self.ax = min(max(0, parent.ax + r.randint(-15, 15)), 255)
            self.ay = min(max(0, parent.ay + r.randint(-15, 15)), 255)
            self.bx = self.ax + t
            self.by = self.ay
            self.dx = self.ax
            self.dy = self.ay - t
            self.cx = self.ax + t
            self.cy = self.ay - t
        if mutate_or_not(self.min_mutate_rate):
            t = min(max(0, parent.ax + r.randint(-3, 3)), 255)
            self.ax = min(max(0, parent.ax + r.randint(-3, 3)), 255)
            self.ay = min(max(0, parent.ay + r.randint(-3, 3)), 255)
            self.bx = self.ax + t
            self.by = self.ay
            self.dx = self.ax
            self.dy = self.ay - t
            self.cx = self.ax + t
            self.cy = self.ay - t
 
 
                # color
        if mutate_or_not(self.max_mutate_rate):
            self.color.r = r.randint(0, 255)
        if mutate_or_not(self.mid_mutate_rate):
            self.color.r = min(max(0, parent.color.r + r.randint(-30, 30)), 255)
        if mutate_or_not(self.min_mutate_rate):
            self.color.r = min(max(0, parent.color.r + r.randint(-10, 10)), 255)
 
        if mutate_or_not(self.max_mutate_rate):
            self.color.g = r.randint(0, 255)
        if mutate_or_not(self.mid_mutate_rate):
            self.color.g = min(max(0, parent.color.g + r.randint(-30, 30)), 255)
        if mutate_or_not(self.min_mutate_rate):
            self.color.g = min(max(0, parent.color.g + r.randint(-10, 10)), 255)
 
        if mutate_or_not(self.max_mutate_rate):
            self.color.b = r.randint(0, 255)
        if mutate_or_not(self.mid_mutate_rate):
            self.color.b = min(max(0, parent.color.b + r.randint(-30, 30)), 255)
        if mutate_or_not(self.min_mutate_rate):
            self.color.b = min(max(0, parent.color.b + r.randint(-10, 10)), 255)
        # alpha
        if mutate_or_not(self.mid_mutate_rate):
            self.color.a = r.randint(95, 115)
        # if mutate_or_not(self.mid_mutate_rate):
        #     self.color.a = min(max(0, parent.color.a + r.randint(-30, 30)), 255)
        # if mutate_or_not(self.min_mutate_rate):
        #     self.color.a = min(max(0, parent.color.a + r.randint(-10, 10)), 255)
 
 
    def draw_it(self, size=(256, 256)):
        self.img_t = Image.new('RGBA', size)
        draw = ImageDraw.Draw(self.img_t)
        draw.polygon([(self.ax, self.ay),
                      (self.bx, self.by),
                      (self.cx, self.cy),
                      (self.dx, self.dy)],
                     fill=(self.color.r, self.color.g, self.color.b, self.color.a))
        return self.img_t
 
 
class Canvas(object):
    '''
    定义每一张图片的类
    属性:
            mutate_rate  : 变异概率
            size         : 图片大小
            target_pixels: 目标图片像素值
    方法:
            add_triangles(self, num=1)      : 在图片类中生成num个三角形
            mutate_from_parent(self, parent): 从父代图片对象进行变异
            calc_match_rate(self)           : 计算环境适应度
            draw_it(self, i)                : 保存图片
    '''
 
 
    mutate_rate = 0.01
    size = (256, 256)
    target_pixels = []
 
 
    def __init__(self):
        self.triangles = []
        self.match_rate = 0
        self.img = None
 
 
    def add_triangles(self, num=1):
        for i in range(0, num):
            triangle = Triangle()
            self.triangles.append(triangle)
 
 
    def mutate_from_parent(self, parent):
        flag = False
        for triangle in parent.triangles:
            t = triangle
            if mutate_or_not(self.mutate_rate):
                flag = True
                a = Triangle()
                a.mutate_from(t)
                self.triangles.append(a)
                continue
            self.triangles.append(t)
        if not flag:
            self.triangles.pop()
            t = parent.triangles[r.randint(0, len(parent.triangles) - 1)]
            a = Triangle()
            a.mutate_from(t)
            self.triangles.append(a)
 
 
    def calc_match_rate(self):
        if self.match_rate > 0:
            return self.match_rate
        self.match_rate = 0
        self.img = Image.new('RGBA', self.size)
        draw = ImageDraw.Draw(self.img)
        draw.polygon([(0, 0), (0, 255), (255, 255), (255, 0)], fill=(255, 255, 255, 255))
        for triangle in self.triangles:
            self.img = Image.alpha_composite(self.img, triangle.img_t or triangle.draw_it(self.size))   
        # 与下方代码功能相同,此版本便于理解但效率低
        # pixels = [self.img.getpixel((x, y)) for x in range(0, self.size[0], 2) for y in range(0, self.size[1], 2)]
        # for i in range(0, min(len(pixels), len(self.target_pixels))):
        #     delta_red   = pixels[i][0] - self.target_pixels[i][0]
        #     delta_green = pixels[i][1] - self.target_pixels[i][1]
        #     delta_blue  = pixels[i][2] - self.target_pixels[i][2]
        #     self.match_rate += delta_red   * delta_red   + \
        #                        delta_green * delta_green + \
        #                        delta_blue  * delta_blue
        arrs = [np.array(x) for x in list(self.img.split())]    # 分解为RGBA四通道
        for i in range(3):                                      # 对RGB通道三个矩阵分别与目标图片相应通道作差取平方加和评估相似度
            self.match_rate += np.sum(np.square(arrs[i]-self.target_pixels[i]))[0]
 
    def draw_it(self, i):
        #self.img.save(os.path.join(PATH, "%s_%d_%d_%d.png" % (PREFIX, len(self.triangles), i, self.match_rate)))
        self.img.save(os.path.join(PATH, "%d.png" % (i)))
 
 
def main():
        global LOOP, PREFIX, PATH, TARGET, TRIANGLE_NUM
        # 声明全局变量
        img = Image.open(TARGET).resize((256, 256)).convert('RGBA')
        size = (256, 256)
        Canvas.target_pixels = [np.array(x) for x in list(img.split())]
        # 生成一系列的图片作为父本,选择其中最好的一个进行遗传
        parentList = []
        for i in range(20):
            print('正在生成第%d个初代个体' % (i))
            parentList.append(Canvas())
            parentList[i].add_triangles(TRIANGLE_NUM)
            parentList[i].calc_match_rate()
        parent = sorted(parentList, key=lambda x: x.match_rate)[0]
        del parentList
        gc.collect()
        # 进入遗传算法的循环
        i = 0
        while i < 30000:
            childList = []
            # 每一代从父代中变异出10个个体
            for j in range(10):
                childList.append(Canvas())
                childList[j].mutate_from_parent(parent)
                childList[j].calc_match_rate()
            child = sorted(childList, key=lambda x: x.match_rate)[0]
            # 选择其中适应度最好的一个个体
            del childList
            gc.collect()
            parent.calc_match_rate()
            if i % LOOP == 0:
                print ('%10d parent rate %11d \t child1 rate %11d' % (i, parent.match_rate, child.match_rate))
            parent = parent if parent.match_rate < child.match_rate else child
            # 如果子代比父代更适应环境,那么子代成为新的父代
            # 否则保持原样
            child = None
            if i % LOOP == 0:
                # 每隔LOOP代保存一次图片
                parent.draw_it(i)
                #print(parent.match_rate)
                #print ('%10d parent rate %11d \t child1 rate %11d' % (i, parent.match_rate, child.match_rate))
            i += 1
 
 
'''
定义全局变量,获取待处理的图片名
'''
NAME = input('请输入原图片文件名:')
LOOP = 100
PREFIX = NAME.split('/')[-1].split('.')[0# 取文件名
PATH = os.path.abspath('.'# 取当前路径
PATH = os.path.join(PATH,'results')
TARGET = NAME  # 源图片文件名
TRIANGLE_NUM = 256  # 三角形个数
 
if __name__ == '__main__':
    #print('开始进行遗传算法')
    main()

(三)运行结果

Python实现遗传算法(虚拟机中运行)

(四)结果描述

  代码是在遗传算法求解三角形火狐拼图改进而来,遗传算法求解正方形拼图游戏只需随机生成一个坐标和一个常数值(作为正方形的边长),通过正方形的性质,可以写出正方形其他三个点的坐标,确定了四个点的坐标之后,进行遗传和变异。

到此这篇关于Python实现遗传算法(虚拟机中运行)的文章就介绍到这了,更多相关Python 遗传算法内容请搜索服务器之家以前的文章或继续浏览下面的相关文章希望大家以后多多支持服务器之家!

原文链接:https://blog.csdn.net/qq_52137561/article/details/121492555

标签:

相关文章

热门资讯

2022年最旺的微信头像大全 微信头像2022年最新版图片
2022年最旺的微信头像大全 微信头像2022年最新版图片 2022-01-10
蜘蛛侠3英雄无归3正片免费播放 蜘蛛侠3在线观看免费高清完整
蜘蛛侠3英雄无归3正片免费播放 蜘蛛侠3在线观看免费高清完整 2021-08-24
背刺什么意思 网络词语背刺是什么梗
背刺什么意思 网络词语背刺是什么梗 2020-05-22
yue是什么意思 网络流行语yue了是什么梗
yue是什么意思 网络流行语yue了是什么梗 2020-10-11
2020微信伤感网名听哭了 让对方看到心疼的伤感网名大全
2020微信伤感网名听哭了 让对方看到心疼的伤感网名大全 2019-12-26
返回顶部