Python使用OpenCV 卷积核 实现康威生命游戏

"Mozart, Beethoven, and Chopin never died. They simply became music."

康威生命游戏规则十分简单，简化后如下：

一个“细胞”（或者说单元）分为生或死两种状态，

如果活相邻细胞有2或3个活细胞 该细胞活，否则死

如果死细胞周围有3个活细胞，该细胞转为活。

再OpenCV中，将二值化图像数值变为0和1之后，使用该内核进行卷积核操作（个位储存相邻活细胞个数，十位储存中间细胞生或死状态）

threshed_value = (convolved/255).astype(np.uint8)
kernel = np.asarray([[1,1,1],
                     [1,10,1],
                     [1,1,1]],dtype='uint8')

需要注意的是进行卷积操作的时候需要设置边界类型，这里将图像边界作为0处理

接下来对图片进行几次掩膜操作，完成一个循环

    convolved = cv.filter2D(src=threshed_value, ddepth=cv.CV_8U,     
    kernel=kernel,borderType=cv.BORDER_CONSTANT)

    #populated:
    convolved[convolved==10] = 0    #单独死亡
    convolved[convolved==11] = 0 
    convolved[convolved>=14] = 0
    #unpopulated:
    convolved[convolved==3] = 255   #复活细胞
    #binarize again
    convolved[convolved>10] = 255    #转化为常值
    convolved[convolved<10] = 0
    #重新将图像数据转化成0和1
    ret,threshed_value = cv.threshold(convolved,10,1,cv.THRESH_BINARY)    

整个代码见下

import cv2 as cv
import numpy as np






image = cv.imread('John_H_Conway_2005_(cropped).jpg',0)
convolved = cv.adaptiveThreshold(image,255,cv.ADAPTIVE_THRESH_GAUSSIAN_C,\
            cv.THRESH_BINARY,11,5)

threshed_value = (convolved/255).astype(np.uint8)

kernel = np.asarray([[1,1,1],
                     [1,10,1],
                     [1,1,1]],dtype='uint8')

for name in ['convolved']:
    cv.namedWindow(name,cv.WINDOW_NORMAL)
    cv.imshow(name,eval(name))

key = cv.waitKey(0)

mode = True

while True:

    cv.imshow('convolved',convolved)
    key = cv.waitKey(10)
    if key == 27:
        break

    convolved = cv.filter2D(src=threshed_value, ddepth=cv.CV_8U, kernel=kernel,borderType=cv.BORDER_CONSTANT)

    #populated:
    convolved[convolved==10] = 0    #solitude
    convolved[convolved==11] = 0 
    convolved[convolved>=14] = 0
    #unpopulated:
    convolved[convolved==3] = 255   #revived
    #binarize again
    convolved[convolved>10] = 255
    convolved[convolved<10] = 0

    ret,threshed_value = cv.threshold(convolved,10,1,cv.THRESH_BINARY)