python在没有opencv库的情况下，使用二进制形态学概念实现侵蚀功能和膨胀功能。

如题所述，没找到理想答案，打了一个python的模板，请大佬参考一下

import cv2
import numpy as np


input = np.array([[1,0,0,0,0],
                  [1,0,0,0,1],
                  [0,1,0,1,0],
                  [0,0,1,0,0],
                  [0,0,1,1,0]],
                 np.float32)

se = np.array([[0,1,0],
               [1,1,1],
               [0,1,0]],
             np.uint8)


def myDilate(input, se):
    output = np.zeros(input.shape)
    for i, row in enumerate(input):
        for j, one in enumerate(row):
            # your code
            output[i][j] = input[i][j]

    return output

def myErode(input, se):
    output = np.zeros(input.shape)
    for i, row in enumerate(input):
        for j, one in enumerate(row):
            # your code
            output[i][j] = input[i][j]
    return output

output = myDilate(input, se)
cv_output = cv2.dilate(input, se, iterations=1)

print(output)
print(cv_output)

if(np.array_equal(output, cv_output)):
    print("Pass")

https://blog.csdn.net/qq_42451251/article/details/107901746