计算机视觉的卷积及Dr的代码，上课没有听懂，需要超级详细讲解

问题遇到的现象和发生背景

Practical task 1: Developing a correlation filter.
Developing a correlation filter.

It is necessary to develop a function - a filter based on correlation. without using ready-made implementations and using only numpy library functions.

It is necessary to use the following types of kernels for filters with normalization:

1. 1 1 1 1 1 2) -1 -1 -1 -1 -1 3) -1 -1 -1 -1 -1

   1 1 1 1 1 0 0 0 0 0 0 0 1 0 0

   1 1 1 1 1 1 1 1 1 1 1 1 1 1 1

For test Image please use picture boat1_resize.jpg.

Plese, Save your Result as the processed image and code to the report file and subscribe it to answer of this task

问题相关代码，请勿粘贴截图

import numpy as np
import cv2 as cv
img=cv.imread("D:/boat1_resize.jpg",0)
cv.imshow("1",img)
cv.waitKey(0)
kernel1=np.array([[1,1,1],                   
                   [0,0,0],
                   [-1,-1,-1]])
img=cv.copyMakeBorder(img,1,1,2,2,cv.BORDER_REPLICATE)
img2=img.copy()
for i in range(0,321,1):
    for j in range(0,492,1):
        a=np.abs(np.sum(np.multiply(img[i:i+3,j:j+3],kernel1)))
        img2[i+1,j+2]=a
img=img[1:323,2:495]
cv.imshow("1",img2)
cv.waitKey(0)
cv.destroyAllWindows()

运行结果及报错内容

我的解答思路和尝试过的方法

我想要达到的结果