力扣——835. 图像重叠

题目描述

思路：
1.图像填充 填充后的图像，即为img1可以平移的所有情况
2.卷积

class Solution(object):
    def largestOverlap(self, img1, img2):
        """
        :type img1: List[List[int]]
        :type img2: List[List[int]]
        :rtype: int
        """
        n=len(img1)
        #填充图像，使得图像各种平移变换都涵盖到新的大图像中
        def filling(img):
            m=2*(n-1)+n
            image=[[0]*m for _ in range(m)]
            for i in range(n):
                for j in range(n):
                    if img[i][j]==1:
                        new_i,new_j=i+n-1,j+n-1
                        image[new_i][new_j]=1
            return image
        image=filling(img1)
        m=len(image)
        #把img2作为卷积核，image作为待卷积图像，计算过程中出现的最大重合数

        #卷积核中为1 得位置记录下来
        one_position = []
        for dx in range(n):
            for dy in range(n):
                if img2[dx][dy] == 1:
                    one_position.append([dx, dy])
        res = 0
        for start_i in range(m- n + 1):
            for start_j in range(m- n + 1):
                tmp = 0
                for dx, dy in one_position:
                    tmp += image[start_i + dx][start_j + dy]
                res = max(res, tmp)
                if res==len(one_position):return res
        return res

调包

from scipy.signal import correlate2d

class Solution:
    def largestOverlap(self, A, B):
        return correlate2d(A, B).max()

官网方法：link