这里我的目标是合并那些元素与其他数组的距离小于3个像素的数组。下面是我使用的代码：

import numpy as np, matplotlib.pyplot as plt
from scipy.signal import convolve2d
from scipy.ndimage import label 

## Example matrix

arr = np.array([[0,0,0,0,0,0,0,0,0,0,0,0],
                [0,0,1,0,0,0,0,0,0,0,0,0],
                [0,0,1,0,0,0,0,0,0,0,0,0],
                [0,0,0,1,0,0,0,0,0,0,0,0],
                [0,0,1,0,0,0,0,1,0,0,0,0],
                [0,0,0,0,0,0,0,0,1,0,0,0],
                [0,0,1,0,0,0,0,0,0,1,0,0],
                [0,0,1,0,0,0,0,0,1,1,0,0],
                [0,0,0,1,0,0,0,0,1,0,0,0],
                [0,0,0,1,0,0,0,0,0,0,0,0],
                [0,0,0,0,0,0,0,0,0,0,0,0],
                [0,0,0,0,0,0,0,0,0,0,0,0],
                [0,0,0,0,0,0,0,0,0,0,0,0]])

## Check if neighboring cells are zero using a 2d convolution:

# The kernel

ker = np.array([[1,1,1],
                [1,0,1],
                [1,1,1]])
ker2= np.array([[1,1,1],
                [1,1,1],
                [1,1,1]])
res = convolve2d(arr,ker,mode='same')

# Eliminate the noise

res = np.int32((res>0)&(arr>0))

# Using label your can segment an image and since an image is basically a matrix:

lab = label(res, ker2)
print(lab)

# Get the coordinate of each line.

coord = []
for ii in range(lab[1]):
    coord.append(np.vstack(np.where(lab[0]==(ii+1))))

for i, e in enumerate(coord):
    xVals = e[1]
    yVals = e[0]
    plt.scatter(xVals, yVals)

plt.xlim(0,len(arr[0]))
plt.ylim(0,len(arr))
plt.yticks(np.linspace(0,len(arr), len(arr)+1))
plt.grid()
plt.show()

这将返回：

请注意，蓝色、绿色和橙子数据点（像素）是独立的数组。我想合并蓝色和绿色数组（相距1个像素），而保留橙色数组（相距〉2个像素）。这里分配给变量“arr”的数组只是示例数据。真实的数据将来自一个大的图像文件，因此该问题的解决方案必须足够有效，以便在更大的范围内实用。
谢谢你的帮助。