围绕点创建方形缓冲区 - 网格数据 - python

Question

I have a dataset for which I would like to create additional training labels, by creating a buffer zone around the true labels in a two-dimensional dataset (lon, lat). For the sake of my question, say that my dataset looks like:

array([[0, 0, 0, 0, 0],
       [0, 0, 0, 0, 0],
       [0, 0, 1, 0, 0],
       [0, 0, 0, 0, 0],
       [0, 0, 0, 0, 0]])

code: df = np.array([0,0,0,0,0,0,0,0,0,0,0,0,1,0,0,0,0,0,0,0,0,0,0,0,0]).reshape(5,5)

After creating the buffer zone. My output data should look something like:

array([[0, 0, 0, 0, 0],
       [0, 1, 1, 1, 0],
       [0, 1, 1, 1, 0],
       [0, 1, 1, 1, 0],
       [0, 0, 0, 0, 0]])

Technically my dataset is 3D with 5000 time variables. I know in ArcGIS there is a tool that does this. However, it only does this for one time at a time. I don't want to export 5000 separate files, as you could understand. Does anyone know how to tackle this issue?

Maybe good to know that all my one 'pixel' is 0.5 by 0.5.

Answer 1

Although it might not be the prettiest of answers. I did find a way on how to tackle it. The code below creates (if possible) a 3x3 grid of true labels around each true label in the original dataset. It handles borders and edges as well without a problem. If anyone knows better / faster solution, please share!

df = np.array([0,0,0,0,0,0,0,0,0,0,0,0,1,0,0,0,0,0,0,0,0,0,0,0,0]).reshape(5,5)

values = np.where(df == 1) # find indexes of true labels
    for ind in range(len(values[0])):
        x = values[0][ind]
        y = values[1][ind]
        #upper left
        if x == 0 and y == 0:
            df[x,y:y+2] = 1
            df[x+1,y:y+2] = 1 
        #upper right
        elif x == 0 and y == df.shape[1]-1:
            df[x,y-1:y+1] = 1
            df[x+1,y-1:y+1] = 1
        #bottom left
        elif x == df.shape[0]-1 and y == 0:
            df[x-1,y:y+2] = 1
            df[x,y:y+2] = 1
        #bottom right
        elif x == df.shape[0]-1 and y == df.shape[1]-1:
            df[x-1,y-1:y+1] = 1
            df[x,y-1:y+1] = 1
        ### along borders
        #along top border
        elif x == 0 and y < df.shape[1]-1:
            df[x,y-1:y+2] = 1
            df[x+1,y-1:y+2] = 1
        #along bottom border
        elif x == df.shape[0]-1 and y < df.shape[1]-1:
            df[x-1,y-1:y+2] = 1
            df[x,y-1:y+2] = 1
        #along left border
        elif x < df.shape[0]-1 and y == 0:
            df[x-1,y:y+2] = 1  
            df[x,y:y+2] = 1
            df[x+1,y:y+2] = 1
        #along right border
        elif x < df.shape[0]-1 and y == df.shape[0]-1:
            df[x-1,y-1:y+1] = 1  
            df[x,y-1:y+1] = 1
            df[x+1,y-1:y+1] = 1
        ### everywhere aside along borders
        else:
            df[x-1,y-1:y+2] = 1
            df[x,y-1:y+2] = 1
            df[x+1,y-1:y+2] = 1