matplotlib.imread() 返回什么？ 如何迭代过滤特定像素？

Question

When I use matplotlib.imread(), it returns a 3D array of shape (497, 1248, 3)

What values are stored in this array? is it the RGB values of the pixels in the photo?

img = mpimg.imread("./pearlite_strain2.jpg")

I want to iterate through this array, and save the distance of all pixels that are a certain RGB value, from the left most side of the photo.

how would I do that?

Answer 1

The values in the array are RGB values (y, x, color) . The array is a standard numpy array, so you can use argwhere and argmax to get the required information ( argmax finds first occurrence of 1 or True in the boolean array of 0 and 1 or False and True ). You desired distances are the x coordinates:

import numpy as np
import matplotlib.pyplot as plt

np.random.seed(0)
img = np.random.randint(0, 200, (2,4,3), np.uint8)
searched = [245, 250, 255]      # almost white bright pixels
img[0,2,:] = img[1,1,:] = img[1,3,:] = searched

plt.imshow(img)

all_distances = np.argwhere((img[...,:]==searched).all(2))[:,1]
# array([2, 1, 3], dtype=int64)

first_distances = np.argmax((img[...,:]==searched).all(2), axis=1)
# array([2, 1], dtype=int64)