将4个通道的图像转变为3个通道的图像

from PIL Import Image
import cv2
import numpy as np
#读取图片 
img = Image.open('./image/unchange.png')
img = np.array(img)
print(img.shape)

# 输出为 （1334,750,4）此图片有四个channels
# 如何将该图片转为三个channels

src = cv2.imread('./image/unchange.png')
cv2.imwrite('./image/change.png')

#最后再读取change.png这张图片
img1 = Image.open('./image/change.png')
img1 = np.array(img1)
print(img1.shape)

#输出为(1334,750,3)