无法将网络图片转换为NumPy数组，请求技术协助

I've run into this exact error before! Let's break down why it's happening and how to fix it:

Why This Error Occurs #

Most often, this happens because the PIL Image object you're working with is in a palette mode (like 'P') instead of a direct pixel mode (like 'RGB'). Numpy can't directly interpret palette-mapped image objects into integer arrays, hence the int() argument must be a string... not 'JpegImageFile' error. It can also pop up if you're using an outdated version of PIL (instead of the maintained Pillow fork).

Solution 1: Convert the Image to RGB Mode First #

This is the simplest fix for palette-mode images. Converting to RGB ensures numpy can read the raw pixel values directly:

from PIL import Image
from io import BytesIO
import requests
import numpy as np

url = r'https://drive.google.com/uc?id=1xrKyFZAIIR_XVhlhYGHlfLvIr5vQ5EnW'
response = requests.get(url)
# Convert to RGB mode to avoid palette issues
img = Image.open(BytesIO(response.content)).convert('RGB')
arr = np.array(img, dtype=np.uint8)

Solution 2: Extract Pixel Data with getdata() and Reshape #

If the first solution doesn't work, you can manually extract the pixel data and reshape it to match the image dimensions:

from PIL import Image
from io import BytesIO
import requests
import numpy as np

url = r'https://drive.google.com/uc?id=1xrKyFZAIIR_XVhlhYGHlfLvIr5vQ5EnW'
response = requests.get(url)
img = Image.open(BytesIO(response.content))
# Get raw pixel data and reshape to (height, width, channels)
arr = np.array(img.getdata(), dtype=np.uint8).reshape(img.height, img.width, len(img.getbands()))

Quick Check: Verify Your Image Mode #

To confirm the root cause, add a print statement to check the image's mode:

print(img.mode)

If the output is 'P', Solution 1 will definitely resolve the error.

Bonus: Update Pillow to the Latest Version #

Old versions of PIL (not Pillow) have compatibility issues with numpy. Make sure you're using the maintained Pillow library:

pip uninstall PIL -y
pip install --upgrade pillow

内容的提问来源于stack exchange，提问作者John