使用Python识别验证码：pytesseract.image_to_string()持续报错求助

Hey there, let's work through that pytesseract.image_to_string() error you're facing! Since your script can generate new images without issues, the problem is almost certainly tied to either pytesseract's underlying setup, how you're accessing the temporary image, or the image's quality for OCR. Here are the most common fixes to try:

1. Make sure the Tesseract OCR engine is properly installed and configured #

pytesseract is just a Python wrapper—it relies on the actual Tesseract OCR engine running under the hood. If the engine isn't installed, or pytesseract can't find its executable, you'll get an error every time you try to run image_to_string().

• For Windows: Download the official Tesseract installer, then explicitly set the path to the executable in your script:

  pytesseract.pytesseract.tesseract_cmd = r'C:\Program Files\Tesseract-OCR\tesseract.exe'
  

• For Linux: Install via your package manager:

  sudo apt install tesseract-ocr
  

  Then confirm the path with which tesseract and set it in your script if needed.
• For macOS: Use Homebrew:

  brew install tesseract
  

2. Verify your temporary image is accessible and intact #

Sometimes the issue isn't pytesseract at all—it's that PIL can't properly open the temporary image file.

• First, print out the filename variable to double-check the path is correct:

  print(f"Trying to open image at: {filename}")
  

• Manually navigate to that path and open the image to make sure it's not corrupted or empty.
• Add error handling to catch issues with the image file:

  try:
      with Image.open(filename) as img:
          img.verify()  # Checks if the image is corrupted
          # Reopen the image since verify() moves the file pointer to the end
          img = Image.open(filename)
          text = pytesseract.image_to_string(img)
  except Exception as e:
      print(f"Failed to open image: {str(e)}")
  

3. Add preprocessing to clean up the captcha image #

Captchas are designed to be hard for OCR—noise, lines, and distorted text can cause pytesseract to fail (or return garbage). Preprocessing will drastically improve results and avoid errors:

from PIL import Image, ImageFilter

# Open the image and convert to grayscale
img = Image.open(filename).convert('L')
# Apply binary thresholding to turn text black and background white
img = img.point(lambda x: 0 if x < 127 else 255)
# Remove small noise with a median filter
img = img.filter(ImageFilter.MedianFilter())

# Now try OCR again
text = pytesseract.image_to_string(img)

4. Update pytesseract and Pillow to compatible versions #

Outdated or mismatched versions of these libraries can cause unexpected errors. Upgrade to the latest stable releases:

pip install --upgrade pytesseract pillow

5. Capture the exact error message #

Since you didn't share the specific error you're getting, this is one of the most important steps. Wrap the problematic line in a try-except block to get detailed info:

import pytesseract
from PIL import Image

try:
    text = pytesseract.image_to_string(Image.open(filename))
except pytesseract.TesseractError as te:
    print(f"Tesseract-specific error: {te}")
except Exception as e:
    print(f"General error: {str(e)}")

The error message will tell you exactly what's wrong—whether it's a missing Tesseract executable, a corrupted image, or something else.

内容的提问来源于stack exchange，提问作者Rajesh Krishna