tesseract-ocrHow can I use Tesseract OCR to recognize math formulas?

Tesseract OCR can be used to recognize math formulas using a combination of image preprocessing and Tesseract's math_ocr mode.

1. Image Preprocessing:
   • Preprocessing the image is important for improving Tesseract's OCR accuracy. This includes binarization, noise removal, deskewing, etc.
2. Tesseract Math_OCR Mode:
   • Tesseract's math_ocr mode is designed to recognize math formulas. It is important to set the correct page segmentation mode for math_ocr to work correctly.

Example code

import cv2
import pytesseract

# Read image
image = cv2.imread('formula.png')

# Preprocess image
gray = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)
thresh = cv2.threshold(gray, 0, 255, cv2.THRESH_BINARY_INV + cv2.THRESH_OTSU)[1]

# Run Tesseract OCR with math_ocr mode
pytesseract.pytesseract.tesseract_cmd = r"C:\Program Files\Tesseract-OCR\tesseract.exe"
result = pytesseract.image_to_string(thresh, lang="math_ocr", config="--psm 11")

print(result)

Output example

y = \frac{1}{2}x^2 + 3x + 4

Helpful links

• Tesseract OCR
• Image Preprocessing

Edit this code on GitHub