/text_extraction_from_image

Primary LanguageJupyter NotebookMIT LicenseMIT

text_extraction_from_image

In this project, I used Tesseract and OpenCV to create a program that extracts text from images. The steps which were taken in developing this program were as follows:

  • Load the image, resize it and then save it.
  • Use the PyTesseract Library to extract the text from the image.
  • Process the text extracted from the image.
  • Used Computer Vision to perform further processing of complex images.
  • Remove noise from the image using the blur function.
  • Perform threshold transformation of the image.
  • Use the erode function of cv2 on the image.
  • Performed other image processing operations.
  • Draw a rectangle around a character/word/pattern.

Results

Original Image:

Screen Shot 2021-12-29 at 3 45 26 AM

Text Extracted Using PyTesseract:

Screen Shot 2021-12-29 at 3 44 04 AM

The results were not quite accurate as it failed to get to 54170 to join our text club.

Rectangle Drawn Around Text After Performing Computer Vision operations:

Screen Shot 2021-12-29 at 3 48 48 AM

After performing Computer Vision Operations using OpenCV, the results were accurate as a rectangle was drawn around the text.