Extracting Meaningful Data from Education Credentials Using OCR and Image Processing

To apply for higher education and job opportunities, a student's marksheet serves as a reference document. The conventional way of manually extracting meaningful information for companies and colleges is timeconsuming, error-prone, and non-automatic. Optical Character Recognition (OCR) helps to facilitate the procedure. The authors have employed OpenCV template matching method using Python, noise removal using the custom autoencoder model, and OCR to provide solutions. The authors have also compared three OCR models using Python: Tesseract OCR, Keras OCR, and Easy OCR. The conclusions are based on observations made about error rate and effectiveness in handling noisy data. In recruiting process, organizations need to extract data from education credentials. The proposed method offers a streamlined and effective option that can help automate the process and enhance accuracy.