Handwritten Word Recognition

Teias Sanap, Saloni Gaikwad, Ishani Tambat, Priya Makarand Shelke, Riddhi Mirajkar, Suruchi Gaurav Dedgaonkar · 2023

Humans are the only species with the ability to recognize anything. The hand transcription is simple for humans to recognize. We have created a Convolutional Neural Network model that accurately converts handwritten words into digital text, aiming to simplify our lives. Despite the challenges posed by variations in patterns across different languages, our system employs advanced techniques such as image processing, feature extraction, and classification to enhance accuracy. Through extensive training, the system learns to recognize similarities and differences among various handwritten samples, improving its ability to identify and interpret text accurately. Our CNN-based approach holds great promise for streamlining tasks that rely on converting handwritten words into digital text.

Read the paper · More papers on PaperTik