CURRENCY DETECTION AND IMAGE TO TEXT CONVERSION

Abstract

The main goal of this project is to create and put into place an integrated system that solves the issues found with OCR and currency classification. With the seamless integration of advanced currency identification and OCR features, the suggested solution seeks to improve efficiency, accuracy, and user experience. The project aims to offer a strong solution for automated financial transactions and document processing through adaptive learning, real-time processing, and security measures. System obsolescence and reduced efficacy may arise from an inability to scale to meet increasing user needs and adjust to changing currency designs. An OCR system reads an image of printed or handwritten characters and transforms it into an editable text document. First, each line is isolated, and then individual characters with spaces are added to break the text picture into sections. Following the extraction of characters, the text image's texture, and topological characteristics, such as corner points, features of distinct regions, character area ratio, and convex area of every character, is computed. Every letter, number, and symbol, both capital and lowercase, has previously had its features saved as a template. The technique uses features matching between the extracted character and the template of all characters as a measure of similarity to identify the precise character based on the texture and topological aspects.