Currency Detection for Blind People Using Image-Processing

Abstract

This research proposes a novel system for assisting visually impaired individuals in identifying Indian currency notes using a webcam-based scanning approach. The aim is to enhance the independence and accessibility of blind users by providing real-time currency recognition through a computer vision system.The system utilizes computer vision algorithms to process live video feed from a webcam, identifying and extracting relevant features from Indian currency notes. The key features considered include color, size, pattern, and denomination-specific characteristics. Machine learning models are employed for robust classification and recognition of various currency denominations. To ensure real-time functionality, the system is designed to operate on a standard personal computer or laptop, making it easily accessible for a broad user base. The user interface is developed with simplicity and user-friendliness in mind, providing auditory or tactile feedback to convey the detected currency denomination. Extensive testing is conducted with a diverse set of Indian currency notes, considering variations in wear and tear, lighting conditions, and viewing angles. The system's accuracy, speed, and reliability are evaluated to ensure its practical utility for visually impaired users in real-world scenarios.