Product identification and text recognition for blind peoples using deep learning algorithm
Mr. NOOR AHAMED J1 M.C.A., M.Phil., SARAVANA KUMAR.S2
1Assistant Professor (SG), Department of Computer Applications, Nehru college of management, Coimbatore, Tamil Nadu, India.
jnamca@gmail.com
2II MCA, Department of Computer Applications, Nehru college of management, Coimbatore, Tamil Nadu, India.
ABSTRACT:
Blind and visually impaired individuals face significant challenges in identifying products and reading text in everyday life. This project proposes a deep learning-based system that enables product identification and text recognition to assist blind people in their daily activities. The system utilizes Convolutional Neural Networks (CNNs) and Optical Character Recognition (OCR) techniques to recognize products and extract text from labels, packaging, and documents. A pre-trained CNN model (such as MobileNet, ResNet, or EfficientNet) is employed for product classification, while Tesseract OCR or deep learning-based EAST (Efficient and Accurate Scene Text Detection) is used for text recognition. The system can be integrated into a mobile application or a wearable device equipped with a camera to capture images of products and text .Once the product is identified or text is extracted, the system converts the information into speech output using Text-to-Speech (TTS) technology, enabling blind users to receive real-time audio feedback. The model is trained on a diverse dataset of product images and text samples to enhance accuracy in different environments, including supermarkets, kitchens, and workspaces. This deep learning-powered
assistive technology aims to enhance independence and accessibility for visually impaired individuals, reducing their reliance on external help for product identification and text reading. The proposed system provides a fast, efficient, and user-friendly solution to improve the quality of life for blind people.
KEYWORDS:
Deep Learning, Product Identification, Text Recognition, Blind Assistance, Convolutional Neural Networks, OCR, Text- to-Speech
