Smart Object and Face Detection Assistant for Visually Impaired

The World Health Organization (WHO) estimates that 253 million people worldwide are blind or visually impaired. 217 million of them have moderate to severe visual impairment, leaving 36 million of them entirely blind. The majority of people who are visually impaired live in low- and middle-income nations, and there is evidence that vision impairment rises with age. Visually challenged people need access to a face and object detection system in order to improve their mobility and capacity for social interaction. The suggested remedy makes use of a Raspberry Pi equipped with a camera module to capture real-time video feeds, OpenCV for image processing, and machine learning techniques for object and face recognition. The proposed methodology includes YOLO Algorithm and Convolutional Neural Network Architecture in implementation. The system is designed to provide real-time feedback to the user via audio and haptic feedback to notify them of the presence of people or objects in their vicinity. The project also includes a user-friendly interface that allows the user to customize the detection settings to suit their individual needs. The system was tested on various scenarios and showed promising results in terms of accuracy and responsiveness. Overall, the proposed system provides a practical solution for visually impaired individuals to overcome their daily navigation challenges and to improve their quality of life. The results are obtained in both software and hardware implementation.