Abstract: One of the most vital senses in the human body is vision, which is also the sense that allows people to perceive their surroundings the best. As a result, thousands of articles on these topics have been published, proposing a range of computer vision services and products by creating new electronic assistive devices for the blind. This work tries to create a system that recovers the identification of nearby objects, a crucial function of the visual system. People who are visually challenged heavily rely on their other senses, such as touch and audio stimuli, to interpret their surroundings. It is quite challenging for a blind person to distinguish what is in front of them without touching it with their hands or any item like walking stick. Physical contact between a person and an object occasionally has the possibility to be fatal. This work employs a Neural Network for the recognition of pre-trained objects. A computer system that has the object identification Neural Network installed to carry out real-time object detection receives input from a camera that is oriented in accordance with the system's predetermined orientation.
Keywords: Computer Vision, Machine Learning, Neural Network, Visually Challenged
Copyright Form