CAPture: A Vision Assistive Cap for People with Visual Impairment

Abstract

With recent technological developments, the field of assistive technology has advanced with enormous opportunity for the development of more efficient assistive devices.While various assistive devicesexist, their enhancement is still in progress. The blind or visually impaired people can use these devices according to their ability and needs. Some of the existing devices for visually handicapped lack advanced features and efficient functionalities.Thus,there is great need for more dynamic features. With the help of technologies like IoT, Machine learning and Computer Vision we propose a prototype of anassistive vision enhancing wearable cap-based system called CAPture for the visually impaired. The system is aimed to be low cost, and will act as the guiding agent for the user to know about the surroundings with precision and ease both in indoor and outdoor environments. The proposed system is designed to be operated in two different modes, i.e., battery saver and normal mode. In the battery saver mode, it can detect the objects with buzzer-based output, whereas in the normal mode it is designed to support enhanced features like object detection, family member detection, social helper detection like police, doctor etc, and scenery visualization using voice-based outputs. The various advanced features are developed using some pre-trained and newly trained deep learning and face recognition models, text to speech module along with buzzer, ultrasonic sensors, camera, Raspberry Pi and Arduino boards with an approximate cost of Rs 5000-6000.