VisionGuide is technology designed to assist visually blind people to be more independent and secure. The smart cap on the head with the camera attached captures real-time visual data, meanwhile there is a remote control with eight buttons, associated with the fulfillment of a variety of functions. With such a system, the visually impaired are given auditory information regarding the surroundings, objects, texts, recognition of known faces, and emotions as well. The latest machine learning concepts of VisionGuide are applied learning algorithms such as CNN for object detection, FaceNet for facial recognition, and Optical Character Recognition (OCR) for reading printed text. It captures visual data from the camera of the cap and gives audio feedback of the processed information through a text-to-speech engine. For navigation, the system uses geolocation services and routes planning to provide voice-guided directions in improving mobility. The major modules of the project are divided into seven: emergency notification, object detection, face recognition, emotion recognition, text reading, caption generation, and navigation. The emergency notification module generates messages with the user’s present location. Other modules help users interact with their environment in a hands-free mode. This solution enhances the independence, mobility, and information access of the visually impaired by addressing more than one challenge in an integrated, user-friendly system. Being an integrated functionality within one device, VisionGuide provides a pragmatic it is thus an efficient tool that helps in improving the quality of life of the visually disabled persons, accessible, convenient, and safe to use for everyday activity.

错误:搜索内容不能为空,请输入英文关键词
错误:关键词超出字数限制,请精简
高级检索

VisionGuide: Empowering Independence for the Visually Impaired

  • Rashi Channawar,
  • Vishwajeet Kushwaha,
  • Shiva Panwar,
  • Chiranjeevisingh Bais,
  • Aditya Turankar

摘要

VisionGuide is technology designed to assist visually blind people to be more independent and secure. The smart cap on the head with the camera attached captures real-time visual data, meanwhile there is a remote control with eight buttons, associated with the fulfillment of a variety of functions. With such a system, the visually impaired are given auditory information regarding the surroundings, objects, texts, recognition of known faces, and emotions as well. The latest machine learning concepts of VisionGuide are applied learning algorithms such as CNN for object detection, FaceNet for facial recognition, and Optical Character Recognition (OCR) for reading printed text. It captures visual data from the camera of the cap and gives audio feedback of the processed information through a text-to-speech engine. For navigation, the system uses geolocation services and routes planning to provide voice-guided directions in improving mobility. The major modules of the project are divided into seven: emergency notification, object detection, face recognition, emotion recognition, text reading, caption generation, and navigation. The emergency notification module generates messages with the user’s present location. Other modules help users interact with their environment in a hands-free mode. This solution enhances the independence, mobility, and information access of the visually impaired by addressing more than one challenge in an integrated, user-friendly system. Being an integrated functionality within one device, VisionGuide provides a pragmatic it is thus an efficient tool that helps in improving the quality of life of the visually disabled persons, accessible, convenient, and safe to use for everyday activity.