Scholarlink

The integration of advanced technologies into assistive devices has significantly enhanced the independence and mobility of visually impaired individuals. This survey explores recent advancements in object detection and navigation systems using technologies such as ultrasonic sensors, facial recognition, and embedded platforms like Raspberry Pi. The study reviews methodologies from key research papers that leverage lightweight machine learning models, wearable devices, and sensor-based solutions to improve obstacle detection and environmental awareness. Innovations such as TensorFlow Lite for resource-constrained systems, MobileNet-Tiny for real-time object detection, and smart belt navigation solutions demonstrate the potential of these technologies. Comparative analysis reveals the increasing influence of newer studies, highlighting the evolving landscape of assistive technology research. Despite significant progress, challenges remain in sensor precision, real-time feedback accuracy, and system optimization for complex environments. This paper provides insights into the state-of-the-art technologies and identifies potential areas for future research, emphasizing the need for further refinement to enhance user experience and system efficiency.

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

Vision Tools for the Visually Impaired: A Survey on Object Detection and Navigation Technologies

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Rakhi Bharadwaj,
  • Snehal Darade Bhushan Bachewar,
  • Harsh Badagandi,
  • Shreya Barsude

摘要

The integration of advanced technologies into assistive devices has significantly enhanced the independence and mobility of visually impaired individuals. This survey explores recent advancements in object detection and navigation systems using technologies such as ultrasonic sensors, facial recognition, and embedded platforms like Raspberry Pi. The study reviews methodologies from key research papers that leverage lightweight machine learning models, wearable devices, and sensor-based solutions to improve obstacle detection and environmental awareness. Innovations such as TensorFlow Lite for resource-constrained systems, MobileNet-Tiny for real-time object detection, and smart belt navigation solutions demonstrate the potential of these technologies. Comparative analysis reveals the increasing influence of newer studies, highlighting the evolving landscape of assistive technology research. Despite significant progress, challenges remain in sensor precision, real-time feedback accuracy, and system optimization for complex environments. This paper provides insights into the state-of-the-art technologies and identifies potential areas for future research, emphasizing the need for further refinement to enhance user experience and system efficiency.