Vision Enhancer: A Voice-Driven System for Obstacle Detection

The eyes serve as the primary sensory organ for individuals, providing vital visual information that aids in perception, recognition, and understanding of the surrounding environment. However, for those with visual impairments, ac- cessing visual information poses significant challenges, limiting their freedom, particularly in navigating through obstacles. The paper attempts to address the obstacles faced by visually impaired individuals by leveraging object detection techniques such as TensorFlow to identify obstacles in their surroundings. By incorporating depth estimation, the system aims to determine the distance to these obstacles, providing crucial spatial awareness. The system generates warnings through voice feedback, aiding users in making informed decisions to navigate safely. The implementation of this detection system is active on handheld devices, predominantly utilizing webcams or mobile devices. Through this approach, the system achieves a commendable accuracy rate of 98%, thereby enhancing the autonomy and safety of individuals with visual impairments in their daily lives.