• Sreenivasulu reddy D
• Dr.K G Chiranjeevi
• Mr N Surendra

Visual impairment limits an individual’s ability to navigate safely and independently. Traditional aids like white canes and guide dogs detect obstacles but cannot describe surroundings or identify specific objects. Existing sensor-based systems improve detection yet lack contextual awareness and adaptability. To address these challenges, this study proposes Smart Vision, an intelligent assistive system integrating deep learning–based object detection with real-time voice guidance. The system employs an Ensemble CNN model combining a fine-tuned MobileNetV2 and a custom CNN for accurate classification of objects such as doors, cabinets, and sofas. Detected objects are communicated through audio feedback to enhance spatial awareness and safety. Experimental results show that Smart Vision provides high detection accuracy, fast processing, and reliable voice assistance, offering a cost-effective and user-friendly solution for visually impaired individuals.

SMART VISION: OBJECT DETECTION AND VOICE ASSISTANCE FOR THE VISUALLY IMPAIRED

"SMART VISION: OBJECT DETECTION AND VOICE ASSISTANCE FOR THE VISUALLY IMPAIRED", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.12, Issue 11, page no.a210-a218, November-2025, Available :http://www.jetir.org/papers/JETIR2511026.pdf

"SMART VISION: OBJECT DETECTION AND VOICE ASSISTANCE FOR THE VISUALLY IMPAIRED", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.12, Issue 11, page no. ppa210-a218, November-2025, Available at : http://www.jetir.org/papers/JETIR2511026.pdf