Real-Time Image Recognition for Intelligent Robots

Authors- C Sanjay, Assistant Professor Dr K Kumtha

Abstract-This project explores an image recognition system for robots, enabling them to interpret and interact with their environment more effectively. By utilizing YOLO (You Only Look Once), PyTorch, OpenCV, and Pillow (PIL), the system processes both static images and real- time webcam feeds to identify and classify objects. The system operates in the following key stages: (1) loading a pre-trained YOLO model for image recognition, (2) processing images to detect and classify objects, (3) enabling real-time recognition via webcam with bounding box overlays, and (4) offering users the flexibility to analyze either static images or live video streams. This image recognition framework significantly enhances robotic vision and has a wide range of applications, including autonomous navigation, object manipulation, and environmental awareness, making it highly beneficial for fields such as self-driving robotics, industrial automation, and smart surveillance.

DOI: /10.61463/ijset.vol.13.issue3.133