Advanced Deep Learning for Real-Time Object Identification and Assistance for the Blind

Advanced Deep Learning for Real-Time Object Identification and Assistance for the Blind

BUVANESWARI S1,POOJA A2,PRATHAP A3,PREMKUMAR B4,Mrs.PREMADEVI P5

1,2,3,4 B.E, Student Department Computer Science And Engineering, Angel College Of Engineering And Technology,       Tiruppur, Tamil Nadu, India.

5 Asst. Professor, Department Computer Science And Engineering, Angel College Of Engineering And Technology, Tiruppur, Tamil Nadu, India.

ABSTRACT :
 

“ Advanced Deep Learning for Real-Time Object Identification and Assistance for the Blind” and is developed  using HTML, CSS, and JavaScript for the front end, with Python (Flask framework) for the back end. Object Detection involves identifying and recognizing real-world objects such as backgrounds, airplanes, bicycles, birds, boats, bottles, buses, cars, cats, chairs, cows, dining tables, dogs, horses, motorbikes, people and humans from images or videos. This technique enables understanding the details of a webcam recognizing, localizing, and detecting multiple objects within it. Cater to various real-world applications. Webcam input allows users to process webcam object for object detection, analyzing each frame in real time Camera input integrates live camera feeds, enabling users to detect objects in their surroundings in real time, making it highly useful for tasks requiring immediate feedback, such as assisting visually impaired individuals. This system is motivated by the need to aid visually impaired individuals, those who face challenges in navigating unfamiliar environments. By utilizing real-time object detection through camera inputs, our system aims to assist visually impaired users by identifying objects around them, and also get a voice of the object thus enhancing their ability to navigate independently.

INDEX TERMS: Real-time object detection, Deep learning, Computer vision