TRAFFIC SIGN CLASSIFICATION AND DETECTION USING DEEP LEARNING
Author(s):
SUNITHA.A , SHANTHI.S
Keywords:
Traffic sign detection and trailing (TSDR), advanced driver help system (ADAS)
Abstract
Driving could be a advanced, continuous, and multitask method that involves driver's noesis, perception, and motor movements. The approach road traffic signs and vehicle data is displayed impacts powerfully driver's attention with enlarged mental work resulting in safety considerations. Drivers should keep their eyes on the road, however will continually use some help in maintaining their awareness and guiding their attention to potential rising hazards. In-vehicle discourse increased Reality (AR) has the potential to supply novel visual feedbacks to drivers for Associate in nursing increased driving expertise. A replacement period approach for quick and correct framework for traffic sign recognition, supported Cascade Deep learning and AR, that superimposes increased virtual objects onto a true scene below every kind of driving things, together with unfavorable weather. Experiments results show that, by deep convolutional neural networks show that the joint learning greatly enhances the potential of detection and still retains its realtime performance. Investigations on vision-based TSDR have received substantial interest within the analysis community, that is especially impelled by 3 factors, that area unit detection, trailing and classification.
Article Details
Unique Paper ID: 149112
Publication Volume & Issue: Volume 6, Issue 11
Page(s): 169 - 175
Article Preview & Download
Share This Article
Join our RMS
Conference Alert
NCSEM 2024
National Conference on Sustainable Engineering and Management - 2024