Detection and Classification of Signage’s from Random Mobile Videos Using Local Binary Patterns

The Traffic-Sign detection and recognition plays significant role in the design of autonomous driverless cars for navigation purpose as well as to assist a driver for alerting and educating him about the tracked signage on the road side. The main objective of this paper is to highlight an automatic...

Full description

Saved in:
Bibliographic Details
Published in:International journal of image, graphics and signal processing graphics and signal processing, 2018-02, Vol.10 (2), p.52-59
Main Authors: S Gornale, Shivanand, Babaleshwar, Ashvini K, Yannawar, Pravin L
Format: Article
Language:English
Subjects:
Autonomous cars
Autonomous navigation
Classification
Light
Pattern recognition
Support vector machines
Traffic signs
Video data
Citations: Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:The Traffic-Sign detection and recognition plays significant role in the design of autonomous driverless cars for navigation purpose as well as to assist a driver for alerting and educating him about the tracked signage on the road side. The main objective of this paper is to highlight an automatic process of detection of Region Of Interest (ROI) which marks or isolates signage’s from color video streams and performs classification of automatically detected signage’s based on support vector machine (SVM) classifiers trained over Local Binary Pattern (LBP) features. The training dataset was captured through 13 mega pixel mobile camera in different illumination and light conditions and due to randomness the data base complexity is very high. The robustness of the proposed system is measured on the bases its of capability of automatic detection and classification of ROI in a given video stream and backed with a comprehensive result analysis presented in this piece of work.
ISSN:2074-9074
2074-9082
DOI:10.5815/ijigsp.2018.02.06