Traffic Light Detection Using Rotated Principal Component Analysis for Video-Based Car Navigation System

• Published in: IEICE Transactions on Information and Systems Vol. E91.D; no. 12; pp. 2884 - 2887
• Main Authors: JOO, Sung-Kwan, KIM, Yongkwon, CHO, Seong Ik, CHOI, Kyoungho, LEE, Kisung
• Format: Journal Article
• Language: English
• Published: Oxford The Institute of Electronics, Information and Communication Engineers 2008; Oxford University Press
• Subjects: Algorithms; Applied sciences; Artificial intelligence; Cameras; car navigation system; Computer science; control theory; systems; Connectionism. Neural networks; crossroad detection; Detection, estimation, filtering, equalization, prediction; Electric, optical and optoelectronic circuits; Electronics; Exact sciences and technology; Filtering; Filtration; Information, signal and communications theory; Neural networks; Polycarbonates; Principal component analysis; Signal and communications theory; Signal, noise; Systems, networks and services of telecommunications; Telecommunications; Telecommunications and information theory; Teletraffic; Traffic engineering; Traffic flow; traffic light; Navigation system; Histogram; Filtering; car navigation system; Threshold detection; Teletraffic; Neural network; Algorithm; crossroad detection; Statistical method; Image quality; Principal component analysis; traffic light
• ISSN: 0916-8532; 1745-1361; 1745-1361
• DOI: 10.1093/ietisy/e91-d.12.2884

This letter presents a novel approach for traffic light detection in a video frame captured by an in-vehicle camera. The algorithm consists of rotated principal component analysis (RPCA), modified amplitude thresholding with respect to the histograms of the PC planes and final filtering with a neural network. The proposed algorithm achieves an average detection rate of 96% and is very robust to variations in the image quality.