심층 합성곱 신경망을 이용한 교통신호등 인식 = Traffic Light Recognition Using a Deep Convolutional Neural Network

다국어 초록 (Multilingual Abstract)

The color of traffic light is sensitive to various illumination conditions. Especially it loses the hue information when oversaturation happens on the lighting area. This paper proposes a traffic light recognition method robust to these illumination variations. The method consists of two steps of traffic light detection and recognition. It just uses the intensity and saturation in the first step of traffic light detection. It delays the use of hue information until it reaches to the second step of recognizing the signal of traffic light. We utilized a deep learning technique in the second step. We designed a deep convolutional neural network(DCNN) which is composed of three convolutional networks and two fully connected networks. 12 video clips were used to evaluate the performance of the proposed method. Experimental results show the performance of traffic light detection reporting the precision of 93.9%, the recall of 91.6%, and the recognition accuracy of 89.4%. Considering that the maximum distance between the camera and traffic lights is 70m, the results shows that the proposed method is effective.

참고문헌 (Reference)

• 1 조평근, "차량용 신호등의 형태적 특징과 연속 영상내의 위치 정보를 이용한 신호등 검출" 제어·로봇·시스템학회 21 (21): 1122-1129, 2015
• 2 김재곤, "적응적 형태학적 분석에 기초한 신호등 인식률 성능 개선" 한국정보통신학회 19 (19): 2129-2137, 2015
• 3 정준익, "성분차 색분할과 검출마스크를 통한실시간 교통신호등 검출과 인식" 대한전자공학회 43 (43): 65-72, 2006
• 4 김민기, "명도와 채도 기반의 점등영역 검출 및 모델 검증에 의한 교통신호등 판별" 한국멀티미디어학회 20 (20): 1729-1740, 2017
• 5 C. Jang, "Traffic Light Recognition Exploiting Map and Localization at Every Stage" 88 : 290-304, 2017
• 6 M. Omachi, "Traffic Light Detection with Color and Edge Information" 284-287, 2009
• 7 H. Moizumi, "Traffic Light Detection Considering Color Saturation Using In-Vehicle Stereo Camera" 24 (24): 349-357, 2016
• 8 M. D. -Cabrera, "Suspended Traffic Lights Detection and Distance Estimation Using Color Features" 1315-1320, 2012
• 9 V. John, "Saliency Map Generation by the Convolutional Neural Network for Real-TIme Traffic Light Detection Using Template Matching" 1 (1): 159-173, 2015
• 10 Z. Cai, "Real-time Recognition System of Traffic Light in Urban Environment" 1-6, 2012
• 11 X. Zhou, "Real-Time Traffic Light Recognition Based on C-HOG Features" 36 (36): 793-814, 2017
• 12 M. Weber, "Deep TLR: A Single Deep Convolutional Network for Detection and Classification of Traffic Lights" 342-348, 2016
• 13 T. Gevers, "Color-based Object Recognition" 32 (32): 453-464, 1999
• 14 Z. Ozcelik, "A Vision Based Traffic Light Detection and Recognition Approach for Intelligent Vehicles" 424-429, 2017
• 15 K. Behrendt, "A Deep Learning Approach to Traffic Lights: Detection, Tracking, and Classification" 1370-1377, 2017