Kernel based automatic traffic sign detection and recognition using SVM

Anjan Gudigar, B. N. Jagadale, Mahesh P.k., Raghavendra U

Research output: Chapter in Book/Report/Conference proceedingConference contribution

10 Citations (Scopus)

Abstract

Traffic sign detection and recognition is an important issue of research recently. Road and traffic signs have been designed according to stringent regulations using special shapes and colors, very different from the natural environment, which makes them easily recognizable by drivers. The human visual perception abilities depend on the individual's physical and mental conditions. In certain conditions, these abilities can be affected by many factors such as fatigue, and observatory skills. Detection of regulatory road signs in outdoor images from moving vehicles will help the driver to take the right decision in good time, which means fewer accidents, less pollution, and better safety. In automatic traffic-sign maintenance and in a visual driver-assistance system, road-sign detection and recognition are two of the most important functions. This paper presents automatic regulatory road-sign detection with the help of distance to borders (DtBs) and distance from centers (DfCs) feature vectors. Our system is able to detect and recognize regulatory road signs. The proposed recognition system is based on the generalization properties of SVMs. The system consists of following processes: segmentation according to the color of the pixel, traffic-sign detection by shape classification using linear SVM and content recognition based on Gaussian-kernel SVM. A result shows a high success rate and a very low amount of false positives in the final recognition stage.

Original languageEnglish
Title of host publicationEco-Friendly Computing and Communication Systems - International Conference, ICECCS 2012, Proceedings
Pages153-161
Number of pages9
Volume305 CCIS
DOIs
Publication statusPublished - 2012
EventInternational Conference on Eco-Friendly Computing and Communication Systems, ICECCS 2012 - Kochi, India
Duration: 09-08-201211-08-2012

Publication series

NameCommunications in Computer and Information Science
Volume305 CCIS
ISSN (Print)1865-0929

Conference

ConferenceInternational Conference on Eco-Friendly Computing and Communication Systems, ICECCS 2012
CountryIndia
CityKochi
Period09-08-1211-08-12

Fingerprint

Traffic signs
Color
Observatories
Accidents
Pollution
Pixels
Fatigue of materials

All Science Journal Classification (ASJC) codes

  • Computer Science(all)

Cite this

Gudigar, A., Jagadale, B. N., P.k., M., & U, R. (2012). Kernel based automatic traffic sign detection and recognition using SVM. In Eco-Friendly Computing and Communication Systems - International Conference, ICECCS 2012, Proceedings (Vol. 305 CCIS, pp. 153-161). (Communications in Computer and Information Science; Vol. 305 CCIS). https://doi.org/10.1007/978-3-642-32112-2_19
Gudigar, Anjan ; Jagadale, B. N. ; P.k., Mahesh ; U, Raghavendra. / Kernel based automatic traffic sign detection and recognition using SVM. Eco-Friendly Computing and Communication Systems - International Conference, ICECCS 2012, Proceedings. Vol. 305 CCIS 2012. pp. 153-161 (Communications in Computer and Information Science).
@inproceedings{e0cad41f511a4770becd1592e04cf210,
title = "Kernel based automatic traffic sign detection and recognition using SVM",
abstract = "Traffic sign detection and recognition is an important issue of research recently. Road and traffic signs have been designed according to stringent regulations using special shapes and colors, very different from the natural environment, which makes them easily recognizable by drivers. The human visual perception abilities depend on the individual's physical and mental conditions. In certain conditions, these abilities can be affected by many factors such as fatigue, and observatory skills. Detection of regulatory road signs in outdoor images from moving vehicles will help the driver to take the right decision in good time, which means fewer accidents, less pollution, and better safety. In automatic traffic-sign maintenance and in a visual driver-assistance system, road-sign detection and recognition are two of the most important functions. This paper presents automatic regulatory road-sign detection with the help of distance to borders (DtBs) and distance from centers (DfCs) feature vectors. Our system is able to detect and recognize regulatory road signs. The proposed recognition system is based on the generalization properties of SVMs. The system consists of following processes: segmentation according to the color of the pixel, traffic-sign detection by shape classification using linear SVM and content recognition based on Gaussian-kernel SVM. A result shows a high success rate and a very low amount of false positives in the final recognition stage.",
author = "Anjan Gudigar and Jagadale, {B. N.} and Mahesh P.k. and Raghavendra U",
year = "2012",
doi = "10.1007/978-3-642-32112-2_19",
language = "English",
isbn = "9783642321115",
volume = "305 CCIS",
series = "Communications in Computer and Information Science",
pages = "153--161",
booktitle = "Eco-Friendly Computing and Communication Systems - International Conference, ICECCS 2012, Proceedings",

}

Gudigar, A, Jagadale, BN, P.k., M & U, R 2012, Kernel based automatic traffic sign detection and recognition using SVM. in Eco-Friendly Computing and Communication Systems - International Conference, ICECCS 2012, Proceedings. vol. 305 CCIS, Communications in Computer and Information Science, vol. 305 CCIS, pp. 153-161, International Conference on Eco-Friendly Computing and Communication Systems, ICECCS 2012, Kochi, India, 09-08-12. https://doi.org/10.1007/978-3-642-32112-2_19

Kernel based automatic traffic sign detection and recognition using SVM. / Gudigar, Anjan; Jagadale, B. N.; P.k., Mahesh; U, Raghavendra.

Eco-Friendly Computing and Communication Systems - International Conference, ICECCS 2012, Proceedings. Vol. 305 CCIS 2012. p. 153-161 (Communications in Computer and Information Science; Vol. 305 CCIS).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

TY - GEN

T1 - Kernel based automatic traffic sign detection and recognition using SVM

AU - Gudigar, Anjan

AU - Jagadale, B. N.

AU - P.k., Mahesh

AU - U, Raghavendra

PY - 2012

Y1 - 2012

N2 - Traffic sign detection and recognition is an important issue of research recently. Road and traffic signs have been designed according to stringent regulations using special shapes and colors, very different from the natural environment, which makes them easily recognizable by drivers. The human visual perception abilities depend on the individual's physical and mental conditions. In certain conditions, these abilities can be affected by many factors such as fatigue, and observatory skills. Detection of regulatory road signs in outdoor images from moving vehicles will help the driver to take the right decision in good time, which means fewer accidents, less pollution, and better safety. In automatic traffic-sign maintenance and in a visual driver-assistance system, road-sign detection and recognition are two of the most important functions. This paper presents automatic regulatory road-sign detection with the help of distance to borders (DtBs) and distance from centers (DfCs) feature vectors. Our system is able to detect and recognize regulatory road signs. The proposed recognition system is based on the generalization properties of SVMs. The system consists of following processes: segmentation according to the color of the pixel, traffic-sign detection by shape classification using linear SVM and content recognition based on Gaussian-kernel SVM. A result shows a high success rate and a very low amount of false positives in the final recognition stage.

AB - Traffic sign detection and recognition is an important issue of research recently. Road and traffic signs have been designed according to stringent regulations using special shapes and colors, very different from the natural environment, which makes them easily recognizable by drivers. The human visual perception abilities depend on the individual's physical and mental conditions. In certain conditions, these abilities can be affected by many factors such as fatigue, and observatory skills. Detection of regulatory road signs in outdoor images from moving vehicles will help the driver to take the right decision in good time, which means fewer accidents, less pollution, and better safety. In automatic traffic-sign maintenance and in a visual driver-assistance system, road-sign detection and recognition are two of the most important functions. This paper presents automatic regulatory road-sign detection with the help of distance to borders (DtBs) and distance from centers (DfCs) feature vectors. Our system is able to detect and recognize regulatory road signs. The proposed recognition system is based on the generalization properties of SVMs. The system consists of following processes: segmentation according to the color of the pixel, traffic-sign detection by shape classification using linear SVM and content recognition based on Gaussian-kernel SVM. A result shows a high success rate and a very low amount of false positives in the final recognition stage.

UR - http://www.scopus.com/inward/record.url?scp=84865230001&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84865230001&partnerID=8YFLogxK

U2 - 10.1007/978-3-642-32112-2_19

DO - 10.1007/978-3-642-32112-2_19

M3 - Conference contribution

AN - SCOPUS:84865230001

SN - 9783642321115

VL - 305 CCIS

T3 - Communications in Computer and Information Science

SP - 153

EP - 161

BT - Eco-Friendly Computing and Communication Systems - International Conference, ICECCS 2012, Proceedings

ER -

Gudigar A, Jagadale BN, P.k. M, U R. Kernel based automatic traffic sign detection and recognition using SVM. In Eco-Friendly Computing and Communication Systems - International Conference, ICECCS 2012, Proceedings. Vol. 305 CCIS. 2012. p. 153-161. (Communications in Computer and Information Science). https://doi.org/10.1007/978-3-642-32112-2_19