Traffic Sign Recognition Using Complementary Features

Traffic sign recognition is difficult due to the low resolution of image, illumination variation and shape distortion. On the public dataset GTSRB, the state-of-the-art performance have been obtained by convolutional neural networks (CNNs), which learn discriminative features automatically to achiev...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Proceedings - IEEE Computer Society Conference on Pattern Recognition and Image Processing S. 210 - 214
Hauptverfasser: Suisui Tang, Lin-Lin Huang
Format: Tagungsbericht
Sprache:Englisch
Veröffentlicht: IEEE 01.11.2013
Schlagworte:
Accuracy
Complementary Features
Feature extraction
Gabor filters
Histograms
Linear SVM
Pattern recognition
Support vector machines
Traffic Sign Recognition
Training
ISSN:0730-6512
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:Traffic sign recognition is difficult due to the low resolution of image, illumination variation and shape distortion. On the public dataset GTSRB, the state-of-the-art performance have been obtained by convolutional neural networks (CNNs), which learn discriminative features automatically to achieve high accuracy but suffer from high computation costs in both training and classification. In this paper, we propose an effective traffic sign recognition method using multiple features which have demonstrated effective in computer vision and are computationally efficient. The extracted features are the histogram of oriented gradients (HOG) feature, Gabor filter feature and local binary pattern (LBP) feature. Using a linear support vector machine (SVM) for classification, each feature yields fairly high accuracy. The combination of three features has shown good complementariness and yielded competitively high accuracy. On the GTSRB dataset, our method reports an accuracy of 98.65%.
ISSN:0730-6512
DOI:10.1109/ACPR.2013.63