Real-time traffic sign recognition from video by class-specific discriminative features

In this paper we address the problem of traffic sign recognition. Novel image representation and discriminative feature selection algorithms are utilised in a traditional three-stage framework involving detection, tracking and recognition. The detector captures instances of equiangular polygons in t...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Pattern recognition Jg. 43; H. 1; S. 416 - 430
Hauptverfasser: Ruta, Andrzej, Li, Yongmin, Liu, Xiaohui
Format: Journal Article
Sprache:Englisch
Veröffentlicht: Kidlington Elsevier Ltd 2010
Elsevier
Schlagworte:
Applied sciences
Artificial intelligence
Colour Distance Transform
Computer science; control theory; systems
Computer vision-based driver assistance
Discriminative local regions
Exact sciences and technology
Forward feature selection
Image processing
Information, signal and communications theory
Miscellaneous
Pattern recognition. Digital image processing. Computational geometry
Signal and communications theory
Signal processing
Signal representation. Spectral analysis
Signal, noise
Telecommunications and information theory
Traffic sign recognition
Forward feature selection
Traffic sign recognition
Computer vision-based driver assistance
Colour Distance Transform
Discriminative local regions
Performance evaluation
Automatic classification
Similarity
Image processing
Learning
Polygonal shape
Image representation
Driver information systems
Learning algorithm
Nearest neighbour
Computer vision
Discriminant analysis
Target tracking
Teletraffic
Distance transformation
Signal representation
Signal classification
Color image
Statistical method
Interest region
Image analysis
Driver assistance
Signal processing
Feature extraction
Open market
Principal component analysis
ISSN:0031-3203, 1873-5142
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:In this paper we address the problem of traffic sign recognition. Novel image representation and discriminative feature selection algorithms are utilised in a traditional three-stage framework involving detection, tracking and recognition. The detector captures instances of equiangular polygons in the scene which is first appropriately filtered to extract the relevant colour information and establish the regions of interest. The tracker predicts the position and the scale of the detected sign candidate over time to reduce computation. The classifier compares a discrete-colour image of the observed sign with the model images with respect to the class-specific sets of discriminative local regions. They are learned off-line from the idealised template sign images, in accordance with the principle of one-vs-all dissimilarity maximisation. This dissimilarity is defined based on the so-called Colour Distance Transform which enables robust discrete-colour image comparisons. It is shown that compared to the well-established feature selection techniques, such as Principal Component Analysis or AdaBoost, our approach offers a more adequate description of signs and involves effortless training. Upon this description we have managed to build an efficient road sign recognition system which, based on a conventional nearest neighbour classifier and a simple temporal integration scheme, demonstrates a competitive performance in the experiments involving real traffic video.
ISSN:0031-3203
1873-5142
DOI:10.1016/j.patcog.2009.05.018