Traffic Sign Recognition Approach Using Artificial Neural Network and Chi-Squared Feature Selection

Sie sind hier: Homepage > Suche

Traffic Sign Recognition Approach Using Artificial Neural Network and Chi-Squared Feature Selection

Vashisht, Manisha / Kumar, Brijesh

With the rising population and vehicular traffic across the globe, driver safety on road has become a huge concern for most governments. Emerging technologies and industrial revolutions have given rise to concept of autonomous cars. The driving systems embedded in these cars identify the traffic signs on the road and then take appropriate action. In spite of all these efforts, the accuracy of traffic sign image detection still remains a challenge for most car manufacturers and drivers, especially under difficult weather conditions. Multiple authors have done research in past and have proposed approaches relevant to identification of traffic sign images. The proposed solutions on traffic sign image detection have been influenced largely by Artificial Intelligence (AI)-based implementation techniques. In this research paper, authors have used Mapillary public traffic image dataset and have proposed an innovative approach using chi-squared ranking algorithm along with ANN for image classification. The effectiveness of proposed approach is compared with some related works. Experimental results showed that the proposed enhanced algorithm based on ANN and chi-squared algorithm provided better results.

Zugriff

Zugriff prüfen

Verfügbarkeit in meiner Bibliothek prüfen

Bestellung bei Subito €

Exportieren, teilen und zitieren

Dokumentinformationen

Titel :

Traffic Sign Recognition Approach Using Artificial Neural Network and Chi-Squared Feature Selection

Weitere Titelangaben:

Lect. Notes in Networks, Syst.

Beteiligte:

Kumar, Raghvendra (Herausgeber:in) / Pattnaik, Prasant Kumar (Herausgeber:in) / R. S. Tavares, João Manuel (Herausgeber:in) / Vashisht, Manisha (Autor:in) / Kumar, Brijesh (Autor:in)

Erschienen in:

Next Generation of Internet of Things ; Kapitel : 45 ; 519-527

Lecture Notes in Networks and Systems ; 445

Erscheinungsdatum :

2022-09-27

Format / Umfang :

9 pages

ISBN :

978-981-19-1412-6 , 978-981-19-1411-9

ISSN :

2367-3389 , 2367-3370

DOI :