This paper presents an algorithm to detect and recognize the information contained in road panels. The aim of this work is to complement the functionality of a traffic signposting inspection system based on computer vision, which is able to collect data related to the maintenance state of traffic signs and panels automatically. In this context, not only a good visibility of the panels is vital for a safe use by road users, but also the suitability of the information contained in the traffic panels. The algorithm presented here, which is based on SIFT descriptors to recognize single characters and also on HMMs to recognize whole words, will be able to make an inventory of the information contained in traffic panels with the aim to check its reliability and brevity automatically. Experimental results and conclusions obtained after analysing a diverse set of real images show the effectiveness of the proposed method.
Automatic information recognition of traffic panels using SIFT descriptors and HMMs
2010-09-01
938518 byte
Conference paper
Electronic Resource
English
Traffic signs recognition based on PCA-SIFT
IEEE | 2014
|HMMs with Mixtures of Trend Functions for Automatic Speech Recognition
British Library Conference Proceedings | 1994
|PCA-SIFT: A More Distinctive Representation for Local Image Descriptors
British Library Conference Proceedings | 2004
|