Title :
M-SIFT: A new method for Vehicle Logo Recognition
Author :
Psyllos, A. ; Anagnostopoulos, Christos-Nikolaos ; Kayafas, Eleftherios
Author_Institution :
Sc. of Appl. Math. & Phys. Sci., NTUA, Athens, Greece
Abstract :
In this paper, a new algorithm for Vehicle Logo Recognition is proposed, on the basis of an enhanced Scale Invariant Feature Transform (Merge-SIFT or M-SIFT). The algorithm is assessed on a set of 1500 logo images that belong to 10 distinctive vehicle manufacturers. A series of experiments are conducted, splitting the 1500 images to a training set (database) and to a testing set (query). It is shown that the MSIFT approach, which is proposed in this paper, boosts the recognition accuracy compared to the standard SIFT method. The reported results indicate an average of 94.6% true recognition rate in vehicle logos, while the processing time remains low (~0.8sec).
Keywords :
image recognition; transforms; vehicles; M-SIFT; merge-scale invariant feature transform; vehicle logo recognition; vehicle manufacturer; Databases; Equations; Feature extraction; Image recognition; Mathematical model; Transforms; Vehicles;
Conference_Titel :
Vehicular Electronics and Safety (ICVES), 2012 IEEE International Conference on
Conference_Location :
Istanbul
Print_ISBN :
978-1-4673-0992-9
DOI :
10.1109/ICVES.2012.6294277
