Title :
Expanded gradient approach in vehicle detection and recognition systems
Author :
Boichenko, Cyril
Author_Institution :
Zaporizhzhya Nat. Univ., Zaporizhzhya, Ukraine
Abstract :
Method of multilayer multi-direct gradient map models construction and its application as a feature system are first proposed. Automatic vehicle classification system which uses the proposed methods and models is developed. The advantages of described method and the optimality of parameter set for the model are proved.
Keywords :
gradient methods; image recognition; road vehicles; traffic engineering computing; multilayer multidirect gradient map models; parameter set optimality; vehicle detection system; vehicle recognition system; Brightness; Computer vision; Control systems; Convolution; Image recognition; Kernel; Nonhomogeneous media; Real time systems; Vehicle detection; Vehicles; binarization; feature transform; object of control; pattern recognition; region of interest;
Conference_Titel :
Modern Problems of Radio Engineering, Telecommunications and Computer Science (TCSET), 2010 International Conference on
Conference_Location :
Lviv-Slavske
Print_ISBN :
978-966-553-875-2
Electronic_ISBN :
978-966-553-901-8
