Title: Application of machine learning approach in detection and classification of cars of an image

Authors: B. Ashwini; B.N. Yuvaraju

Addresses: Department of Information Science and Engineering, NMAMIT (Affiliated to V.T.U. Belgavi), Nitte, Karnataka 574110, India ' Department of Computer Science and Engineering, NIE (Affiliated to V.T.U. Belagavi), Mysuru, Karnataka, India

Abstract: Support vector machine (SVM) qualified on histogram orientation gradients (HOG) features is a genuine standard across many visual awareness responsibilities. Due to the change in the illumination and scene complexity, moving vehicle detection has become one of the very important components. Therefore, in this paper, a HOG feature descriptor is proposed. HOG features are not perceptive to illumination change and performance is better in characterising object shape and appearance. A feature vector is built by combining all the HOG features, which are required to train a linear SVM classifier for classification of vehicles.

Keywords: classification; HOG features; machine learning; SVM classifier; vehicle detection.

DOI: 10.1504/IJSISE.2017.084564

International Journal of Signal and Imaging Systems Engineering, 2017 Vol.10 No.1/2, pp.8 - 13

Received: 02 Nov 2016
Accepted: 24 Jan 2017
Published online: 14 Jun 2017 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article