Title :
Car recognition for multiple data sets based on histogram of oriented gradients and support vector machines
Author_Institution :
Equipe RITM, Univ. Mohammed V, Rabat, Morocco
Abstract :
In this paper we implement a soft margin support vector machine (SVM) using linear and nonlinear kernels for image classification. The sequential minimal optimization (SMO) method will be used to find the optimal separating hyperplane between the training data that consist of a multidimensional histogram of oriented gradient (HOG) features vectors extracted from multiple data sets.
Keywords :
feature extraction; image classification; image recognition; optimisation; support vector machines; HOG; SMO method; SVM; car recognition; features vectors extraction; histogram of oriented gradients; image classification; linear kernels; multidimensional histogram of oriented gradient; nonlinear kernels; sequential minimal optimization method; soft margin support vector machine; Support vector machines; Testing; Support vector machines; histogram of oriented gradients; machine learning;
Conference_Titel :
Multimedia Computing and Systems (ICMCS), 2012 International Conference on
Conference_Location :
Tangier
Print_ISBN :
978-1-4673-1518-0
DOI :
10.1109/ICMCS.2012.6320284
