Title: Autonomous approach for moving object detection and classification in road applications

Authors: Imane El Manaa; Fadwa Benjelloun; My Abdelouahed Sabri; Ali Yahyaouy; Abdellah Aarab

Addresses: LISAC Laboratory, Faculty of Sciences Dhar Mahraz, Sidi Mohamed Ben Abdellah University, Fez, Morocco ' LISAC Laboratory, Faculty of Sciences Dhar Mahraz, Sidi Mohamed Ben Abdellah University, Fez, Morocco ' LISAC Laboratory, Faculty of Sciences Dhar Mahraz, Sidi Mohamed Ben Abdellah University, Fez, Morocco ' LISAC Laboratory, Faculty of Sciences Dhar Mahraz, Sidi Mohamed Ben Abdellah University, Fez, Morocco ' LISAC Laboratory, Faculty of Sciences Dhar Mahraz, Sidi Mohamed Ben Abdellah University, Fez, Morocco

Abstract: Our paper presents robust approaches for all moving object detection processes. First of all, we propose an automatic and non-parametric method in the segmentation phase based on Delaunay triangulation applied to the image histogram. For the feature extraction phase, we proceed by the GLCM technique for textural feature extraction and the HSV histogram method for the colour feature extraction. Those features will be used as input of the support vector machine (SVM) algorithm to design a robust classification model that will be used to differentiate between moving and static objects. Thus, static objects will be considered as a part of background, and in the other hand moving objects are surrounded by a bounding box in furtherance of careful tracking.

Keywords: moving object detection; segmentation; classification; computer vision; training; discriminating classifier; Delaunay triangulation; SVM; GLCM; feature extraction.

DOI: 10.1504/IJCAET.2023.127798

International Journal of Computer Aided Engineering and Technology, 2023 Vol.18 No.1/2/3, pp.211 - 223

Received: 20 Mar 2020
Accepted: 26 Jun 2020
Published online: 19 Dec 2022 *

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