Authors: Sabrine Hamdi; Souhir Sghaier; Hassene Faiedh; Chokri Souani
Addresses: National Engineering School of Sousse, Sousse University, Sousse, 4023, Tunisia ' University College of Ranyah, Taif University, Taif, 21974, Saudi Arabia ' Higher Institute of Applied Sciences and Technology of Sousse, Sousse University, Sousse, 4003, Tunisia ' Higher Institute of Applied Sciences and Technology of Sousse, Sousse University, Sousse, 4003, Tunisia
Abstract: Vision-based pedestrian detection is a challenge task for a variety of applications such as driving assistance systems, especially in case of insufficient illumination. Effective fusion of complementary information acquired by multispectral images (visible and infrared) allows robust pedestrian detection under various lighting conditions (e.g., day and nighttime). In this paper, we propose a multispectral pedestrian detection approach that combines visible and infrared images. Firstly, an Otsu thresholding is applied to infrared images to detect hot spots most likely representing a pedestrian, after applying some morphological operations to enhance the original image and compensate for clothing-based distortions. The significant regions of interest obtained in the infrared image are mapped into corresponding visible image. Secondly, multispectral aggregated channel features are used with a thermal discrete cosine transform, as descriptor combined with a support vector machine (SVM) classifier. Our approach is evaluated on the KAIST multispectral dataset to prove its efficiency.
Keywords: pedestrian detection; day/night-time; visible image; infrared image; multispectral data; Otsu thresholding; morphological operations; aggregated channel features; discrete cosine transform; SVMs; support vector machines; KAIST dataset.
International Journal of Vehicle Design, 2020 Vol.83 No.2/3/4, pp.140 - 171
Received: 02 Mar 2020
Accepted: 21 Oct 2020
Published online: 09 May 2021 *