Moving object detection under different weather conditions using full-spectrum light sources

Abstract

The moving object detection always remains an active field of research given the variety of challenges related to this topic. In fact, most of the challenges related to the low illumination and weather conditions (fog, snow, rain, etc.) remain unresolved and require more developments. In this paper, our intrinsic objective is to overcome these challenges using an effective moving object detection method. Unlike most works in the literature that use one of the two infrared or visible spectra independently, we proposed a Moving Object Detection method based on background modeling using the Full-Spectrum Light Sources (FSLS-MOD). To better ensure the adaptability and independence of the moving object speeds and sizes, the principle of the inter-frame differences’ methods is introduced in the background modeling stage. Furthermore, we applied a new strategy to switch between the spectra allowing us to benefit from the advantages of each spectrum and carry out a better moving object detection even in bad weather conditions. An experimental study by quantitative and qualitative evaluations proved the robustness and effectiveness of our proposed method of moving object detection using the switching strategy between full-spectrum light sources under different illuminations and weather conditions.