A new pedestrian detection method for intelligent surveillance systems
Redouan Lahmyed, Mohamed El Ansari, Lahcen Koutti · 2025
In this chapter, we present an intelligent surveillance-based system for fast robust pedestrian detection. The new approach detects pedestrians in a walking pose from a single visible camera, consisting of two components. The first component generates the pedestrian regions of interest (ROI) hypothesis by exploiting the motion information. The second component consists of pedestrian hypothesis validation by using the so-called Colored Gradient Local Binary Patterns (GLBP)-COLOR, which is an extension of the classical GLBP feature to the RGB color space together with a support vector machine (SVM) classifier to detect the pedestrians from the extracted ROIs. The proposed method has been tested on INO Video Analytics dataset and the obtained results justify its effectiveness.