Nighttime Foreground Pedestrian Detection Based on Three-Dimensional Voxel Surface Model

Abstract

Pedestrian detection is among the most frequently-used preprocessing tasks in many surveillance application fields, from low-level people counting to high-level scene understanding. Even though many approaches perform well in the daytime with sufficient illumination, pedestrian detection at night is still a critical and challenging problem for video surveillance systems. To respond to this need, in this paper, we provide an affordable solution with a near-infrared stereo network camera, as well as a novel three-dimensional foreground pedestrian detection model. Specifically, instead of using an expensive thermal camera, we build a near-infrared stereo vision system with two calibrated network cameras and near-infrared lamps. The core of the system is a novel voxel surface model, which is able to estimate the dynamic changes of three-dimensional geometric information of the surveillance scene and to segment and locate foreground pedestrians in real time. A free update policy for unknown points is designed for model updating, and the extracted shadow of the pedestrian is adopted to remove foreground false alarms. To evaluate the performance of the proposed model, the system is deployed in several nighttime surveillance scenes. Experimental results demonstrate that our method is capable of nighttime pedestrian segmentation and detection in real time under heavy occlusion. In addition, the qualitative and quantitative comparison results show that our work outperforms classical background subtraction approaches and a recent RGB-D method, as well as achieving comparable performance with the state-of-the-art deep learning pedestrian detection method even with a much lower hardware cost.

Tuning of Classifiers to Speed-Up Detection of Pedestrians in Infrared Images

“…They are widely used due to the relatively low cost of cameras and their small size, especially in closed circuit television (CCTV). However, they have a limited detection range and require additional illuminators [4]. NIR detectors can also be dazzled by oncoming vehicle headlights (or illuminators) and perform much worse than the FIR ones in fog.…”

An Adaptive Framework for Multi-Vehicle Ground Speed Estimation in Airborne Videos

“…However, the target in the image is small, the background in the scene is complex, and the motion blurring will be caused by the vibration of the UAV, which brings new problems for target detection. Currently, there are many popular target detection algorithms, such as background modeling methods like the visual background extractor (ViBe) [32], stereo vision methods [33], and deep learning methods like YOLOv3 [31]. In the system, we used YOLOv3 [31] to obtain an accurate detection bounding box of the vehicle in the image.…”

Development of a Sensor to Measure Physician Consultation Times

“…An advanced application such as this requires stereo/video capturing, covering a wider area and specific analysis methods [4][5][6]. Li et al [7] added a near-infrared stereo camera system for application in low light levels. Consequently, this method is available for nearly every condition but raises significant privacy issues.…”

“…The first tests were conducted with an Arduino Uno (Figure 1a), which has been replaced by an Arduino Nano for the final version due to the smaller dimensions (Figure 1d,e). (4), Arduino (5), Micro SD card module (6), support structure (7), and power supply (8).…”