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.

Nighttime Pedestrian Detection With A Normal Camera Using Svm Classifier

A Novel Approach to Design the Fast Pedestrian Detection for Video Surveillance System

Research on Hierarchical Pedestrian Detection Based on SVM Classifier with Improved Kernel Function

Efficient And Real-Time Pedestrian Detection At Night-Time Environments

Nighttime Pedestrian Detection With Near Infrared Using Cascaded Classifiers

Fast Pedestrian Detection And Tracking Based On Vibe Combined Hog-Svm Scheme

Nighttime vehicle detection based on illumination visibility recognition

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

Nighttime Pedestrian Detection Method for Driver Assistance Systems

Detection of Pedestrians for Far-Infrared Automotive Night Vision Systems Using Learning-Based Method and Head Validation

A multi-stage classifier based algorithm of pedestrian detection in night with a near infrared camera in a moving car.

Real-time pedestrian detection and tracking at nighttime for driver-assistance systems

Pedestrian detection based on modified dynamic background using gaussian mixture models and hog-svm detection

Adaptive Hysteresis Thresholding Based Pedestrian Detection in Nighttime Using A Normal Camera

Pedestrian Detection Based on Center, Temperature, Scale and Ratio Prediction in Thermal Imagery

The Algorithm of Nighttime Pedestrian Detection in Intelligent Surveillance for Renewable Energy Power Stations

A SVM-based Classifier with Shape and Motion Features for a Pedestrian Detection System

A Fast Detection and Recognition Algorithm for Pedestrian at Night Based on Entropy Weight Fast Support Vector Machine

Nighttime Pedestrian Detection Based on Feature Attention and Transformation

An Improved Pedestrian Detection Approach For Cluttered Background In Nighttime

A Pedestrian Detection Method Based on Hierarchical Tree Cascade Classification at Nighttime