Traffic Accident Detection Based on Deformable Frustum Proposal and Adaptive Space Segmentation

Peng Chen,Yongxiang Tian,Ziyao Xiao,Weiwei Zhang

doi:10.32604/cmes.2022.016632

Abstract

Road accident detection plays an important role in abnormal scene reconstruction for Intelligent Transportation Systems and abnormal events warning for autonomous driving. This paper presents a novel 3D object detector and adaptive space partitioning algorithm to infer traffic accidents quantitatively. Using 2D region proposals in an RGB image, this method generates deformable frustums based on point cloud for each 2D region proposal and then frustum-wisely extracts features based on the farthest point sampling network (FPS-Net) and feature extraction network (FE-Net). Subsequently, the encoder-decoder network (ED-Net) implements 3D-oriented bounding box (OBB) regression. Meanwhile, the adaptive least square regression (ALSR) method is proposed to split 3D OBB. Finally, the reduced OBB intersection test is carried out to detect traffic accidents via separating surface theorem (SST). In the experiments of KITTI benchmark, our proposed 3D object detector outperforms other state-of-the-art methods. Meanwhile, collision detection algorithm achieves the satisfactory performance of 91.8% accuracy on our SHTA dataset.

Highlights

Vision-based object detection algorithms have been extensively exploited for traffic accident detection, which generates object location, motion information, and object category
(2) This study proposes the adaptive least square regression model to split 3D oriented bounding box (OBB), which is followed by separating surface theorem to infer traffic accidents
4.2 Evaluation of 3D OBB Regression 4.2.1 Comparing with SOTA

Summary

Introduction

Vision-based object detection algorithms have been extensively exploited for traffic accident detection, which generates object location, motion information, and object category. Vision-based researches on traffic accident detection [1,2] can achieve 2D bounding box regression and classification prediction from monocular images, and utilize trajectory information to identify accidents. CMES, 2022, vol.130, no.1 detection followed by Centroid Tracking algorithm from RGB images, and capitalized on speed and trajectory anomalies to infer traffic accidents. These methods achieved poor accuracy and recall when objects are truncated, especially occluded. These works only provide texture information but lack depth information, which makes it difficult to describe objects for collision detection in 3D real-world scenes. 3D object detection method becomes the main research issue on traffic accident detection

Methods

Results

Conclusion