Distorted Vehicle Detection and Distance Estimation by Metric Learning-Based SSD

Authors

Fanghui Zhang^{1, 2, †,} , Yi Jin^{3, †}, Shichao Kan^{1, 2,} , Linna Zhang⁴, Yigang Cen^{1, 2, *}, Wen Jin⁵

¹Institute of Information Science, Beijing Jiaotong University, Beijing, 100044, China

²Beijing Key Laboratory of Advanced Information Science and Network Technology, Beijing, 100044, China

³School of Computer and Information Technology, Beijing Jiaotong University, Beijing, 100044, China

⁴School of Mechanical engineering, Guizhou University, Guiyang, 550025, China

⁵Jiangsu Jinhai Star Navigation Technology Co., Ltd., Zhenjiang, 212002, China

^†

The first two authors (Fanghui Zhang and Yi Jin) contribute equally.

^*Corresponding author. Email: ygcen@bjtu.edu.cn

DOI

10.2991/ijcis.d.210419.001How to use a DOI?

Keywords

Object detection; Vehicle distance estimation; Metric learning; Scalable overlapping partition-pooling

Abstract

Object detection and distance estimation based on videos are important issues in advanced driver-sssistant system (ADAS). In practice, fisheye cameras are widely used to capture images with a large field of view, which will produce distorted image frames. But most of the object detection algorithms were designed for the nonfisheye camera videos without distortion, which is not suitable for the application of ADAS since one always expects the panorama stitching and object detection system should share one set of cameras. The research of vehicle detection based on fisheye cameras is relatively rare. In this paper, vehicle detection and distance estimation based on fisheye cameras are studied. First, a multi-scale partition preprocessing is proposed, which can enlarge the size of small targets to improve the detection accuracy of small targets. Second, parameters learned from the public datasets without distortion is transferred to our fisheye video dataset. Then metric learning-based single shot multibox detector (MLSSD) is proposed to improve the accuracy of distorted vehicle detection. Combining metric learning and SSD network, MLSSD can significantly reduce the missing and false detection rates. Moreover, a scalable overlapping partition pooling method is proposed to explore the relations among the adjacent features in a feature map. Finally, the distance between the driving vehicle and vehicles around this vehicle is estimated based on the object detection results by the method of marker points. Experimental results show that our proposed MLSSD network significantly outperforms other networks for distorted object detection.

Copyright

© 2021 The Authors. Published by Atlantis Press B.V.

Open Access

This is an open access article distributed under the CC BY-NC 4.0 license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)
View full text (HTML)

TY  - JOUR
AU  - Fanghui Zhang
AU  - Yi Jin
AU  - Shichao Kan
AU  - Linna Zhang
AU  - Yigang Cen
AU  - Wen Jin
PY  - 2021
DA  - 2021/04/28
TI  - Distorted Vehicle Detection and Distance Estimation by Metric Learning-Based SSD
JO  - International Journal of Computational Intelligence Systems
SP  - 1426
EP  - 1437
VL  - 14
IS  - 1
SN  - 1875-6883
UR  - https://doi.org/10.2991/ijcis.d.210419.001
DO  - 10.2991/ijcis.d.210419.001
ID  - Zhang2021
ER  -