Evaluation of Semantic Segmentation Performance for a Multimodal Roadside Vehicle Detection System on the Edge
Discretely monitoring traffic systems and tracking payloads on vehicle targets can be challenging when traversal occurs off main roads where overhead traffic cameras are not present. This work proposes a portable roadside vehicle detection system as part of a solution for tracking traffic along any...
Saved in:
Main Authors: | , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2025-01-01
|
Series: | Sensors |
Subjects: | |
Online Access: | https://www.mdpi.com/1424-8220/25/2/370 |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
_version_ | 1832587499647533056 |
---|---|
author | Lauren Ervin Max Eastepp Mason McVicker Kenneth Ricks |
author_facet | Lauren Ervin Max Eastepp Mason McVicker Kenneth Ricks |
author_sort | Lauren Ervin |
collection | DOAJ |
description | Discretely monitoring traffic systems and tracking payloads on vehicle targets can be challenging when traversal occurs off main roads where overhead traffic cameras are not present. This work proposes a portable roadside vehicle detection system as part of a solution for tracking traffic along any path. Training semantic segmentation networks to automatically detect specific types of vehicles while ignoring others will allow the user to track payloads present only on certain vehicles of interest, such as train cars or semi-trucks. Different vision sensors offer varying advantages for detecting targets in changing environments and weather conditions. To analyze the benefits of both, corresponding LiDAR and camera data were collected at multiple roadside sites and then trained on separate semantic segmentation networks for object detection. A custom CNN architecture was built to handle highly asymmetric LiDAR data, and a network inspired by DeepLabV3+ was used for camera data. The performance of both networks was evaluated, and showed comparable accuracy. Inferences run on embedded platforms showed real-time execution matching the performance on the training hardware for edge deployments anywhere. Both camera and LiDAR semantic segmentation networks were successful in identifying vehicles of interest from the proposed viewpoint. These highly accurate vehicle detection networks can pair with a tracking mechanism to establish a non-intrusive roadside detection system. |
format | Article |
id | doaj-art-689d1d1a494b43969f3437ef95ceab87 |
institution | Kabale University |
issn | 1424-8220 |
language | English |
publishDate | 2025-01-01 |
publisher | MDPI AG |
record_format | Article |
series | Sensors |
spelling | doaj-art-689d1d1a494b43969f3437ef95ceab872025-01-24T13:48:40ZengMDPI AGSensors1424-82202025-01-0125237010.3390/s25020370Evaluation of Semantic Segmentation Performance for a Multimodal Roadside Vehicle Detection System on the EdgeLauren Ervin0Max Eastepp1Mason McVicker2Kenneth Ricks3Electrical and Computer Engineering Department, The University of Alabama, Tuscaloosa, AL 35487, USAElectrical and Computer Engineering Department, The University of Alabama, Tuscaloosa, AL 35487, USAElectrical and Computer Engineering Department, The University of Alabama, Tuscaloosa, AL 35487, USAElectrical and Computer Engineering Department, The University of Alabama, Tuscaloosa, AL 35487, USADiscretely monitoring traffic systems and tracking payloads on vehicle targets can be challenging when traversal occurs off main roads where overhead traffic cameras are not present. This work proposes a portable roadside vehicle detection system as part of a solution for tracking traffic along any path. Training semantic segmentation networks to automatically detect specific types of vehicles while ignoring others will allow the user to track payloads present only on certain vehicles of interest, such as train cars or semi-trucks. Different vision sensors offer varying advantages for detecting targets in changing environments and weather conditions. To analyze the benefits of both, corresponding LiDAR and camera data were collected at multiple roadside sites and then trained on separate semantic segmentation networks for object detection. A custom CNN architecture was built to handle highly asymmetric LiDAR data, and a network inspired by DeepLabV3+ was used for camera data. The performance of both networks was evaluated, and showed comparable accuracy. Inferences run on embedded platforms showed real-time execution matching the performance on the training hardware for edge deployments anywhere. Both camera and LiDAR semantic segmentation networks were successful in identifying vehicles of interest from the proposed viewpoint. These highly accurate vehicle detection networks can pair with a tracking mechanism to establish a non-intrusive roadside detection system.https://www.mdpi.com/1424-8220/25/2/370semantic segmentationCNNLiDARcameraroadside detection systeminference |
spellingShingle | Lauren Ervin Max Eastepp Mason McVicker Kenneth Ricks Evaluation of Semantic Segmentation Performance for a Multimodal Roadside Vehicle Detection System on the Edge Sensors semantic segmentation CNN LiDAR camera roadside detection system inference |
title | Evaluation of Semantic Segmentation Performance for a Multimodal Roadside Vehicle Detection System on the Edge |
title_full | Evaluation of Semantic Segmentation Performance for a Multimodal Roadside Vehicle Detection System on the Edge |
title_fullStr | Evaluation of Semantic Segmentation Performance for a Multimodal Roadside Vehicle Detection System on the Edge |
title_full_unstemmed | Evaluation of Semantic Segmentation Performance for a Multimodal Roadside Vehicle Detection System on the Edge |
title_short | Evaluation of Semantic Segmentation Performance for a Multimodal Roadside Vehicle Detection System on the Edge |
title_sort | evaluation of semantic segmentation performance for a multimodal roadside vehicle detection system on the edge |
topic | semantic segmentation CNN LiDAR camera roadside detection system inference |
url | https://www.mdpi.com/1424-8220/25/2/370 |
work_keys_str_mv | AT laurenervin evaluationofsemanticsegmentationperformanceforamultimodalroadsidevehicledetectionsystemontheedge AT maxeastepp evaluationofsemanticsegmentationperformanceforamultimodalroadsidevehicledetectionsystemontheedge AT masonmcvicker evaluationofsemanticsegmentationperformanceforamultimodalroadsidevehicledetectionsystemontheedge AT kennethricks evaluationofsemanticsegmentationperformanceforamultimodalroadsidevehicledetectionsystemontheedge |