HMD-Net: A Vehicle Hazmat Marker Detection Benchmark
Vehicles carrying hazardous material (hazmat) are severe threats to the safety of highway transportation, and a model that can automatically recognize hazmat markers installed or attached on vehicles is essential for intelligent management systems. However, there is still no public dataset for bench...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-03-01
|
| Series: | Entropy |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1099-4300/24/4/466 |
| _version_ | 1827628924534784000 |
|---|---|
| author | Lei Jia Jianzhu Wang Tianyuan Wang Xiaobao Li Haomin Yu Qingyong Li |
| author_facet | Lei Jia Jianzhu Wang Tianyuan Wang Xiaobao Li Haomin Yu Qingyong Li |
| author_sort | Lei Jia |
| collection | DOAJ |
| description | Vehicles carrying hazardous material (hazmat) are severe threats to the safety of highway transportation, and a model that can automatically recognize hazmat markers installed or attached on vehicles is essential for intelligent management systems. However, there is still no public dataset for benchmarking the task of hazmat marker detection. To this end, this paper releases a large-scale vehicle hazmat marker dataset named VisInt-VHM, which includes 10,000 images with a total of 20,023 hazmat markers captured under different environmental conditions from a real-world highway. Meanwhile, we provide an compact hazmat marker detection network named HMD-Net, which utilizes a revised lightweight backbone and is further compressed by channel pruning. As a consequence, the trained-model can be efficiently deployed on a resource-restricted edge device. Experimental results demonstrate that compared with some established methods such as YOLOv3, YOLOv4, their lightweight versions and popular lightweight models, HMD-Net can achieve a better trade-off between the detection accuracy and the inference speed. |
| first_indexed | 2024-03-09T13:42:30Z |
| format | Article |
| id | doaj.art-a369e59f40b74ed3970c175a1b808705 |
| institution | Directory Open Access Journal |
| issn | 1099-4300 |
| language | English |
| last_indexed | 2024-03-09T13:42:30Z |
| publishDate | 2022-03-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Entropy |
| spelling | doaj.art-a369e59f40b74ed3970c175a1b8087052023-11-30T21:05:05ZengMDPI AGEntropy1099-43002022-03-0124446610.3390/e24040466HMD-Net: A Vehicle Hazmat Marker Detection BenchmarkLei Jia0Jianzhu Wang1Tianyuan Wang2Xiaobao Li3Haomin Yu4Qingyong Li5Beijing Key Lab of Traffic Data Analysis and Mining, Beijing Jiaotong University, Beijing 100044, ChinaBeijing Key Lab of Traffic Data Analysis and Mining, Beijing Jiaotong University, Beijing 100044, ChinaShenzhen Urban Transport Planning Center Co., Ltd., Shenzhen 518000, ChinaBeijing Key Lab of Traffic Data Analysis and Mining, Beijing Jiaotong University, Beijing 100044, ChinaBeijing Key Lab of Traffic Data Analysis and Mining, Beijing Jiaotong University, Beijing 100044, ChinaBeijing Key Lab of Traffic Data Analysis and Mining, Beijing Jiaotong University, Beijing 100044, ChinaVehicles carrying hazardous material (hazmat) are severe threats to the safety of highway transportation, and a model that can automatically recognize hazmat markers installed or attached on vehicles is essential for intelligent management systems. However, there is still no public dataset for benchmarking the task of hazmat marker detection. To this end, this paper releases a large-scale vehicle hazmat marker dataset named VisInt-VHM, which includes 10,000 images with a total of 20,023 hazmat markers captured under different environmental conditions from a real-world highway. Meanwhile, we provide an compact hazmat marker detection network named HMD-Net, which utilizes a revised lightweight backbone and is further compressed by channel pruning. As a consequence, the trained-model can be efficiently deployed on a resource-restricted edge device. Experimental results demonstrate that compared with some established methods such as YOLOv3, YOLOv4, their lightweight versions and popular lightweight models, HMD-Net can achieve a better trade-off between the detection accuracy and the inference speed.https://www.mdpi.com/1099-4300/24/4/466vehicles for hazmat transportationhazmat marker detectionsparse regularizationchannel pruningYOLOv5MobileNet |
| spellingShingle | Lei Jia Jianzhu Wang Tianyuan Wang Xiaobao Li Haomin Yu Qingyong Li HMD-Net: A Vehicle Hazmat Marker Detection Benchmark Entropy vehicles for hazmat transportation hazmat marker detection sparse regularization channel pruning YOLOv5 MobileNet |
| title | HMD-Net: A Vehicle Hazmat Marker Detection Benchmark |
| title_full | HMD-Net: A Vehicle Hazmat Marker Detection Benchmark |
| title_fullStr | HMD-Net: A Vehicle Hazmat Marker Detection Benchmark |
| title_full_unstemmed | HMD-Net: A Vehicle Hazmat Marker Detection Benchmark |
| title_short | HMD-Net: A Vehicle Hazmat Marker Detection Benchmark |
| title_sort | hmd net a vehicle hazmat marker detection benchmark |
| topic | vehicles for hazmat transportation hazmat marker detection sparse regularization channel pruning YOLOv5 MobileNet |
| url | https://www.mdpi.com/1099-4300/24/4/466 |
| work_keys_str_mv | AT leijia hmdnetavehiclehazmatmarkerdetectionbenchmark AT jianzhuwang hmdnetavehiclehazmatmarkerdetectionbenchmark AT tianyuanwang hmdnetavehiclehazmatmarkerdetectionbenchmark AT xiaobaoli hmdnetavehiclehazmatmarkerdetectionbenchmark AT haominyu hmdnetavehiclehazmatmarkerdetectionbenchmark AT qingyongli hmdnetavehiclehazmatmarkerdetectionbenchmark |