MSCD-YOLO: A Lightweight Dense Pedestrian Detection Model with Finer-Grained Feature Information Interaction
Pedestrian detection is widely used in real-time surveillance, urban traffic, and other fields. As a crucial direction in pedestrian detection, dense pedestrian detection still faces many unresolved challenges. Existing methods suffer from low detection accuracy, high miss rates, large model paramet...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2025-01-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/25/2/438 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832587479062937600 |
|---|---|
| author | Qiang Liu Zhongmin Li Lei Zhang Jin Deng |
| author_facet | Qiang Liu Zhongmin Li Lei Zhang Jin Deng |
| author_sort | Qiang Liu |
| collection | DOAJ |
| description | Pedestrian detection is widely used in real-time surveillance, urban traffic, and other fields. As a crucial direction in pedestrian detection, dense pedestrian detection still faces many unresolved challenges. Existing methods suffer from low detection accuracy, high miss rates, large model parameters, and poor robustness. In this paper, to address these issues, we propose a lightweight dense pedestrian detection model with finer-grained feature information interaction called MSCD-YOLO, which can achieve high accuracy, high performance and robustness with only a small number of parameters. In our model, the light-weight backbone network MobileViT is used to reduce the number of parameters while efficiently extracting both local and global features; the SCNeck neck network is designed to fuse the extracted features without losing information; and the DEHead detection head is utilized for multi-scale feature fusion to detect the targets. To demonstrate the effectiveness of our model, we conducted tests on the highly challenging dense pedestrian detection datasets Crowdhuman and Widerperson. Compared to the baseline model YOLOv8n, MSCD-YOLO achieved a 4.6% and 1.8% improvement in mAP@0.5, and a 5.3% and 2.6% improvement in mAP@0.5:0.95 on the Crowdhuman and Widerperson datasets, respectively. The experimental results show that under the same experimental conditions, MSCD-YOLO significantly outperforms the original model in terms of detection accuracy, efficiency, and model complexity. |
| format | Article |
| id | doaj-art-17cc77174e054364afe6735502e687a5 |
| institution | Kabale University |
| issn | 1424-8220 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj-art-17cc77174e054364afe6735502e687a52025-01-24T13:48:56ZengMDPI AGSensors1424-82202025-01-0125243810.3390/s25020438MSCD-YOLO: A Lightweight Dense Pedestrian Detection Model with Finer-Grained Feature Information InteractionQiang Liu0Zhongmin Li1Lei Zhang2Jin Deng3School of Information Engineering, Nanchang Hangkong University, Nanchang 330063, ChinaSchool of Information Engineering, Nanchang Hangkong University, Nanchang 330063, ChinaSchool of Information Engineering, Nanchang Hangkong University, Nanchang 330063, ChinaSchool of Information Engineering, Nanchang Hangkong University, Nanchang 330063, ChinaPedestrian detection is widely used in real-time surveillance, urban traffic, and other fields. As a crucial direction in pedestrian detection, dense pedestrian detection still faces many unresolved challenges. Existing methods suffer from low detection accuracy, high miss rates, large model parameters, and poor robustness. In this paper, to address these issues, we propose a lightweight dense pedestrian detection model with finer-grained feature information interaction called MSCD-YOLO, which can achieve high accuracy, high performance and robustness with only a small number of parameters. In our model, the light-weight backbone network MobileViT is used to reduce the number of parameters while efficiently extracting both local and global features; the SCNeck neck network is designed to fuse the extracted features without losing information; and the DEHead detection head is utilized for multi-scale feature fusion to detect the targets. To demonstrate the effectiveness of our model, we conducted tests on the highly challenging dense pedestrian detection datasets Crowdhuman and Widerperson. Compared to the baseline model YOLOv8n, MSCD-YOLO achieved a 4.6% and 1.8% improvement in mAP@0.5, and a 5.3% and 2.6% improvement in mAP@0.5:0.95 on the Crowdhuman and Widerperson datasets, respectively. The experimental results show that under the same experimental conditions, MSCD-YOLO significantly outperforms the original model in terms of detection accuracy, efficiency, and model complexity.https://www.mdpi.com/1424-8220/25/2/438deep learningdense pedestrian detectionYOLOv8nMobile-ViTSCNeckDEHead |
| spellingShingle | Qiang Liu Zhongmin Li Lei Zhang Jin Deng MSCD-YOLO: A Lightweight Dense Pedestrian Detection Model with Finer-Grained Feature Information Interaction Sensors deep learning dense pedestrian detection YOLOv8n Mobile-ViT SCNeck DEHead |
| title | MSCD-YOLO: A Lightweight Dense Pedestrian Detection Model with Finer-Grained Feature Information Interaction |
| title_full | MSCD-YOLO: A Lightweight Dense Pedestrian Detection Model with Finer-Grained Feature Information Interaction |
| title_fullStr | MSCD-YOLO: A Lightweight Dense Pedestrian Detection Model with Finer-Grained Feature Information Interaction |
| title_full_unstemmed | MSCD-YOLO: A Lightweight Dense Pedestrian Detection Model with Finer-Grained Feature Information Interaction |
| title_short | MSCD-YOLO: A Lightweight Dense Pedestrian Detection Model with Finer-Grained Feature Information Interaction |
| title_sort | mscd yolo a lightweight dense pedestrian detection model with finer grained feature information interaction |
| topic | deep learning dense pedestrian detection YOLOv8n Mobile-ViT SCNeck DEHead |
| url | https://www.mdpi.com/1424-8220/25/2/438 |
| work_keys_str_mv | AT qiangliu mscdyoloalightweightdensepedestriandetectionmodelwithfinergrainedfeatureinformationinteraction AT zhongminli mscdyoloalightweightdensepedestriandetectionmodelwithfinergrainedfeatureinformationinteraction AT leizhang mscdyoloalightweightdensepedestriandetectionmodelwithfinergrainedfeatureinformationinteraction AT jindeng mscdyoloalightweightdensepedestriandetectionmodelwithfinergrainedfeatureinformationinteraction |