Pedestrian Detection Based on Feature Enhancement in Complex Scenes
Pedestrian detection has always been a difficult and hot spot in computer vision research. At the same time, pedestrian detection technology plays an important role in many applications, such as intelligent transportation and security monitoring. In complex scenes, pedestrian detection often faces s...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2024-01-01
|
| Series: | Algorithms |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1999-4893/17/1/39 |
| _version_ | 1797340214918643712 |
|---|---|
| author | Jiao Su Yi An Jialin Wu Kai Zhang |
| author_facet | Jiao Su Yi An Jialin Wu Kai Zhang |
| author_sort | Jiao Su |
| collection | DOAJ |
| description | Pedestrian detection has always been a difficult and hot spot in computer vision research. At the same time, pedestrian detection technology plays an important role in many applications, such as intelligent transportation and security monitoring. In complex scenes, pedestrian detection often faces some challenges, such as low detection accuracy and misdetection due to small target sizes and scale variations. To solve these problems, this paper proposes a pedestrian detection network PT-YOLO based on the YOLOv5. The pedestrian detection network PT-YOLO consists of the YOLOv5 network, the squeeze-and-excitation module (SE), the weighted bi-directional feature pyramid module (BiFPN), the coordinate convolution (coordconv) module and the wise intersection over union loss function (WIoU). The SE module in the backbone allows it to focus on the important features of pedestrians and improves accuracy. The weighted BiFPN module enhances the fusion of multi-scale pedestrian features and information transfer, which can improve fusion efficiency. The prediction head design uses the WIoU loss function to reduce the regression error. The coordconv module allows the network to better perceive the location information in the feature map. The experimental results show that the pedestrian detection network PT-YOLO is more accurate compared with other target detection methods in pedestrian detection and can effectively accomplish the task of pedestrian detection in complex scenes. |
| first_indexed | 2024-03-08T09:59:47Z |
| format | Article |
| id | doaj.art-5b7c3efa34f6453bb810e7c7dcebbb2e |
| institution | Directory Open Access Journal |
| issn | 1999-4893 |
| language | English |
| last_indexed | 2024-03-08T09:59:47Z |
| publishDate | 2024-01-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Algorithms |
| spelling | doaj.art-5b7c3efa34f6453bb810e7c7dcebbb2e2024-01-29T13:41:35ZengMDPI AGAlgorithms1999-48932024-01-011713910.3390/a17010039Pedestrian Detection Based on Feature Enhancement in Complex ScenesJiao Su0Yi An1Jialin Wu2Kai Zhang3School of Electrical Engineering, Xinjiang University, Urumqi 830000, ChinaSchool of Electrical Engineering, Xinjiang University, Urumqi 830000, ChinaSchool of Control Science and Engineering, Dalian University of Technology, Dalian 116023, ChinaSchool of Electrical Engineering, Xinjiang University, Urumqi 830000, ChinaPedestrian detection has always been a difficult and hot spot in computer vision research. At the same time, pedestrian detection technology plays an important role in many applications, such as intelligent transportation and security monitoring. In complex scenes, pedestrian detection often faces some challenges, such as low detection accuracy and misdetection due to small target sizes and scale variations. To solve these problems, this paper proposes a pedestrian detection network PT-YOLO based on the YOLOv5. The pedestrian detection network PT-YOLO consists of the YOLOv5 network, the squeeze-and-excitation module (SE), the weighted bi-directional feature pyramid module (BiFPN), the coordinate convolution (coordconv) module and the wise intersection over union loss function (WIoU). The SE module in the backbone allows it to focus on the important features of pedestrians and improves accuracy. The weighted BiFPN module enhances the fusion of multi-scale pedestrian features and information transfer, which can improve fusion efficiency. The prediction head design uses the WIoU loss function to reduce the regression error. The coordconv module allows the network to better perceive the location information in the feature map. The experimental results show that the pedestrian detection network PT-YOLO is more accurate compared with other target detection methods in pedestrian detection and can effectively accomplish the task of pedestrian detection in complex scenes.https://www.mdpi.com/1999-4893/17/1/39pedestrian detectioncoordinate convolutionfeature pyramidSE attention mechanismfeature enhancement |
| spellingShingle | Jiao Su Yi An Jialin Wu Kai Zhang Pedestrian Detection Based on Feature Enhancement in Complex Scenes Algorithms pedestrian detection coordinate convolution feature pyramid SE attention mechanism feature enhancement |
| title | Pedestrian Detection Based on Feature Enhancement in Complex Scenes |
| title_full | Pedestrian Detection Based on Feature Enhancement in Complex Scenes |
| title_fullStr | Pedestrian Detection Based on Feature Enhancement in Complex Scenes |
| title_full_unstemmed | Pedestrian Detection Based on Feature Enhancement in Complex Scenes |
| title_short | Pedestrian Detection Based on Feature Enhancement in Complex Scenes |
| title_sort | pedestrian detection based on feature enhancement in complex scenes |
| topic | pedestrian detection coordinate convolution feature pyramid SE attention mechanism feature enhancement |
| url | https://www.mdpi.com/1999-4893/17/1/39 |
| work_keys_str_mv | AT jiaosu pedestriandetectionbasedonfeatureenhancementincomplexscenes AT yian pedestriandetectionbasedonfeatureenhancementincomplexscenes AT jialinwu pedestriandetectionbasedonfeatureenhancementincomplexscenes AT kaizhang pedestriandetectionbasedonfeatureenhancementincomplexscenes |