LPCF-YOLO: A YOLO-Based Lightweight Algorithm for Pedestrian Anomaly Detection with Parallel Cross-Fusion

To address the issue of high complexity in current pedestrian anomaly detection network models, which hinders real-world deployment, this paper proposes a lightweight anomaly detection network called LPCF-YOLO (Lightweight Parallel Cross-Fusion YOLO) based on the YOLOv8n model. Firstly, the FPC-F (F...

Full description

Saved in:
Bibliographic Details
Main Authors: Peiyi Jia, Hu Sheng, Shijie Jia
Format: Article
Language:English
Published: MDPI AG 2025-04-01
Series:Sensors
Subjects:
Online Access:https://www.mdpi.com/1424-8220/25/9/2752
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850275300095557632
author Peiyi Jia
Hu Sheng
Shijie Jia
author_facet Peiyi Jia
Hu Sheng
Shijie Jia
author_sort Peiyi Jia
collection DOAJ
description To address the issue of high complexity in current pedestrian anomaly detection network models, which hinders real-world deployment, this paper proposes a lightweight anomaly detection network called LPCF-YOLO (Lightweight Parallel Cross-Fusion YOLO) based on the YOLOv8n model. Firstly, the FPC-F (Fast Parallel Cross-Fusion) module, which incorporates PConv, and the S-EMCP (Space-efficient Merging Convolution Pooling) module are designed in the backbone network to replace C2F and SPPF at various scale branches. Additionally, an ADown module is introduced in the third layer to reduce the computational cost. In the neck network, a Lightweight High-level Screening Feature Pyramid Network (L-HSFPN) is designed to replace the PAFPN structure. Furthermore, the Wise-IoU loss function is employed to enhance the model’s localization performance and generalization ability. The experimental results in the UCSD-Ped1 and UCSD-Ped2 datasets show that, compared to YOLOv8n, the proposed approach reduces parameters by 30.33% and FLOPs by 79.01%, achieving 2.09 M parameters and 1.7 G FLOPs; it also results in a 179.62% increase in FPS to 43.9. Meanwhile, the mean average precision (mAP@0.5) is either maintained (in the UCSD-Ped2 dataset) or slightly improved (in the UCSD-Ped1 dataset).
format Article
id doaj-art-a48dddb0d8f54868b5a0c316fb2dc50b
institution OA Journals
issn 1424-8220
language English
publishDate 2025-04-01
publisher MDPI AG
record_format Article
series Sensors
spelling doaj-art-a48dddb0d8f54868b5a0c316fb2dc50b2025-08-20T01:50:46ZengMDPI AGSensors1424-82202025-04-01259275210.3390/s25092752LPCF-YOLO: A YOLO-Based Lightweight Algorithm for Pedestrian Anomaly Detection with Parallel Cross-FusionPeiyi Jia0Hu Sheng1Shijie Jia2School of Rail Intelligent Engineering, Dalian Jiaotong University, Dalian 116028, ChinaSchool of Rail Intelligent Engineering, Dalian Jiaotong University, Dalian 116028, ChinaSchool of Rail Intelligent Engineering, Dalian Jiaotong University, Dalian 116028, ChinaTo address the issue of high complexity in current pedestrian anomaly detection network models, which hinders real-world deployment, this paper proposes a lightweight anomaly detection network called LPCF-YOLO (Lightweight Parallel Cross-Fusion YOLO) based on the YOLOv8n model. Firstly, the FPC-F (Fast Parallel Cross-Fusion) module, which incorporates PConv, and the S-EMCP (Space-efficient Merging Convolution Pooling) module are designed in the backbone network to replace C2F and SPPF at various scale branches. Additionally, an ADown module is introduced in the third layer to reduce the computational cost. In the neck network, a Lightweight High-level Screening Feature Pyramid Network (L-HSFPN) is designed to replace the PAFPN structure. Furthermore, the Wise-IoU loss function is employed to enhance the model’s localization performance and generalization ability. The experimental results in the UCSD-Ped1 and UCSD-Ped2 datasets show that, compared to YOLOv8n, the proposed approach reduces parameters by 30.33% and FLOPs by 79.01%, achieving 2.09 M parameters and 1.7 G FLOPs; it also results in a 179.62% increase in FPS to 43.9. Meanwhile, the mean average precision (mAP@0.5) is either maintained (in the UCSD-Ped2 dataset) or slightly improved (in the UCSD-Ped1 dataset).https://www.mdpi.com/1424-8220/25/9/2752LPCF-YOLOlightweight feature extractionpedestrian abnormal detectionparallel cross-fusion
spellingShingle Peiyi Jia
Hu Sheng
Shijie Jia
LPCF-YOLO: A YOLO-Based Lightweight Algorithm for Pedestrian Anomaly Detection with Parallel Cross-Fusion
Sensors
LPCF-YOLO
lightweight feature extraction
pedestrian abnormal detection
parallel cross-fusion
title LPCF-YOLO: A YOLO-Based Lightweight Algorithm for Pedestrian Anomaly Detection with Parallel Cross-Fusion
title_full LPCF-YOLO: A YOLO-Based Lightweight Algorithm for Pedestrian Anomaly Detection with Parallel Cross-Fusion
title_fullStr LPCF-YOLO: A YOLO-Based Lightweight Algorithm for Pedestrian Anomaly Detection with Parallel Cross-Fusion
title_full_unstemmed LPCF-YOLO: A YOLO-Based Lightweight Algorithm for Pedestrian Anomaly Detection with Parallel Cross-Fusion
title_short LPCF-YOLO: A YOLO-Based Lightweight Algorithm for Pedestrian Anomaly Detection with Parallel Cross-Fusion
title_sort lpcf yolo a yolo based lightweight algorithm for pedestrian anomaly detection with parallel cross fusion
topic LPCF-YOLO
lightweight feature extraction
pedestrian abnormal detection
parallel cross-fusion
url https://www.mdpi.com/1424-8220/25/9/2752
work_keys_str_mv AT peiyijia lpcfyoloayolobasedlightweightalgorithmforpedestriananomalydetectionwithparallelcrossfusion
AT husheng lpcfyoloayolobasedlightweightalgorithmforpedestriananomalydetectionwithparallelcrossfusion
AT shijiejia lpcfyoloayolobasedlightweightalgorithmforpedestriananomalydetectionwithparallelcrossfusion