Design and Implementation of an Intelligent Pest Status Monitoring System for Farmland
This study proposes an intelligent agricultural pest monitoring system that integrates mechanical control with deep learning to address issues in traditional systems, such as pest accumulation interference, image contrast degradation under complex lighting, and poor balance between model accuracy an...
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2025-05-01
|
| Series: | Agronomy |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2073-4395/15/5/1214 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1849711301310283776 |
|---|---|
| author | Xinyu Yuan Zeshen He Caojun Huang |
| author_facet | Xinyu Yuan Zeshen He Caojun Huang |
| author_sort | Xinyu Yuan |
| collection | DOAJ |
| description | This study proposes an intelligent agricultural pest monitoring system that integrates mechanical control with deep learning to address issues in traditional systems, such as pest accumulation interference, image contrast degradation under complex lighting, and poor balance between model accuracy and real-time performance. A three-axis coordinated separation device is employed, achieving a 92.41% single-attempt separation rate and 98.12% after three retries. Image preprocessing combines the Multi-Scale Retinex with Color Preservation (MSRCP) algorithm and bilateral filtering to enhance illumination correction and reduce noise. For overlapping pest detection, EfficientNetv2-S replaces the YOLOv5s backbone and is combined with an Adaptive Feature Pyramid Network (AFPN), achieving 95.72% detection accuracy, 94.04% mAP, and 127 FPS. For pest species recognition, the model incorporates a Squeeze-and-Excitation (SE) attention module and α-CIoU loss function, reaching 91.30% precision on 3428 field images. Deployed on an NVIDIA Jetson Nano, the system demonstrates a detection time of 0.3 s, 89.64% recall, 86.78% precision, and 1.136 s image transmission delay, offering a reliable solution for real-time pest monitoring in complex field environments. |
| format | Article |
| id | doaj-art-cc2c49c121d0408f9dc3a8c1eba077a1 |
| institution | DOAJ |
| issn | 2073-4395 |
| language | English |
| publishDate | 2025-05-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Agronomy |
| spelling | doaj-art-cc2c49c121d0408f9dc3a8c1eba077a12025-08-20T03:14:39ZengMDPI AGAgronomy2073-43952025-05-01155121410.3390/agronomy15051214Design and Implementation of an Intelligent Pest Status Monitoring System for FarmlandXinyu Yuan0Zeshen He1Caojun Huang2College of Engineering, Heilongjiang Bayi Agricultural University, Daqing 163319, ChinaCollege of Information and Electrical Engineering, Heilongjiang Bayi Agricultural University, Daqing 163319, ChinaCollege of Information and Electrical Engineering, Heilongjiang Bayi Agricultural University, Daqing 163319, ChinaThis study proposes an intelligent agricultural pest monitoring system that integrates mechanical control with deep learning to address issues in traditional systems, such as pest accumulation interference, image contrast degradation under complex lighting, and poor balance between model accuracy and real-time performance. A three-axis coordinated separation device is employed, achieving a 92.41% single-attempt separation rate and 98.12% after three retries. Image preprocessing combines the Multi-Scale Retinex with Color Preservation (MSRCP) algorithm and bilateral filtering to enhance illumination correction and reduce noise. For overlapping pest detection, EfficientNetv2-S replaces the YOLOv5s backbone and is combined with an Adaptive Feature Pyramid Network (AFPN), achieving 95.72% detection accuracy, 94.04% mAP, and 127 FPS. For pest species recognition, the model incorporates a Squeeze-and-Excitation (SE) attention module and α-CIoU loss function, reaching 91.30% precision on 3428 field images. Deployed on an NVIDIA Jetson Nano, the system demonstrates a detection time of 0.3 s, 89.64% recall, 86.78% precision, and 1.136 s image transmission delay, offering a reliable solution for real-time pest monitoring in complex field environments.https://www.mdpi.com/2073-4395/15/5/1214pest monitoring systemYOLOv5edge computingmachine visionimage detection |
| spellingShingle | Xinyu Yuan Zeshen He Caojun Huang Design and Implementation of an Intelligent Pest Status Monitoring System for Farmland Agronomy pest monitoring system YOLOv5 edge computing machine vision image detection |
| title | Design and Implementation of an Intelligent Pest Status Monitoring System for Farmland |
| title_full | Design and Implementation of an Intelligent Pest Status Monitoring System for Farmland |
| title_fullStr | Design and Implementation of an Intelligent Pest Status Monitoring System for Farmland |
| title_full_unstemmed | Design and Implementation of an Intelligent Pest Status Monitoring System for Farmland |
| title_short | Design and Implementation of an Intelligent Pest Status Monitoring System for Farmland |
| title_sort | design and implementation of an intelligent pest status monitoring system for farmland |
| topic | pest monitoring system YOLOv5 edge computing machine vision image detection |
| url | https://www.mdpi.com/2073-4395/15/5/1214 |
| work_keys_str_mv | AT xinyuyuan designandimplementationofanintelligentpeststatusmonitoringsystemforfarmland AT zeshenhe designandimplementationofanintelligentpeststatusmonitoringsystemforfarmland AT caojunhuang designandimplementationofanintelligentpeststatusmonitoringsystemforfarmland |