基于改进 ATSS 模型的水稻害虫检测

  • 打印
  • 收藏
收藏成功


打开文本图片集

中图分类号:TP391.41;S435.11 文献标志码:A

Rice pest detection based on improved ATSS model

(1 Collge ofMathematicsandInformatics,SouthChinaAgriculturalUnversity,Guangzhu50642,China;2KeyLaboratoryof

Smart Agricultural Technology in Tropical South China,MinistryofAgriculture andRural Afairs, Guangzhou 51064,China; 3 Zhuhai Wanlida Electric Automation Co.,Ltd., Zhuhai 519085, China; 4 School of Chemistry and Chemical Engineering, South China University of Technology, Guangzhou 510641, China)

Abstract: 【Objective】 To address issues including data scarcity, low accuracy, and poor real-time performance in rice pest detection,this study aims to construct a specialized dataset and develop an efficient detection method. 【Method】 The Pest5 dataset was built based on insect-atracting light traps. Within the adaptive training sample selection (ATSs) framework, an improved model, PestDet, was proposed. Improvements included: A combined data augmentation strategy and anchor optimization were adopted to enhance sample diversity and target matching capability; GHM-Cand DIoU were used as the clasification and regression losses, respectively,to improve robustness and localization accuracy; Inflated convolutions were introduced to reconstruct the feature pyramid for enhanced multi-scale feature perception; The detection head architecture was simplified and the coordinate atention (CA) mechanism was embedded to accelerate inference and strengthen key information extraction. 【Result】 PestDet achieved a detection mean average precision (mAP) of 92.0% and frame per second (FPS) of 40.2 on the Pest5 dataset, surpassing the original ATSS by 7.0 percentage points and 7.O respectively, and outperformed other mainstream models.【Conclusion】 PestDet demonstrates high accuracy with high efficiency, enabling eective identification of rice pests in complex backgrounds, and provides technical support for intelligent pest monitoring and precision control.

Keywords: Rice pest; Object detection; Data augmentation; Attention mechanism; ATSS

水稻作为我国关键的粮食作物,其生产的稳定性直接影响国家粮食安全和民生稳定。(剩余13222字)

monitor