GC-Faster RCNN: The Object Detection Algorithm for Agricultural Pests Based on Improved Hybrid Attention Mechanism

文献类型： 外文期刊

第一作者： Guan, Bolun

作者： Guan, Bolun;Zhu, Jingbo;Kong, Juanjuan;Dong, Wei;Wu, Yaqian

作者机构：

关键词： object detection; hybrid attention mechanism; optimization function; agricultural pests

期刊名称：PLANTS-BASEL （ 影响因子：4.1； 五年影响因子：4.5 ）

ISSN： 2223-7747

年卷期： 2025 年 14 卷 7 期

页码：

收录情况： SCI

摘要： Pest infestations remain a critical threat to global agriculture, significantly compromising crop yield and quality. While accurate pest detection forms the foundation of precision pest management, current approaches face two primary challenges: (1) the scarcity of comprehensive multi-scale, multi-category pest datasets and (2) performance limitations in detection models caused by substantial target scale variations and high inter-class morphological similarity. To address these issues, we present three key contributions: First, we introduce Insect25-a novel agricultural pest detection dataset containing 25 distinct pest categories, comprising 18,349 high-resolution images. This dataset specifically addresses scale diversity through multi-resolution acquisition protocols, significantly enriching feature distribution for robust model training. Second, we propose GC-Faster RCNN, an enhanced detection framework integrating a hybrid attention mechanism that synergistically combines channel-wise correlations and spatial dependencies. This dual attention design enables more discriminative feature extraction, which is particularly effective for distinguishing morphologically similar pest species. Third, we implement an optimized training strategy featuring a cosine annealing scheduler with linear warm-up, accelerating model convergence while maintaining training stability. Experiments have shown that compared with the original Faster RCNN model, GC-Faster RCNN has improved the average accuracy mAP0.5 on the Insect25 dataset by 4.5 percentage points, and mAP0.75 by 20.4 percentage points, mAP0.5:0.95 increased by 20.8 percentage points, and the recall rate increased by 16.6 percentage points. In addition, experiments have also shown that the GC-Faster RCNN detection method can reduce interference from multiple scales and high similarity between categories, improving detection performance.

分类号：

  • 相关文献
作者其他论文 更多>>
  • Efficient Triple Attention and AttentionMix: A Novel Network for Fine-Grained Crop Disease Classification

    作者：Zhang, Yanqi;Zhang, Ning;Chai, Xiujuan;Zhu, Jingbo;Dong, Wei;Sun, Tan

    关键词：crop pests and diseases; CNNs; channel attention; spatial attention; data augmentation

  • Assessment of recombinant Lactobacillus vector-based expressing IFITM3 protein anti-PRV infection in mice.

    作者：Zheng, Fan;Zhu, Maokun;Ran, Hengdong;Bai, Binghan;Feng, Simeng;Dong, Wei;Yuan, Xiaomin;He, Jiayi;Li, Bin;Wen, Lixin

    关键词：Recombinant lactic acid bacteria; Interferon-induced transmembrane protein 3; Pseudorabies virus; Antiviral

  • Construction of a Multi-Source, Heterogeneous Rice Disease and Pest Knowledge Graph Based on the MARBC Model

    作者：Li, Chunchun;Yang, Siyi;Liang, Dong;Chen, Peng;Li, Chunchun;Liang, Dong;Chen, Peng;Dong, Wei

    关键词：RDP; MARBC; multi-source heterogeneous; knowledge graph; ontology

  • Genome-Wide Analysis of the APETALA2/Ethylene-Responsive Factor Gene Family in Carthamus tinctorius L.

    作者：Tan, Zheng-Wei;Lu, Dan-Dan;Yu, Yong-Liang;Li, Lei;Xu, Lan-Jie;Dong, Wei;Li, Chun-Ming;Yang, Qing;Liang, Hui-Zhen

    关键词：APETALA2/ethylene-responsive factor; Carthamus tinctorius L; expression pattern; multiple sequence alignment; transcription factor

  • Integrated metabolomic and transcriptomic analyses reveal anthocyanin biosynthesis mechanisms and the regulatory role of LjAN2 in Lonicera japonica

    作者：Tan, Zhengwei;Lu, Dandan;Li, Lei;Yu, Yongliang;Su, Xiaoyu;Sun, Yao;Cao, Yiwen;Li, Chunming;Dong, Wei;Yang, Hongqi;Yang, Qing;An, Sufang;Liang, Huizhen;Tan, Zhengwei;Lu, Dandan;Li, Lei;Yu, Yongliang;Su, Xiaoyu;Sun, Yao;Cao, Yiwen;Li, Chunming;Dong, Wei;Yang, Hongqi;Yang, Qing;An, Sufang;Liang, Huizhen

    关键词：Lonicera japonica; Flower pigmentation; Anthocyanins composition; Anthocyanin synthase; Transcription factors; MYB

  • Fungal diversity notes 1919-2016: taxonomic and phylogenetic contributions to fungal taxa

    作者：Cao, Bin;Wang, Long;Li, Jia-Xin;Han, Xi-Xi;Yang, Wen-Qiang;Wang, Shi-Hui;He, Mao-Qiang;Zhao, Rui-Lin;Cao, Bin;Phurbu, Dorji;Zhao, Rui-Lin;Ralaiveloarisoa, Anna;Liimatainen, Kare;Niskanen, Tuula;Liimatainen, Kare;Niskanen, Tuula;Ramirez-Cruz, Virginia;Cortes-Perez, Alonso;Guzman-Davalos, Laura;Castro-Jauregui, Oscar;Bradshaw, Alexander James;Dentinger, Bryn T. M.;Dentinger, Bryn T. M.;Ramirez-Guillen, Florencia;Villalobos-Arambula, Alma Rosa;da Silva, Paula Santos;Day, Rory;Davoodian, Naveed;Lebel, Teresa;May, Tom W.;Lebel, Teresa;Castellano, Michael;Toome, Merje;Vasey, Jack;Hofer, Katharina;Thangavel, Rajaram;Braithwaite, Mark;Braithwaite, Lewis;Dutta, Arun Kumar;Chattopadhyay, Pinaki;Roy, Niranjan;Tanti, Bhaben;Biswas, Pinky Rani;Roy, Niranjan;Arumugam, Elangovan;Kezo, Kezhocuyi;Kaliyaperumal, Malarvizhi;Murugadoss, Ramesh;Ji, Jing-Xin;Kakishima, Makoto;Cooper, Jerry;Nuytinck, Jorinde;Lagaet, Phaedra;De Lange, Ruben;Verbeken, Annemieke;Tondeleir, Lowie;Nuytinck, Jorinde;Luo, Zong-Long;Wang, Wen-Peng;Zhang, Xian;Li, Hua;Xiong, Yin-Ru;Wu, Na;Shu, Yong-Xin;Zhao, Hai-Jun;Liao, Chun-Fang;Yang, Hong-De;Jayawardena, Ruvishika Shehali;Thongklang, Naritsada;Han, Xi-Xi;Hyde, Kevin D.;Zhang, Xian;Wen, Ting-Chi;Zhang, Xian;Wen, Ting-Chi;Jia, Ao-Li;Fan, Xin-Lei;Jiang, Ning;Jiao, Ning;Zhang, Ying;Manawasinghe, Ishara S.;Li, Hua;Xiong, Yin-Ru;Zhou, Tao;Tan, Qing-Ling;Li, Hua;Xiong, Yin-Ru;Wu, Na;Shu, Yong-Xin;Zhao, Hai-Jun;Liao, Chun-Fang;Yang, Hong-De;Jayawardena, Ruvishika Shehali;Thongklang, Naritsada;Han, Xi-Xi;Hoshino, Tamotsu;Hoshino, Tamotsu;Manz, Cathrin;Hampe, Felix;Zhao, Chang-Lin;Yang, Yang;Zhang, Gui-Qing;Dai, Dong-Qin;Tomsovsky, Michal;Denchev, Teodor T.;Denchev, Cvetomir M.;Denchev, Teodor T.;Denchev, Cvetomir M.;Kemler, Martin;Leveille-Bourret, Etienne;Kemler, Martin;Chen, Yan-Peng;Maharachchikumbura, Sajeewa S. N.;Liu, Jian-Kui;Du, Hong-Zhi;Wu, Na;Feng, Zi-Xuan;Wang, Qi-Ming;Wan, Shan-Ping;Yu, Fu-Qiang;Yu, Fu-Qiang;Du, Hong-Zhi;Du, Hong-Zhi;Hu, Hong-Li;Su, Ji-Yu;Wang, Zong-Hua;Hu, Yan-Ping;Yu, Hao;Wang, Zong-Hua;Wang, Jing;Yang, Yi-Hua;Dong, Wei;Shu, Yong-Xin;Zhao, Hai-Jun;Liao, Chun-Fang;Doilom, Mingkwan;Wang, Chao-Qun;Xia, Wen-Xiao;Li, Guo-Jie;Suwannarach, Nakarin;Senwanna, Chanokned;Suwannarach, Nakarin;Senwanna, Chanokned;Gafforov, Yusufjon;Gafforov, Yusufjon;Flakus, Adam;Suchan, Tomasz;Rodriguez-Flakus, Pamela;Plata, Oscar;Tamang, Juna;Acharya, Krishnendu;Tiendrebeogo, Assiata;Somda, Irenee;Decock, Cony;Tiendrebeogo, Assiata;Legreve, Anne;Yang, Yu;Xiao, Yuan-Pin;Zhou, Xian-Zhi;Li, Jia-Xin;Zhao, Rui-Lin;Wang, Shi-Hui;He, Mao-Qiang

    关键词：Ascomycota; Basidiomycota; New combination; New genus; New name; New record; New species

  • Dual-Attention-Enhanced MobileViT Network: A Lightweight Model for Rice Disease Identification in Field-Captured Images

    作者：Zhang, Meng;Lin, Zichao;Tang, Shuqi;Lin, Chenjie;Zhong, Nan;Zhang, Meng;Zhang, Liping;Dong, Wei;Zhang, Meng;Lin, Zichao;Tang, Shuqi;Lin, Chenjie;Zhong, Nan;Zhang, Meng;Lin, Zichao;Tang, Shuqi;Lin, Chenjie;Zhong, Nan

    关键词：rice diseases; lightweight model; deep learning; attention mechanism; visualization

微信小程序