MACNet: A More Accurate and Convenient Pest Detection Network

被引:0
|
作者
Hu, Yating [1 ]
Wang, Qijin [2 ,3 ]
Wang, Chao [1 ]
Qian, Yu [1 ]
Xue, Ying [1 ]
Wang, Hongqiang [3 ]
机构
[1] Anhui Jianzhu Univ, Sch Elect & Informat Engn, Hefei 230601, Peoples R China
[2] Anhui Xinhua Univ, Sch Big Data & Artificial Intelligence, Hefei 230088, Peoples R China
[3] Inst Intelligent Machines, Hefei Inst Phys Sci, Chinese Acad Sci, Hefei 230031, Peoples R China
关键词
object detection; YOLOv8s; feature sampling; convolution; agricultural pests;
D O I
10.3390/electronics13061068
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
Pest detection: This process is essential for the early warning of pests in the agricultural sector. However, the challenges posed by agricultural pest datasets include but are not limited to species diversity, small individuals, high concentration, and high similarity, which greatly increase the difficulty of pest detection and control. To effectively solve these problems, this paper proposes an innovative object detection model named MACNet. MACNet is optimized based on YOLOv8s, introducing a content-based feature sampling strategy to obtain richer object feature information, and adopts distribution shifting convolution technology, which not only improves the accuracy of detection but also successfully reduces the size of the model, making it more suitable for deployment in the actual environment. Finally, our test results on the Pest24 dataset verify the good performance of MACNet; its detection accuracy reaches 43.1 AP which is 0.5 AP higher than that of YOLOv8s, and the computational effort is reduced by about 30%. This achievement not only demonstrates the efficiency of MACNet in agricultural pest detection, but also further confirms the great potential and practical value of deep learning technology in complex application scenarios.
引用
收藏
页数:15
相关论文
共 50 条
  • [1] MORE ACCURATE, MORE CONVENIENT ECHOCARDIOGRAPHY
    不详
    [J]. SIEMENS REVIEW, 1986, 53 (05): : 26 - 26
  • [2] Personal computer based image analysis - more precise, more accurate and more convenient
    Borley, NR
    Warren, BF
    [J]. JOURNAL OF PATHOLOGY, 1999, 187 : 19A - 19A
  • [3] Burn image segmentation based on Mask Regions with Convolutional Neural Network deep learning framework: more accurate and more convenient
    Jiao, Chong
    Su, Kehua
    Xie, Weiguo
    Ye, Ziqing
    [J]. BURNS & TRAUMA, 2019, 7
  • [4] A novel ruler for more convenient and accurate measurement of penile length only
    Moon, Hyun Joon
    [J]. INTERNATIONAL JOURNAL OF UROLOGY, 2016, 23 : 83 - 83
  • [5] LMANet: A Lighter and More Accurate Multiobject Detection Network for UAV Remote Sensing Imagery
    Fu, Qingwei
    Zheng, Qianying
    Yu, Fan
    [J]. IEEE GEOSCIENCE AND REMOTE SENSING LETTERS, 2024, 21
  • [6] BOUNDARY INFORMATION MATTERS MORE: ACCURATE TEMPORAL ACTION DETECTION WITH TEMPORAL BOUNDARY NETWORK
    Zhang, Tao
    Liu, Shan
    Li, Thomas
    Li, Ge
    [J]. 2019 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH AND SIGNAL PROCESSING (ICASSP), 2019, : 1642 - 1646
  • [7] BETTER, ACCURATE, MORE RAPID AND CONVENIENT PHYSICIAN-OFFICE SEROLOGIC TEST FOR DETECTION OF HELICOBACTER-PYLORI INFECTION
    GRAHAM, DY
    EVANS, DG
    EVANS, DJ
    [J]. GASTROENTEROLOGY, 1994, 106 (04) : A83 - A83
  • [8] A convenient and accurate method for the determination and detection of carbon monoxide in blood
    Christman, AA
    Randall, EL
    [J]. JOURNAL OF BIOLOGICAL CHEMISTRY, 1933, 102 (02) : 595 - 609
  • [9] OSNA enables more accurate detection of micrometastases
    Peter Sidaway
    [J]. Nature Reviews Clinical Oncology, 2018, 15 : 68 - 68
  • [10] MACNet: Multi-Attention and Context Network for Polyp Segmentation
    Hao, Xiuzhen
    Pan, Haiwei
    Zhang, Kejia
    Chen, Chunling
    Bian, Xiaofei
    He, Shuning
    [J]. WEB AND BIG DATA, PT II, APWEB-WAIM 2022, 2023, 13422 : 369 - 384