A lightweight real-time algorithm for plum harvesting detection in orchards under complex conditions

被引:0
作者
Zhang, Dongyan [1 ]
Chen, Nuo [1 ]
Mao, Siyu [1 ]
Wu, Chenxv [1 ]
Gu, Dandan [1 ]
Zhang, Lanxiang [1 ]
机构
[1] Univ Northeast Forestry, Coll Comp & Control Engn, Harbin 150040, Peoples R China
基金
中国国家自然科学基金;
关键词
Deep learning; Plum detection; Target detection; Smart picking; Real-time;
D O I
10.1007/s11760-025-03864-8
中图分类号
TM [电工技术]; TN [电子技术、通信技术];
学科分类号
0808 ; 0809 ;
摘要
Accurate and efficient algorithms for detection are essential technologies utilized by robots for plum harvesting. To facilitate effective implementation on edge computing devices that have restricted processing capabilities, this research introduces a lightweight plum detection algorithm derived from YOLOv5s. Initially, the backbone network of the original architecture is substituted with the Revcol network. Subsequently, the feature space pyramid pooling (SPPF) is refined by integrating a Large Separable Kernel Attention (LSKA) mechanism, and the C3 module of the original model is improved through the use of Distributed Shift Convolutions (DSConv). Finally, a custom-designed lightweight detection head is employed to construct the lightweight plum detection model. The results show that the average mAP of the improved detection model is 96.3%, and the number of floating point operations (FLOPs) and parameters are reduced to 84.8% and 91.2% respectively. On edge computing devices, the inference speed reaches 44.64 frames per millisecond. This algorithm features high accuracy, lightweight design, and fast inference speed, providing valuable references for real-time object detection deployment of plum harvesting robots on edge computing devices.
引用
收藏
页数:12
相关论文
共 26 条
  • [1] Real-Time Monitoring Method of Strawberry Fruit Growth State Based on YOLO Improved Model
    An, Qilin
    Wang, Kai
    Li, Zhongyang
    Song, Chengyuan
    Tang, Xiuying
    Song, Jian
    [J]. IEEE ACCESS, 2022, 10 : 124363 - 124372
  • [2] Cai YX, 2022, Arxiv, DOI arXiv:2212.11696
  • [3] Chen Hanting, 2024, Adv. Neural Inf. Process. Syst., V36
  • [4] Run, Don't Walk: Chasing Higher FLOPS for Faster Neural Networks
    Chen, Jierun
    Kao, Shiu-Hong
    He, Hao
    Zhuo, Weipeng
    Wen, Song
    Lee, Chul-Ho
    Chan, S. -H. Gary
    [J]. 2023 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR), 2023, : 12021 - 12031
  • [5] Table grape inflorescence detection and clamping point localisation based on channel pruned YOLOV7-TP
    Du, Wensheng
    Jia, Zihang
    Sui, Shunshun
    Liu, Ping
    [J]. BIOSYSTEMS ENGINEERING, 2023, 235 : 100 - 115
  • [6] A detection algorithm based on improved YOLOv5 for coarse-fine variety fruits
    Fan, Yanru
    Cai, Yuliang
    Yang, Huijun
    [J]. JOURNAL OF FOOD MEASUREMENT AND CHARACTERIZATION, 2024, 18 (02) : 1338 - 1354
  • [7] Automatic detection of pecan fruits based on Faster RCNN with FPN in orchard
    Hu, Chunhua
    Shi, Zefeng
    Wei, Hailin
    Hu, Xiangdong
    Xie, Yuning
    Li, Pingping
    [J]. INTERNATIONAL JOURNAL OF AGRICULTURAL AND BIOLOGICAL ENGINEERING, 2022, 15 (06) : 189 - 196
  • [8] Large Separable Kernel Attention: Rethinking the Large Kernel Attention design in CNN
    Lau, Kin Wai
    Po, Lai-Man
    Rehman, Yasar Abbas Ur
    [J]. EXPERT SYSTEMS WITH APPLICATIONS, 2024, 236
  • [9] Li Y., 2022, Advances in Neural Information Processing Systems, V35, P12934, DOI DOI 10.48550/ARXIV.2206.01191
  • [10] Research on efficient feature extraction: Improving YOLOv5 backbone for facial expression detection in live streaming scenes
    Li, Zongwei
    Song, Jia
    Qiao, Kai
    Li, Chenghai
    Zhang, Yanhui
    Li, Zhenyu
    [J]. FRONTIERS IN COMPUTATIONAL NEUROSCIENCE, 2022, 16