Tea detection based on YOLOv8 and PyQt5
Keywords:
YOLOv8, PyQt5, Tea detect, Attention fusion, Asymmetric convolutionAbstract
Under the trend of intelligent transformation in modern agriculture, the contradiction between the efficiency and quality of tea picking urgently needs to be resolved. This study conducts tea detection optimization based on the YOLOv8 algorithm and constructs a complete technical path from theory to practice. The study first analyzes the network architecture of YOLOv8, combines the characteristics of the One-Stage algorithm, and clarifies its advantages in real-time tea detection. Through training and deployment on public datasets, the detection accuracy has been improved by 6.7% compared to the original YOLOv8, providing algorithmic support for mechanized picking. For the complex environment of tea gardens, a module optimization strategy is proposed: introducing a fusion attention module to generate a new C2fM module, and introducing asymmetric convolution to generate the ACBSPPF module. Through ablation experiments and cross-validation, the optimization effect was verified using mAP and FPS as indicators. The model has reached the industry-leading level in terms of real-time performance and accuracy. Research shows that the optimized YOLOv8 algorithm effectively solves the problem of tea detection. Finally, a tea detection system is designed using PyQt5, providing a feasible solution for the industrialization of intelligent picking technology.
References
Li Linshan.Research Progress on Tea Picking Techniques and Tea Picking Machinery[J].Southern Agricultural Machinery,2024,55(13):61-64.
He Yu.Research on Tea Bud Recognition Based on Deep Learning[D].Xihua University,2023.
Xu Zheng.Manual Picking Techniques for Fresh Tea Leaves[J].Science Planting and Breeding,2018(03):21.
Zheng Hang,Fu Tong,Xue Xianglei, et al.Research Status and Prospect of Mechanized Tea Picking Technology[J].Chinese Journal of Agricultural Mechanization,2023,44(09):28-35.
Dong Chunwang. Innovative Thoughts on Intelligent Processing Technology of Tea[J].China Tea,2019,41(03):53-55.
Chen X,Gupta A.An implementation of faster rcnn with study for region sampling[J].arxiv preprint arxiv:1702.02138,2017.
Liu W,Anguelov D,Erhan D,et al.SSD:Single shot multibox detector[C]//European Conference on Computer Vision,2016:21-37.
Redmon J,Divvala S,Girshick R,et al.You only look once:Unified,real-time object detection[C]//Proceedings of the IEEE conference on computer vision and pattern recognition.2016:779-788.
Reis D,Kupec J,Hong J,et al.Real-time flying object detection with YOLOv8[J].arxiv preprint arxiv:2305.09972,2023.
Simonyan K,Zisserman A.Very deep convolutional networks for large-scale image recognition[J].arXiv preprint arXiv:1409.1556,2014.
He K,Zhang X,Ren S,et al.Deep residual learning for image recognition[C]//Proceedings of the IEEE conference on computer vision and pattern recognition.2016:770-778.
Xu Yanwei,Li Jun,Dong Yuanfang,et al.Review of YOLO Series Object Detection Algorithms [J].Journal of Frontiers of Computer Science & Technology,24,18(9).
WOO S,PARK J,LEE J Y,et al.CBAM:Convolutional block attention module[C]//Computer Vision-ECCV 2018.Cham:Springer International Publishing,2018:3⁃19.
Xu Jun. Research on Computer Vision Image Description Based on Deep Learning [J]. Information and Computer (Theoretical Edition),2023,35(19):155-157.
WANG Q L,WU B G,ZHU P F,et al.ECA-Net:Efficient Channel Attention for Deep Convolutional Neural Networks[C]//2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition(CVPR).Seattle:IEEE,2020:11531-11539.
YangL,WangS,Chen X,et al.High-fidelity permeabilityandporositypredictionusingdeeplearning withtheself-attentionmechanism[J].IEEETransactions on Neural Networksand Learning Systems,2022,34(7):3429-3443.
Zhang H,Goodfellow I,Metaxas D,et al.Self-attention genera⁃tive adversarial networks[C]//International conference on ma⁃chine learning.PMLR,2019:7354-7363.
Song Yi, Ding Geyuan. Research on Foreign Object Detection Algorithm for Transmission Lines Based on Improved YOLOv8[J].Industrial Control Computer,2020,38(06):49-51.
Zhao Xiaoxiao Vehicle License Plate Detection Based on YOLOv10n+SE Attention Mechanism [J].Information and Computer,2020,37(10):6-8.
MAO Ziwei, Zhou Zhengkang, Tang Jiashan. Research on Road Vehicle Fire Detection Algorithm Based on Improved YOLOv8 [J]. Radio Engineering,2020,55(05):920-927.
Zhu Yuhua,Zhang Yuhuan,Li Zhihui,et al.Research on Grain Storage Temperature Prediction Based on TCN-BiGRU Combined with Self-Attention Mechanism[J].Chinese Journal of Agricultural Mechanization,24,45(12):133-139.
Yang Bingzhen,Lin Yuanhao, Ji Lihua,et al.Bearing Fault Diagnosis Method Combining Multi-head Attention Mechanism with Generative Adversarial Neural Network[J].Mechanical and Electrical Engineering Technology,2020,54(11):158-163.
Gao Min,Chen Gaohua,Gu Jiaxin,et al.FLM-YOLOv8:A Lightweight Mask-Wearing Detection Algorithm [J].Computer Engineering and Applications,2024,60(17):203-215.
Ouyang Jianquan, Tang Huanrong, Lu Jiaxiong. Research on Target Detection of Pig Counting Based on YOLOv8[J].Journal of Xiangtan University(Natural Science Edition),2025,1-13.
Wang Yongsen, Liu Qian, Liu Libo.ACGFN:Speech Recognition Model Based on Asymmetric Convolution and Gated Feedforward Neural Network[J].Journal of Chinese Information Processing,2020,39(01):167-174.
Kabir M M,Hafiz M S,Bandyopadhyaa S,et al.Tea leaf age quality:age-stratified tea leaf quality classification dataset[J].Data in Brief,2024,54:110462.
Zhang Wenguang,Zeng Xiangjiu,Liu Chongyang.Research on Graphic Element Recognition Method of Electric Power Dispatching Control System Based on Improved YOLOv7[J].Electrical Technology,2025,26(5):1-9.
WU Binbin, ZHANG Lihua, LIU Junwei, DONG Junjun. Improved EP-RTDETR based surface defect detection on PCB[J].Manufacturing Technology & Machine Tool,2025,(3):139-148.
CHEN Wei，JIANG Zhicheng，TIAN Zijian，et al. Unsafe action detection algorithm of underground personnel in coal mine based on YOLOv8[J].Coal Science and Technology,2024,52(S2):267-283.
Qi Xiangyu, Su Qinghua, Zhang Zhichao, et al. Defect Detection Algorithm for Improving YOLOv11[J].Computer Science and Applications,2025,15:108.
Guo Jing,Hu Meng,Li Weishan,et al.Research on Cinema Ticketing System Detection Platform Based on PyQt5 and SpringBoot[J].Modern Information Science and Technology, 2020,9(01):88-92+99.
Liu Runji. Research on Finger Vein Recognition Algorithm Based on Lightweight Network[D].North China Unive
