Article Details

Abstract

Aiming at the existing problems of insufficient accuracy, the difficulty of identification under complex backgrounds, and the obvious limitations of traditional methods in feature extraction in fruit ripeness detection, a fruit ripeness detection algorithm (AGLU-YOLOv11) based on improved YOLOv11 is proposed, to meet the demands for efficient data and reliable collection in fruit ripeness detection. AGLU-YOLOv11 designs the C3k2_AddBlock_CGLU module by optimizing the C3k2 module in the YOLOv11 backbone network and integrating CATM (Conv Additive Self-Attention) and CGLU (Convolutional Gated Linear Unit), and significantly enhances feature extraction capability and adaptability of multi-variety and multi-stage ripeness fruits. At the same time, the AFCA Attention Mechanism is introduced in the feature fusion stage to strengthen global feature expression and adaptability to complex backgrounds, and achieve efficient fruit quality detection and labeling. Experimental results show that AGLU-YOLOv11 performs better in precision, robustness and multi-scale object adaptability than other detection models in Precision, Recall, mAP@0.5 and mAP@0.5:0.95 indicators, and can better meet the demands for identifying fruit ripeness.

Keywords

YOLO; Object Detection; CGLU; CATM; fruit ripeness detection

References

1. [1] Zhao Jiebin, Deng Hao, Xu Zhenlin Risk assessment of pesticide residues in vegetables and fruits in Taishan City based on the food safety index method [J]. Modern Food Science and Technology, 2023, 39 (6): 277-283
2. [2] Tian Youwen, Wu Wei, Lu Shilead, etc Application of Deep Learning in Fruit Quality Detection and Grading Classification [J]. Food Science, 2021, 42 (19): 260-270
3. [3] CUBERO S，ALEIXOS N，ALBERT F，et al. Optimised Computer Vision System for Automatic Pre-Grading of Citrus Fruit in the Field Using a Mobile Platform [J].Precision Agriculture，2014，15（1）：80-94.
4. [4] OHALI A Y. Computer Vision Based Date Fruit Grading System: Design and Implementation [J].Journal of King Saud University-Computer and Information Sciences，2010，23（1）：29-36.
5. [5] Song Lihang, Zhang Yi, Shi Yinhao A method for detecting the maturity of peaches in natural environments based on improved YOLOv5s [J]. China Agricultural Machinery Chemistry Journal, 2024, 45 (9): 250-257+2
6. [6] ZHAO J F，DU C F，LI Y，et al. YOLO-Granada: A Lightweight Attentioned Yolo for Pomegranates Fruit Detection [J/OL].Scientific Reports，2024，14（1）：16848[2024-10-26].https://doi.org/10.1038/s41598-024-67526-4.
7. [7] REDMON J. You Only Look Once, Unified，Real-Time Object Detection [C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.Las Vegas：IEEE，2016：779-788.
8. [8] Chang Kai, Ma Jidong, Zhang Hao, etc Research on Wood Surface Defect Detection Based on Improved YOLOv8 [J/OL]. Journal of Forestry Engineering, 2024:1-12 [2024-12-03] https://doi.org/10.13360/j.issn.2096-1359.202410002.
9. [9] ARISTOTELES，SYARIF A，SUTYARSO，et al. Identification of Human Sperm Based on Morphology Using the You Only Look Once Version 4 Algorithm [J/OL].International Journal of Advanced Computer Science and Applications (IJACSA)，2022，13（7）[2024-11-02].https://dx.doi.org/10.14569/IJACSA.2022.0130752.
10. [10] ZHANG T，LI L，ZHOU Y，et al. CAS-ViT: Convolutional Additive Self-Attention Vision Transformers for Efficient Mobile Applications [J/OL].arXiv:2408.03703 [cs.CV].[2024-10-28].https://arxiv.org/abs/2408.03703?context=cs.
11. [11] SHI D. TransNeXt: Robust Foveal Visual Perception for Vision Transformers [C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. Seattle：IEEE，2024：17773-17783.
12. [12] HU J，SHEN L，SUN G. Squeeze-and-Excitation Networks [J].IEEE Transactions on Pattern Analysis and Machine Intelligence，2019，42（8）：2011-2023.
13. [13] SUN H，WEN Y，FENG H，et al. Unsupervised Bidirectional Contrastive Reconstruction and Adaptive Fine-Grained Channel Attention Networks for Image Dehazing [J/OL].Neural Networks，2024，176：106314[2024-11-02].https://doi.org/10.1016/j.neunet.2024.106314.

Funding

Key Project of Science and Technology Strategy Research in Shanxi Province (202304031401011)

About the authors

Zhao Peng (1973-), male, Han nationality, from Taiyuan, Shanxi, professor, doctoral student, research direction: software engineering, big data, blockchain; Qiang Guanglei (2001-), male, Han ethnicity, from Wuhu, Anhui Province, is currently pursuing a master's degree in blockchain and object detection; Corresponding author: Lu Bo (1988-), male, Han ethnicity, from Jinzhong, Shanxi Province, PhD, research direction: blockchain, protocol security analysis; Gao Yang (2000-), male, Han ethnicity, from Xinzhou, Shanxi, currently pursuing a master's degree in image processing and object detection; Zhang Qianxiang (2000-), male, Han ethnicity, from Taiyuan, Shanxi, is currently pursuing a master's degree in feature recognition, embedded systems, and IoT engineering.