Article
Version 1
Preserved in Portico This version is not peer-reviewed
Highly Accuracy and Lightweight Detection of Apple Leaf Diseases Based on YOLO
Version 1
: Received: 23 May 2024 / Approved: 23 May 2024 / Online: 23 May 2024 (12:43:57 CEST)
How to cite: Sun, Z.; Feng, Z.; Chen, Z. Highly Accuracy and Lightweight Detection of Apple Leaf Diseases Based on YOLO. Preprints 2024, 2024051550. https://doi.org/10.20944/preprints202405.1550.v1 Sun, Z.; Feng, Z.; Chen, Z. Highly Accuracy and Lightweight Detection of Apple Leaf Diseases Based on YOLO. Preprints 2024, 2024051550. https://doi.org/10.20944/preprints202405.1550.v1
Abstract
Aiming at the problem of small size of some spots on apple leaves and the difficulty of accurate detection of spot targets brought by the complex background of the orchard, this paper takes the alternaria leaf spot, rust, brown spot, grey spot and frog eye leaf spot on apple leaves as the research object, and proposes a high-accuracy detection model YOLOv5-Res and lightweight detection model YOLOv5-Res4 are proposed. Firstly, a multiscale feature extraction module ResBlock is designed by combining the Inception multi-branch structure and ResNet residual idea. Secondly, a lightweight feature fusion module C4 is designed to reduce the number of model parameters while improving the detection ability of small targets. Finally, a parameter streamlining strategy based on optimized model architecture is proposed. The experimental results show that the performance of YOLOv5-Res model and YOLOv5-Res4 model is significantly improved, the [email protected] value is improved by 2.8% and 2.2% compared with YOLOv5s model and YOLOv5n model, respectively, and the size of YOLOv5-Res model and YOLOv5-Res4 model is only 10.8MB and 2.4MB, and the number of model parameters is reduced compared with YOLOv5s model and YOLOv5n model. counts are reduced by 22% and 38.3% compared to the YOLOv5s model and YOLOv5n model.
Keywords
leaf disease detection; YOLO; high-accuracy detection; lightweight detection
Subject
Computer Science and Mathematics, Robotics
Copyright: This is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Comments (0)
We encourage comments and feedback from a broad range of readers. See criteria for comments and our Diversity statement.
Leave a public commentSend a private comment to the author(s)
* All users must log in before leaving a comment