基于深度学习的桃树叶片病害识别系统设计与实现
河南科技学院, 信息工程学院, 河南新乡 453003
| 摘 要： | 桃树在我国有广泛栽培，随着人们生活水平的提高，对桃果的质量要求不断提高。桃树叶片病害的发生严重影响到了桃树果子的产量和质量。生产高品质的桃果需要加强对桃树叶片病害的管理。传统的病害叶片的识别方法是人工识别，识别过程中的主观性和难度都比较大。本文提出使用Anaconda+Python+Pycharm软件，与增加了注意力机制CBAM后的YOLOv5模型相结合的桃树叶片病害的检测方法，并开发出GUI界面式的识别系统供用户操作。软件设计界面主要有登录界面、注册界面、检测主界面。果农将采集的桃树叶片病害图片导入识别系统，就可以在本软件中检测主界面的选择权重模块，选择合适的权重。然后对模型进行初始化，通过图片检测模块，与本系统包括的桃树缩叶病、细菌性穿孔病、黄叶病、褐斑穿孔病、炭疽病五种叶片病害数据集类型进行自动对比，进而获取桃树叶片病害类型，根据叶片病类型提供相应的防治建议。本系统的开发使桃树叶片病害的识别准确率提升了2.15%。IOU提高了3.02%，达到82.37%。减少了果农的误诊率，增加果农对桃树叶片病识别准确率，采取正确的防治措施，减少叶片病危害，提高果农收入。 |
| 关 键 词： | 桃树叶片病; 深度学习; 界面设计; YOLOv5; 目标检测 |
| DOI： | 10.57237/j.cst.2023.01.001 |
View:
Downloads:
收稿日期: 2022-12-07
接受日期: 2023-01-16
在线出版日期: 2023-02-03
Design and Implementation of Peach Leaf Disease Identification System
School of Information Engineering, Henan Institute of Science and Technology, Xinxiang 453003, China
| Abstract: | Peach trees are widely cultivated in China. With the improvement of people's living standards, the requirements for the quality of peaches are constantly improving. The occurrence of peach leaf diseases has seriously affected the yield and quality of peach fruits. The production of high-quality peach fruit requires the necessary management of peach leaf diseases. The traditional identification method of diseased leaves is manual identification, which is subjective and difficult. This paper proposes a method of peach leaf disease detection using Anaconda+Python+Pycharm software and YOLOv5 model after adding attention mechanism CBAM, and develops a GUI interface recognition system for users to operate. The software design interface mainly includes login interface, registration interface and main detection interface. Fruit growers can import the collected peach leaf disease pictures into the recognition system, and then check the selection weight module of the main interface in this software to select the appropriate weight. Then initialize the model, and automatically compare it with the data set types of five leaf diseases including peach leaf shrinkage disease, bacterial perforation disease, yellow leaf disease, brown spot perforation disease and anthrax in this system through the picture detection module, so as to obtain the type of peach leaf disease, and provide corresponding prevention and control suggestions according to the type of leaf disease. The development of this system has improved the identification accuracy of peach leaf diseases by 2.15%. IOU increased by 3.02% to 82.37%. It reduces the misdiagnosis rate of fruit growers, increases the recognition accuracy of peach leaf disease, takes correct control measures, reduces the harm of leaf disease, and improves the income of fruit growers. |
| Keywords: | Peach Leaf Disease; Deep Learning; Interface Design; YOLOv5; Disease Detection |
基金
| 1. | 河南省科技厅科技公关项目《基于机器学习的农作物病害识别与物理防控》 (222102210116) |
| 2. | 《基于AIoT和物理技术防控害虫的绿色生态果园》 (212102310553). |
参考文献
| [1] | Salunkhe D K, Desai BB. Postharvest Biotechnology of Fruits [J]. CRC Press, Boca Raton, FL. 1984, p. 168. |
| [2] | 刘天赐等. 智慧农业病虫害巡检系统设计与实现 [J]. 福建电脑, 2022, 38 (01): 89-92. |
| [3] | BRAHIMI M, BOUKHALFA K, MOUSSAOUI A. Deep Learning for Tomato Diseases: Classification and Symptoms Visualization [J]. Applied Artificial Intelligence, 2017, 31 (4): 299-315. |
| [4] | 雷声渊等. 基于微调卷积神经网络迁移学习模式下被害棉叶图像的识别 [J]. 新疆农业科学, 2019, 56 (7): 1288- 1295. |
| [5] | 鲍文霞等. 基于多路卷积神经网络的大田小麦赤霉病图像识别 [J]. 农业工程学报, 2020, 36 (11): 174-181. |
| [6] | KIRATIRATANAPRUK K, TEMNIRANRAT P, KIT‐ VIMONRAT A, et al., Using deep learning techniques to detect rice diseases from images of rice fields [C] // International Conference on Industrial, Engineering and Other Applications of Applied Intelligent Systems. Berlin, German: Springer, 2020: 225-237. |
| [7] | REDMON J, DIVVALA S, GIRSHICK R, et al. You only look once: Unified, real-time object detection [C] // The IEEE Conference on Computer Vision and Pattern Recognition. Piscataway, New York, USA: IEEE, 2016: 779-788. |
| [8] | 沈凯文, 李浩伟, 韩进, 房若民. 基于卷积神经网络的病虫害可视化监测系统设计 [J]. 软件导刊, 2020, 19 (09): 122-126. |
| [9] | 贾少鹏, 高红菊, 杭潇. 基于深度学习的农作物病虫害图像识别技术研究进展 [J]. 农业机械学报, 2019, 50 (S1): 313-317. |
| [10] | 曾弟先等. 基于卷积神经网络的水稻病虫害识别分类系统设计与实现 [J]. 电脑与信息技术, 2021. 09, 29 (3) 16-21. |
| [11] | 邵明月等: 深度学习在植物叶部病害检测与识别的研究进展 [J]. 智慧农业 (中英文), 2022, 4 (01), 29-46. |
| [12] | 朱逢乐, 郑增威. 基于图像和卷积神经网络的蝴蝶兰 种 苗 生 长 势 评 估 [J]. 农 业 工 程 学 报, 2020, 36 (9): 185-194. |
| [13] | Ji J, Pang W, Zhou C, et al. Afuzzy k-prototype clustering algorithm for mixed numeric and categorical data [J]. Neurocomputing, 2012, 30 (10): 129-135. |
| [14] | 刘芳, 刘玉坤, 林森, 等. 基于改进型YOLO的复杂环境下番茄果实快速识别方法 [J]. 农业机械学报, 2020, 51 (6): 229-237. |
| [15] | Padilla R, Netto S L, da Silva E A B. A survey on performance metrics for object-detection algorithms [C] // 2020 International Conference on Systems, Signals and Image Processing (IWSSIP). IEEE, 2020: 237-242. |
| [16] | 蒋镕圻, 彭月平, 谢文宣, 等. 嵌入 scSE 模块的改进YOLOv4小目标检测算法[J]. 图学学报, 2021, 42(4): 546-555. |
| [17] | 龙洁花等. 改进YOLOv4的温室环境下草莓生育期识别方法 [J]. 智慧农业（中英文）, 2021, 3 (4): 99-110. |
