1 材料与方法
1.1 试验数据集采集
1.2 试验数据集构建
1.3 试验环境与训练参数
表1 试验环境参数配置 |
| 试验环境 | 具体内容 |
|---|---|
| 镜像 | PyTorch1.7.0 Python3.8(ubuntu18.04) Cuda 11.0 |
| GPU | Tesla T4(16 GB) * 1 |
| CPU | 8 vCPU Intel Xeon Processor (Skylake,IBRS) |
| 内存 | 56 GB |
安徽农学通报 >
2025 , Vol. 31 >Issue 2: 97 - 100
DOI: https://doi.org/10.16377/j.cnki.issn1007-7731.2025.02.018
基于YOLOv8s的水稻害虫图片智能识别
|
邓相红(1986—),男,湖南宁乡人,硕士,讲师,从事农作物疾病计算机人工智能图像识别与处理研究。 |
Copy editor: 李媛
收稿日期: 2024-10-29
网络出版日期: 2025-01-24
基金资助
湖南机电职业技术学院院级重点项目“基于深度学习的水稻病虫害智能检测与识别系统研究”(YJA202401)
Intelligent recognition of rice pest images based on YOLOv8s
Received date: 2024-10-29
Online published: 2025-01-24
邓相红 . 基于YOLOv8s的水稻害虫图片智能识别[J]. 安徽农学通报, 2025 , 31(2) : 97 -100 . DOI: 10.16377/j.cnki.issn1007-7731.2025.02.018
The distribution of pests during rice cultivation is characterized by small scale and high density, making identification challenging. This article was based on deep learning and the classic YOLOv8s lightweight model was used to train and recognize 14 types of rice pests, including rice leaf roller, rice leaf caterpillar, and rice stem maggot, etc. The model training and verification results were obtained. The training results showed that the model has good convergence speed and stability; the verification results indicated that the model has good performance, with the recognition accuracy of 0.788, the recall rate of 0.721, and the recognition accuracy of 0.809, mAP@0.5 of 0.772 for 14 rice pests. Overall, the model had good performance and can meet the requirements of rice pest detection. The research results provide references for the identification of rice pest.
Key words: deep learning; rice pest; YOLOv8s; target detection
表1 试验环境参数配置 |
| 试验环境 | 具体内容 |
|---|---|
| 镜像 | PyTorch1.7.0 Python3.8(ubuntu18.04) Cuda 11.0 |
| GPU | Tesla T4(16 GB) * 1 |
| CPU | 8 vCPU Intel Xeon Processor (Skylake,IBRS) |
| 内存 | 56 GB |
| 1 |
李辉,罗敏,岳佳欣. 基于计算机视觉技术的水稻病害图像识别研究进展[J]. 湖北农业科学,2022,61(4):9-15.
|
| 2 |
吕盛坪,李灯辉,冼荣亨. 深度学习在我国农业中的应用研究现状[J]. 计算机工程与应用,2019,55(20):24-33,51.
|
| 3 |
王聃,柴秀娟. 机器学习在植物病害识别研究中的应用[J]. 中国农机化学报,2019,40(9):171-180.
|
| 4 |
黄双萍,孙超,齐龙,等. 基于深度卷积神经网络的水稻穗瘟病检测方法[J]. 农业工程学报,2017,33(20):169-176.
|
| 5 |
邱靖,刘继荣,曹志勇,等. 基于卷积神经网络的水稻病害图像识别研究[J]. 云南农业大学学报(自然科学版),2019,34(5):884-888.
|
| 6 |
|
| 7 |
|
| 8 |
|
| 9 |
|
| 10 |
杨红云,肖小梅,黄琼,等. 基于卷积神经网络和迁移学习的水稻害虫识别[J]. 激光与光电子学进展,2022,59(16):333-340.
|
| 11 |
吴小燕,郭威,朱轶萍,等. 基于改进YOLOv8s的大田甘蓝移栽状态检测算法[J]. 智慧农业(中英文),2024,6(2):107-117.
|
| 12 |
李林轩. 农业多目标虫害的小目标检测[D]. 荆州:长江大学,2024.
|
/
| 〈 |
|
〉 |