农业大数据学报 ›› 2023, Vol. 5 ›› Issue (2): 91-96.doi: 10.19788/j.issn.2096-6369.230214

• 数据论文 • 上一篇    下一篇

2023年南京农业病虫害图像识别数据集

王伯元1(), 管志浩1, 杨杨1, 胡林2,3,*(), 王晓丽2,3,*()   

  1. 1.南京理工大学计算机科学与工程学院,南京 210094
    2.中国农业科学院农业信息研究所,北京 100081
    3.国家农业科学数据中心,北京 100081
  • 收稿日期:2023-06-21 出版日期:2023-06-26 发布日期:2023-08-15
  • 通讯作者: 胡林,王晓丽
  • 作者简介:王伯元,硕士研究生,研究方向:数据挖掘;E-mail:boyuanwang@njust.edu.cn
  • 基金资助:
    农业育种科学数据自主应用软件研发(2022YFF 0712100);中央级公益性科研院所基本科研业务费专项(Y2023LM01)

The Agricultural Pest and Disease Image Recognition Dataset in Nanjing, Jiangsu Province, in 2023

WANG BoYuan1(), GUAN ZhiHao1, YANG Yang1, HU Lin2,3,*(), WANG XiaoLi2,3,*()   

  1. 1. School of Computer Science and Engineering, Nanjing University of Science and Technology, Nanjing 210094, China
    2. Agricultural Information Institute of Chinese Academe of Agricultural Sciences, Beijing 100081, China
    3. National Agriculture Science Data Center, Beijing 100081, China
  • Received:2023-06-21 Online:2023-06-26 Published:2023-08-15
  • Contact: HU Lin,WANG XiaoLi

摘要:

农业病虫害对农作物的产量和品质造成了严重的威胁,因此准确、高效地检测和识别病虫害是农业生产中的重要任务。本文介绍了一个综合的农业病虫害数据集,由农业虫害检测数据集、农业病害检测数据集、农业病害分类数据集和水稻表型分割数据集组成,包含55个类别、48576张,共4.14 GB的图像样本。从公开数据源和学术论文中收集和整理数据,保证了数据集的多样性和代表性。在数据的筛选、清洗和标注过程中,采用了严格的质量控制和验证措施,以确保数据集的准确性和可靠性。该数据集可用于农业病虫害识别和水稻表型鉴定等农业视觉任务,能够为农业病虫害研究提供有价值的资源,并促进农业生产的可持续发展。

关键词: 水稻, 病虫害, 图像

Abstract:

Agricultural pests and diseases pose a serious threat to crop yield and quality, making accurate and efficient detection and identification of pests and diseases crucial in agricultural production. In this paper, we propose a comprehensive agricultural pests and diseases dataset, which includes agricultural pest detection dataset, agricultural disease detection dataset, agricultural disease classification dataset, and rice phenotype segmentation dataset. By collecting and curating data from public sources and academic papers, we ensured the diversity and representativeness of the dataset. Rigorous quality control and validation measures were implemented during the data filtering, cleaning, and annotation processes to ensure the accuracy and reliability of the dataset. This dataset can be used for agricultural pest and disease recognition, as well as rice phenotype identification and other agricultural visual tasks. It provides valuable resources for agricultural pest and disease research and contributes to the sustainable development of agricultural production.

Key words: rice, pests and diseases, images