| 利用深度学习进行GF-6影像枣园检测识别 |
| |
| 引用本文： | 段晨阳,冯建中,全斌,白林燕,王盼盼.利用深度学习进行GF-6影像枣园检测识别[J].测绘通报,2022,0(3):54-59. |
| |
| 作者姓名： | 段晨阳 冯建中 全斌 白林燕 王盼盼 |
| |
| 作者单位： | 1. 西安科技大学, 陕西 西安 710054;2. 中国农业科学院, 北京 100081;3. 中国科学院空天信息创新研究院, 北京 100094;4. 新疆生产建设兵团第十四师农业科学研究所, 新疆 昆玉 848100 |
| |
| 基金项目： | 新疆生产建设兵团重点领域科技攻关计划;中国农业科学院科技创新工程项目 |
| |
| 摘 要： | 针对新疆南疆大规模枣园的检测识别,本文提出了一种基于泛化迁移深度学习的枣园目标检测识别方法。以GF-6卫星影像数据为基础制作了Jujube数据集,并将其泛化扩充增强;以Faster R-CNN体系为基础,利用多态协同模式实现数据集的有效关联和优化重构,进行检测识别模型的迁移深度学习以提高对目标对象检测识别的泛化能力。结果表明,模型算法的验证识别精确率、召回率和调和平均值分别达0.979、0.952和0.965,在应用测试中,3个指标平均值均大于0.929,优于传统检测方法,且本文模型方法总体分类精度为0.97,Kappa系数为0.93,均高于面向对象最邻近法,能够有效地满足研究区规模化枣园目标检测识别的精度和效率的要求,为精细化枣园田间管理提供基础依据。
|
| 关 键 词： | 枣园目标检测 Faster R-CNN 泛化迁移学习 数据增强 GF-6 |
| 收稿时间： | 2021-03-19 |
| 修稿时间： | 2021-05-11 |
Jujube garden detection and recognition in GF-6 image using deep learning |
| |
| DUAN Chenyang,FENG Jianzhong,QUAN Bin,BAI Linyan,WANG Panpan.Jujube garden detection and recognition in GF-6 image using deep learning[J].Bulletin of Surveying and Mapping,2022,0(3):54-59. |
| |
| Authors: | DUAN Chenyang FENG Jianzhong QUAN Bin BAI Linyan WANG Panpan |
| |
| Institution: | 1. Xi'an University of Science and Technology, Xi'an 710054, China;2. Chinese Academy of Agricultural Sciences, Beijing 100081, China;3. Aerospace Information Research Institute, Chinese Academy of Sciences, Beijing 100094, China;4. Institute of Agricultural Sciences, the 14th Division of Xinjiang Production and Construction Corps, Kunyu 848100, China |
| |
| Abstract: | Focusing on the large-scale jujube fields in the southern Xinjiang,this paper proposes a jujube orchard detection method based on a generalized deep transfer learning principle.From GF-6 satellite imagery,a jujube field dataset is made,and then it is augmented effectively.Grounded on a Faster R-CNN system,a multi-modally cooperative mode is used to realize the effective correlation and optimization reconstruction of the expanded dataset,and a transfer deep learning of detection and recognition model is thus carried out to improve the generalization ability of the detection and recognition of target object on jujube fields.The results show that the precision,recall and F1-score of the model algorithm reached 0.979,0.952 and 0.965,respectively.In the application tests,the average values of the three indexes are all more than 0.929,which could better than traditional detection method,and the overall classification accuracy and Kappa coefficient of this model method are 0.97 and 0.93,which are higher than the object-oriented nearest neighbor method,and effectively meet the requirements of high-efficient and accurate large-scale jujube orchard detection in the study area.Then it provides the basis for fine jujube orchard field management. |
| |
| Keywords: | jujube orchard detection Faster R-CNN generalized transfer learning data augmentation GF-6 |
| 本文献已被 万方数据 等数据库收录！ |
| 点击此处可从《测绘通报》浏览原始摘要信息 |
| 点击此处可从《测绘通报》下载免费的PDF全文 |
|
