首页|基于深度学习的NDVI时空数据融合模型

基于深度学习的NDVI时空数据融合模型

扫码查看
原文链接
  • NETL
  • NSTL
  • 万方数据
  • 维普
卫星遥感提供了大尺度空间和长时间序列的地球表面变化信息,已广泛应用于生态学研究.人类活动可能对较小尺度的空间产生影响,并可能在较长时间内被检测到,这需要具有较高空间和时间分辨率的遥感数据.时空数据融合算法的发展为这些需求提供了机会.本文基于深度学习,提出了一种残差卷积神经网络(Res-CNN)模型,利用全新的网络架构来融合来自Landsat 8和MODIS图像的NDVI检测结果,从而显著提高融合结果.通过与现有算法的比较,在两个不同地区进行的实验都展示了改进效果.模型性能通过预测值与观测值之间的线性回归进行评估,并通过决定系数(R2)、回归斜率(Slope),并与两种传统的模型(ESTARFM、FSDAF)进行比较.结果显示,Res-CNN模型预测的NDVI对研究区上具有较高的解释能力(R2 分别为 0.773和 0.804,slope为 1.01和 0.989).该研究证明,本文开发的Res-CNN模型具有较高的精度和较强的稳健性,优于传统模型.这项研究具有广泛的应用意义,因为它不仅提供了一个时空数据融合模型,还可以为区域尺度农业和草地生态系统的管理和利用提供长时间序列的数据.
A Deep Learning-based Spatio-temporal NDVI Data Fusion Model
Satellite remote sensing provides the changes information of Earth surface on large spatial scale in a long time series and has been widely used in ecology.However,the possible impact from human activities gener-ally occurs on a smaller spatial scale and could be detected in a longer time,which requires the remote sensing data having the both higher spatial and temporal resolution.Meanwhile,the development of the spatiotemporal data fusion algorithm provides an opportunity for the requirements.In this paper,based on deep learning,we pro-posed a residual convolutional neural network(Res-CNN)model to improve the fusion result considerably with brand-new network architecture to fuse the NDVI retrievals from Landsat 8 and MODIS images.Experiments conducted in two different areas demonstrate improvements by comparing them with existing algorithms.The model performance was evaluated by a linear regression between predictions and observations and quantified by determination coefficients(R2),regressive ecoefficiency(slope).The two excellent models,ESTARFM and FSDAF,were compared with the new model on their performance.The results showed that the predicted NDVI had the higher exploitational on the variability in the Landsat-based NDVI with the R2 of 0.768 and 0.807 at the urban and grassland sites.The predicted NDVI was well consistent with the observations with the slope of 1.01 and 0.989,and the R-RMSE of 95.76%and 93.58%at the urban and grassland sites respectively.This study demonstrated that the Res-CNN model developed in this paper exhibits higher accuracy and stronger robustness than the traditional models.This research is full implications because it not only provides a model on the spatio-temporal data fusion,but also can provide the data of a long time series for the management and utilization of agriculture and grassland ecosystems on the regional scale.

NDVIspatio-temporal data fusionLandsatMODISconvolutional neural network

孙梓煜、欧阳熙煌、李浩、王军邦

展开 >

中国科学院地理科学与资源研究所,生态系统网络观测与建模重点实验室,国家生态系统科学数据中心,北京 100101

中国科学院大学,北京 100049

雄安创新研究院,河北保定 071899

NDVI 时空数据融合 Landsat MODIS 卷积神经网络

国家自然科学基金国家自然科学基金Joint Research Project of the People's Government of Qinghai Province and Chinese Academy of Sciences

3197150731861143015LHZX-2020-07

2024

10.5814/j.issn.1674-764x.2024.01.019

资源与生态学报(英文版)
中国科学院地理科学与资源研究所

资源与生态学报(英文版)

CSTPCD
影响因子：0.388
ISSN：1674-764X
年,卷(期)：2024.15(1)
  • 1