FVC-Net： A fusion network for producing fine spatial resolution fractional vegetation cover

ZHANG Zhihao; WANG Qunming; DING Xinyu

doi:10.11834/jrs.20243112

Views : 0 下载量: 115 CSCD: 0 更多指标

R-PDF
PDF
Export
Share
Collection
Album

FVC-Net： A fusion network for producing fine spatial resolution fractional vegetation cover
Pages: 1-13(2024)
Published Online： 12 March 2024 ，
DOI： 10.11834/jrs.20243112

扫描看全文

张智昊，王群明，丁欣宇.XXXX.FVC-Net：一种高空间分辨率植被覆盖度的融合网络.遥感学报，XX（XX）： 1-13

ZHANG Zhihao，WANG Qunming，DING Xinyu. XXXX. FVC-Net： A fusion network for producing fine spatial resolution fractional vegetation cover. National Remote Sensing Bulletin， XX（XX）：1-13
张智昊，王群明，丁欣宇.XXXX.FVC-Net：一种高空间分辨率植被覆盖度的融合网络.遥感学报，XX（XX）： 1-13 DOI： 10.11834/jrs.20243112.

ZHANG Zhihao，WANG Qunming，DING Xinyu. XXXX. FVC-Net： A fusion network for producing fine spatial resolution fractional vegetation cover. National Remote Sensing Bulletin， XX（XX）：1-13 DOI： 10.11834/jrs.20243112.

摘要

植被覆盖度（FVC）是描述地表植被分布的定量指标之一。通过遥感卫星（如Landsat和Sentinel-2）获取大尺度下的高空间分辨（如10 m级） FVC，能为全球生态系统研究提供重要基础数据。然而，由于云雾干扰以及卫星重返时间分辨率有限等问题，高空间分辨FVC在时域上存在大量缺失。本文考虑协同30 m Landsat-8和10 m Sentinel-2数据，实现二者在时域上的互补。为解决二者空间分辨率不一致的问题，本文提出了一种基于FVC-Net的深度学习方法，通过融合10 m Sentinel-2归一化植被指数（NDVI）数据，将30 m Landsat FVC降尺度至10 m。FVC-Net方法构建双分支结构下的通道注意力模块用于FVC和NDVI的多尺度特征采集与融合，随后利用空间注意力模块将选择的特征进行细节增强，以有效描述不同获取时间下的10 m NDVI与30 m FVC之间的非线性映射关系。实验中，与四种典型非深度学习方法和四种深度学习方法相比，FVC-Net获得了更高精度的融合结果。FVC-Net有望应用于全球尺度下的30 m Landsat FVC产品的降尺度，为相关领域研究提供更为精细的数据支撑。

Abstract

Objective Fractional vegetation cover (FVC) is an important indicator to characterize the spatial distribution of vegetation on the land surface. Remote Sensing Satellites (such as Landsat and Sentinel-2) can acquire fine spatial resolution FVC data at 10 m level

which are crucial source for researches on global ecosystem. However

due to cloud contamination and limited temporal resolution of the satellites

a large amount of fine spatial resolution FVC data are not available in the temporal domain. This paper considers the collaboration of 30 m Landsat-8 and 10 m Sentinel-2 to increase the temporal frequency of the observations.Method To deal with the difference in the spatial resolution

a deep learning-based method named FVC-Net is proposed in this paper. FVC-Net fuses 30 m Landsat FVC with 10 m Sentinel-2 normalized difference vegetation index (NDVI) directly

producing 10 m Landsat FVC. Specifically

a two-branch network based on multi-scale attention mechanism is designed

in which the channel enhancement blocks are used in both FVC and NDVI branches for feature extraction and fusion. Then

the spatial attention blocks are used to increase the spatial details of the fused FVC features. The scheme designed in FVC-Net can help to characterize the non-linear relationship between 10 m NDVI and 30 m FVC effectively.Result In the experiments

the proposed FVC-Net was compared with four typical non-deep learning-based and four deep learning-based fusion methods. It was found that FVC-Net is consistently more accurate than the eight benchmark methods. The 10 m FVC results can display more spatial details than original 30 m FVC.Conclusion The proposed FVC-Net is an effective solution to downscale 30 m Landsat FVC to 10 m by fusion with 10 m Sentinel-2 NDVI

which can effectively overcome the differences between Sentinel-2 and Landsat data at different time points. FVC-Net has the potential to be applied to downscale the current 30 m Landsat FVC products at the global scale

of which the predictions can support the researches in the related fields greatly.

关键词

植被覆盖度（FVC）归一化植被指数（NDVI）深度学习降尺度数据融合

Keywords

Fractional vegetation cover (FVC)normalized difference vegetation index (NDVI)deep learningdownscalingdata fusion

references

BOYD D S, FOODY G M, RIPPLE W J. Evaluation of approaches for forest cover estimation in the Pacific Northwest, USA, using remote sensing [J]. Applied Geography, 2002, 22(4): 375-92.

CUI T X, GONG Z N, ZHAO W J, ZHAO Y L, LIN C. Research on estimating wetland vegetation abundance based on spectral mixture analysis with different endmember model: a case study in Wild Duck Lake wetland，Beijing. Acta Ecologica Sinica，2013，33(4):1160-1171．

崔天翔，宫兆宁，赵文吉，赵雅莉，林川.不同端元模型下湿地植被覆盖度的提取方法——以北京市野鸭湖湿地自然保护区为例．生态学报，2013，33( 4) : 1160-1171．

CHAVEZ J P, SIDES S, ANDERSON J. Comparison of Three Different Methods to Merge Multiresolution and Multispectral Data: Landsat TM and SPOT Panchromatic [J]. Photogrammetric Engineering and Remote Sensing, 1991, 57: 265-303.

CHEN Y, GE Y. Spatiotemporal image fusion using multiscale attention-aware two-stream convolutional neural networks[J]. Science of Remote Sensing, 2022, 6: 100062.

CHEN Y H, LI X B, SHI P J. Landscape Spatial-Temporal Pattern Analysis on Change in the Fraction of Green Vegetation Based on Remotely Sensed Data：A Case Study in Haidian District，Beijing[J]. Acta Ecologica Sinica，2002, 22(10)：1581-1586．

陈云浩,李晓兵,史培军.基于遥感的植被覆盖变化景观分析——以北京海淀区为例[J].生态学报,2002, 22(10):1581-1586.

DEGERICKX J, ROBERTS D A, SOMERS B. Enhancing the performance of Multiple Endmember Spectral Mixture Analysis (MESMA) for urban land cover mapping using airborne lidar data and band selection [J]. Remote Sensing of Environment, 2019, 221: 260-73.

GUERSCHMAN J P, HILL M J, RENZULLO L J, et al. Estimating fractional cover of photosynthetic vegetation, non-photosynthetic vegetation and bare soil in the Australian tropical savanna region upscaling the EO-1 Hyperion and MODIS sensors [J]. Remote Sensing of Environment, 2009, 113(5): 928-45.

GUTMAN G, IGNATOV A. The derivation of the green vegetation fraction from NOAA/AVHRR data for use in numerical weather prediction models [J]. International Journal of Remote Sensing, 2010, 19(8): 1533-43.

HUANG C, SONG K, KIM S, et al. Use of a dark object concept and support vector machines to automate forest cover change analysis [J]. Remote Sensing of Environment, 2008, 112(3): 970-85.

HANSEN M C, DEFRIES R S, TOWNSHEND J R G, et al. Towards an operational MODIS continuous field of percent tree cover algorithm: examples using AVHRR and MODIS data [J]. Remote Sensing of Environment, 2002, 83(1-2): 303-19.

JIA D, CHENG C, SONG C, et al. A hybrid deep learning-based spatiotemporal fusion method for combining satellite images with different resolutions[J]. Remote Sensing, 2021, 13(4): 645.

KUANG W. Mapping global impervious surface area and green space within urban environments[J]. Science China Earth Sciences,2019, 49(07):1151-1168.

DOI: 10.1007/s11430-018-9342-3. 匡文慧.全球城市人居环境不透水面与绿地空间特征制图[J].中国科学:地球科学,2019,49(07):1151-1168. DOI: 10.1007/s11430-018-9342-3http://dx.doi.org/10.1007/s11430-018-9342-3.

KHAN M M, CHANUSSOT J, CONDAT L, et al. Indusion: Fusion of multispectral and panchromatic images using the induction scaling technique [J]. Ieee Geoscience and Remote Sensing Letters, 2008, 5(1): 98-102.

LIU J G. Smoothing Filter-based Intensity Modulation: A spectral preserve image fusion technique for improving spatial details [J]. International Journal of Remote Sensing, 2010, 21(18): 3461-72.

MASI G, COZZOLINO D, VERDOLIVA L, et al. Pansharpening by Convolutional Neural Networks [J]. Remote Sensing, 2016, 8(7).

NORTH P R J. Estimation of fAPAR, LAI, and vegetation fractional cover from ATSR-2 imagery [J]. Remote Sensing of Environment, 2002, 80(1): 114-21

RANCHIN T, AIAZZI B, ALPARONE L, et al. Image fusion—the ARSIS concept and some successful implementation schemes [J]. ISPRS Journal of Photogrammetry and Remote Sensing, 2003, 58(1-2): 4-18.

SONG D-X, WANG Z, HE T, et al. Estimation and validation of 30 m fractional vegetation cover over China through integrated use of Landsat 8 and Gaofen 2 data [J]. Science of Remote Sensing, 2022, 6.

SZEGEDY C, LIU W, JIA Y Q, et al. Going Deeper with Convolutions [J]. Proc Cvpr Ieee, 2015: 1-9.

SU L. Optimizing support vector machine learning for semi-arid vegetation mapping by using clustering analysis [J]. ISPRS Journal of Photogrammetry and Remote Sensing, 2009, 64(4): 407-13.

VAN DE VOORDE T, VLAEMINCK J, CANTERS F. Comparing Different Approaches for Mapping Urban Vegetation Cover from Landsat ETM+ Data: A Case Study on Brussels [J]. Sensors (Basel), 2008, 8(6): 3880-902.

WALD L. Quality of high resolution synthesised images: Is there a simple criterion ?; proceedings of the Third conference "Fusion of Earth data: merging point measurements, raster maps and remotely sensed images", Sophia Antipolis, France, F 2000-01-26, 2000 [C].

WANG Q, TANG Y, ATKINSON P M. The effect of the point spread function on downscaling continua [J]. ISPRS Journal of Photogrammetry and Remote Sensing, 2020, 168: 251-67.

WANG Q, SHI W, ATKINSON P M. Area-to-point regression kriging for pan-sharpening [J]. ISPRS Journal of Photogrammetry and Remote Sensing, 2016, 114: 151-65.

XIE W, CUI Y, LI Y, et al. HPGAN: Hyperspectral Pansharpening Using 3-D Generative Adversarial Networks [J]. IEEE Transactions on Geoscience and Remote Sensing, 2021, 59(1): 463-77.

XIAO J, MOODY A. A comparison of methods for estimating fractional green vegetation cover within a desert-to-upland transition zone in central New Mexico, USA [J]. Remote Sensing of Environment, 2005, 98(2-3): 237-50.

YANG J F, FU X Y, HU Y W, et al. PanNet: A deep network architecture for pan-sharpening [J]. Ieee I Conf Comp Vis, 2017: 1753-61.

ZENG X, RAO P, DEFRIES R S, et al. Interannual Variability and Decadal Trend of Global Fractional Vegetation Cover from 1982 to 2000 [J]. Journal of Applied Meteorology, 2003, 42(10): 1525-30.

ZHANG H, MA J, CHEN C, et al. NDVI-Net: A fusion network for generating high-resolution normalized difference vegetation index in remote sensing[J]. ISPRS Journal of Photogrammetry and Remote Sensing, 2020, 168: 182-196.

ZHOU W, BOVIK A C. A universal image quality index [J]. IEEE Signal Processing Letters, 2002, 9(3): 81-4.

Alert me when the article has been cited

提交

Downscaling of Landsat 8 land surface temperature products based on deep learning

Remote sensing data fusion by combining STARFM and downscaling mixed pixel algorithm

SAR ship detection through generative knowledge transfer

MtSCCD： Land-use scene classification and change-detection dataset for deep learning

Improving the quality of remotely sensed precipitation product from GPM satellites by using a spatial random forest