Spatial Resolution Matching of Microwave Radiometer Data with Convolutional Neural Network

Abstract

Passive multi-frequency microwave remote sensing is often plagued with the problems of low- and non-uniform spatial resolution. In order to adaptively enhance and match the spatial resolution, an accommodative spatial resolution matching (ASRM) framework, composed of the flexible degradation model, the deep residual convolutional neural network (CNN), and the adaptive feature modification (AdaFM) layers, is proposed in this paper. More specifically, a flexible degradation model, based on the imaging process of the microwave radiometer, is firstly proposed to generate suitable datasets for various levels of matching tasks. Secondly, a deep residual CNN is introduced to jointly learn the complicated degradation factors of the data, so that the resolution can be matched up to fixed levels with state of the art quality. Finally, the AdaFM layers are added to the network in order to handle arbitrary and continuous resolution matching problems between a start and an end level. Both the simulated and the microwave radiation imager (MWRI) data from the Fengyun-3C (FY-3C) satellite have been used to demonstrate the validity and the effectiveness of the method.

Comparison of the Remapping Algorithms for the Advanced Technology Microwave Sounder (ATMS)

“…Some filter-based deconvolution algorithms similar to the AFA have already incorporated the along-scan deformation of FOVs using space-variant PSFs [17,[30][31][32]. In these algorithms, filtering is completed along each column of the image with a PSF at that specific scan position, since the relative geometry changes of the data in the along-track direction almost stay the same.…”

“…Figure 1 illustrates that the relative geometries of the samples change significantly along the scan. Because the remapping algorithms are highly dependent on the overlaps between the raw antenna pattern and the expected one, the geometric deformation of the FOVs over the scan has a very important effect on the remapping algorithms [17,[30][31][32].…”

“…Recently, many deep-learning-based super-resolution technologies have been proposed to improve the image quality of the natural environment [24][25][26][27][28][29]. Since the development by Hu et al [30] of a deconvolution algorithm based on a convolutional neural network, the deep learning technique has been applied to microwave radiometer data to enhance their resolution [31][32][33]. However, among these algorithms, only a few manage to match the intrinsic antenna…”

A New Passive Microwave Tool for Operational Forest Fires Detection: A Case Study of Siberia in 2019

“…If the unmanned helicopter is used as a remote sensing platform, the spatial resolution can be 2.5-25 m. For satellite altitude, H=600-800 km, pixel size ranges from 15-20 km at χ=2.25 cm to 140-190 km at χ=21 cm, while other conditions are equal. If the microwave radiometers of wavelengths 0.8, 2.25 and 21 cm are mounted on an unmanned aircraft vehicle, the DMS can safely detect fire of about 1-2 square meters under the trees at an altitude of about 80-100 m [24][25][26]. Land observation satellites have spatial resolution ranging from tens of meters to one hundred kilometers depending on the sensors.…”

“…Satellite measurements such as vegetation indices Normalized Difference Vegetation Index (NDVI) and Leaf Area Index (LAI) are used to calculate vegetation parameters, including biomass and evapotranspiration [22]. In general, a more detailed examination of the problem of spatial resolution is carried out when using passive multi-frequency microwave remote sensing by several publications [e.g., 24,25]. The IL-18 remote sensing platform provides spatial resolution depending on the H-altitude of the remote sensing platform and the wavelengths of the microwave sensors.…”