A Model-free Four Component Scattering Power Decomposition for Polarimetric SAR Data

S. Dey, A. Bhattacharya, A. C. Frery, C. López-Martínez and Y. S. Rao, “A Model-free Four Component Scattering Power Decomposition for Polarimetric SAR Data,” in IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, Early Access, 2021

➲ Full paper

Summary

Target decomposition methods of polarimetric Synthetic Aperture Radar (PolSAR) data explain scattering information from a target. In this regard, several conventional model-based methods utilize scattering power components to analyze polarimetric SAR data. However, the typical hierarchical process to enumerate power components uses various branching conditions, leading to several limitations. These techniques assume \textit{ad hoc} scattering models within a radar resolution cell. Therefore, the use of several models makes the computation of scattering powers ambiguous. Some common issues of model-based decompositions are related to the compensation of the orientation angle about the radar line of sight and the negative power components’ occurrence. We propose a model-free four-component scattering power decomposition that alleviates these issues. In the proposed approach, we use the non-conventional 3D Barakat degree of polarization to obtain the scattered electromagnetic wave’s polarization state. The degree of polarization is used to obtain the even-bounce, odd-bounce, and diffused scattering power components. Along with this, a measure of target scattering asymmetry is also proposed, which is then suitably utilized to obtain the helicity power. All the power components are roll-invariant, non-negative and unambiguous. In addition to this, we propose an unsupervised clustering technique that preserves the dominance of the scattering power components for different targets. This clustering technique assists in understanding the importance of diverse scattering mechanisms based on target characteristics. The technique adequately captures the clusters’ variations from one target to another according to their physical and geometrical properties.

IEEE GRSS Remote Sensing Training Materials

In the framework of the Global Activities of the GRSS Geoscience and Remote Sensing Society of IEEE, we have released a 18 hour on-line tutorial on East Remote Sensing, recorded both in English and Spanish, through its GRSS YouTube channel. The lectures cover different remote sensing technologies radar, SAR, radiometry, GNSSr, optical, lidar and thermal), as well as techniques and applications.

The English playlist covers the following lectures:

  • Introduction to Remote Sensing
  • Earth Observation Tools (GIS, Clasification…)
  • Radar and SAR Principles
  • Synthetic Aperture Radar: Applications (InSAR, PolSAR, PolInSAR, Multi-temporal, multi-frequency)
  • Microwave Radiometers: Principles, Technologies and Sensors
  • Microwave Radiometers: Applications (incl. SMOS-SM, SM @ 1 km, OS)
  • Multi- and Hyperspectral Sensors: Principles, Technologies and Sensors
  • Multi- and Hyperspectral Sensors: Applications
  • LIDAR: Principles, Technologies and Sensors
  • LIDAR: Applications
  • Agriculture Applications of Synthetic Aperture Radar
  • Statistical Modeling, Processing and Analysis of SAR Images
  • Remote Sensing Using GNSS (reflected) Signals of Opportunity
  • Electromagnetic Scattering in Earth Remote Sensing: the “machine code approach”
  • Thermal Infrared Remote Sensing: Principles and Applications
  • Water Quality Monitoring with Optical Methods
  • Machine Learning for Remote Sensing Data Analysis
  • Real-Time Hyperspectral Imaging

The Spanish playlist covers the following lectures:

  • Introducción a la Teledetección
  • Herramientas para la Observación de la Tierra (GIS, Clasificación)
  • Principios del Radar y el SAR
  • Radar de Apertura Sintética: Aplicaciones (InSAR, PolSAR, PolInSAR, Multitemporal, multifrecuencia)
  • Radiómetros de Microondas: Principios, Tecnologías y Sensores
  • Radiómetros de Microondas: Aplicaciones (incl. SMOS-SM, SM @ 1 km, OS)
  • Sensores Multi e Hiper Espectrales: Principios, Tecnologías y Sensores
  • Sensores Multi e Hiper Espectrales: Aplicaciones
  • LIDAR: Principios, Tecnologías y Sensores
  • LIDAR: Aplicaciones
  • Aplicaciones Agrícolas del Radar de Apertura Sintética
  • Modelado Estadístico, Procesado y Análisis de Imágenes SAR
  • Teledetección basada en GNSS (Reflectrometría) Señales de Oportunidad
  • Retrodispersión Electromagnética en Teledetección Terrestre: una aproximación “machine code”
  • Teledetección Térmica de Infrarrojos: Principios y Aplicaciones
  • Monitorización de la Calidad del Agua Mediante Métodos Ópticos
  • Machine Learning para el Análisis de Datos de Teledetección
  • Imagen Hiper Espectral en Tiempo Real

SinCohMap Project Satellite Radar Interferometry Effective for Mapping Crops

The European Space Agency has recently highlighted our SInCohMap project, and in particular the research leaded by our colleagues Alejandro Mestre-Quereda and Juan M. López-Sánchez from the University of Alicante. This work, entitled “Time-Series of Sentinel-1 Interferometric Coherence and Backscatter for Crop-Type Mapping” has been recently published in the IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing journal.

Crop changes over one year

This work explores the potential use of the interferometric coherence measured with Sentinel-1 satellites as input feature for crop classification. A one-year time-series of Sentinel-1 images acquired over an agricultural area in Spain, in which 17 crop species are present, is exploited for this purpose. Different options regarding temporal baselines, polarization, and combination with radiometric data (backscattering coefficient) are analyzed in the associated pater. The presented results show that both radiometric and interferometric features provide notable classification accuracy when used individually, where the overall accuracy lies between 70% and 80%. It is found that the shortest temporal baseline coherences (6 days) and the use of all available intensity images perform best, hence proving the advantage of the 6-day revisit time provided by the Sentinel-1 constellation with respect to longer revisit times. It is also shown that dual-pol data always provide better classification results than single-pol ones. More importantly, when both coherence and backscattering coefficient are jointly used, a significant increase in accuracy is obtained (greater than 7% in overall accuracies). Individual accuracies of all crop types are increased, and an overall accuracy above 86% is reached. This proves that both features provide complementary information, and that the combination of interferometric and radiometric radar data constitutes a solid information source for this application.

➲ Full open access paper

© Credits

AMERSIE 2020 School on Advanced Methods for RS Information Extraction

In November 2020 we contributed to the online fall school AMERSIE on Advanced methods for remote sensing information extraction, jointly organized by the IEEE Geoscience and Remote Sensing Society chapters of Norway and Spain, within the framework of the ChapNet initiative. The school aimed at introducing to advanced signal processing methods for information retrieval in large scale datasets collected by multiple remote sensors.

The new developments in remote sensing technologies are moving towards smaller, specialized satellite systems, constellations involving many platforms, and exploitation of multi-sensor (multi-modal) observations. Together with the ever-increasing capabilities for higher spatial, spectral, temporal, and radiometric resolution, this results in an explosive growth in the amount of RS data available for Earth science applications. The proliferation of remote sensing data also increases the complexity of these data, like larger diversity and higher dimensionality. This development goes hand in hand with the rise of new, innovative signal processing methodologies to take advantage of the new monitoring capabilities and opportunities.

The school had the following programme:

  • High performance computing for hyperspectral remote sensing information extraction by Antonio Plaza
  • Extracting physical information in multichannel radar remote sensing by Carlos López-Martínez
  • Applied mathematics for SAR and PolSAR information extraction by Avik Bhattacharya
  • Statistical information theory for remote sensing data analysis & Best practice guidelines for publishing a scientific paper by Alejandro Frery
  • Sensing soil and vegetation water with space-borne microwave radiometers by Maria Piles
  • Challenges and opportunities in multimodal RS information extraction by Andrea Marinoni

The schools finished with a panel debato on the role of artificial intelligence on remote sensing data exploitation.

The school material is completely available in the IEEE Geoscience and Remote Sensing Society YouTube channel.

IEEE GRSS Webinar Series – Front Range Chapter

On November 11th, 2020, I was invited by the Front Range Chapter of the IEEE GRSS Geoscience and Remote Sensing Society, located in the Denver Colorado (USA) region, to give a webinar within the Distinguished Lecturer Program. This webinar was also co-sponsored by the University of Colorado, Boulder.

The talk of the webinar was entitled Signal Processing of Polarimetric SAR: Detection and Parameter Extraction, where I presented:

  • The principles of Polarimetric Synthetic Aperture Radar remote sensing imaging for Earth observation
  • How to explore the physical information content of Polarimetric Synthetic Aperture Radar data
  • Several applications of Polarimetric Synthetic Aperture Radar

This time we had more than 80 attendees from different countries.