VISION: VIdeo StabilisatION using automatic features selection for image velocimetry analysis in rivers

VISION is open-source software written in MATLAB for video stabilisation using automatic features detection. It can be applied for any use, but it has been developed mainly for image velocimetry applications in rivers. It includes a number of options that can be set depending on the user’s needs and intended application: 1) selection of different feature detection algorithms (seven to be selected with the flexibility to choose two simultaneously), 2) definition of the percentual value of the strongest features detected to be considered for stabilisation, 3) geometric transformation type, 4) definition of a region of interest on which the analysis can be performed, and 5) visualisation in real-time of stabilised frames. One case study was deemed to illustrate VISION stabilisation capabilities on an image velocimetry experiment. In particular, the stabilisation impact was quantified in terms of velocity errors with respect to field measurements obtaining a significant error reduction of velocities. VISION is an easy-to-use software that may support research operating in image processing, but it can also be adopted for educational purposes.

How to cite: Pizarro, A., S.F. Dal Sasso, S. Manfreda, VISION: VIdeo StabilisatION using automatic features selection for image velocimetry analysis in rivers, SoftwareX, Volume 19,  101173, 2022. [pdf]

Best Paper Award 2022

I’m proud to announce that our recent paper “Current Practices in UAS-based Environmental Monitoring” has been awarded with the Best Paper Award 2022 by Remote Sensing MDPI. This paper has been the result of a joint effort carried out by several authors collaborating within the framework of COST Action – HARMONIOUS

PAPERS AWARD

The recognition should be shared with the colleagues Goran Tmušić, Salvatore Manfreda, Helge Aasen, Mike R James, Gil Gonçalves, Eyal Ben-Dor, Anna Brook, Maria Polinova, Jose Juan Arranz, János Mészáros, Ruodan Zhuang, Kasper Johansen, Yoann Malbeteau, Isabel Pedroso de Lima, Corine Davids, Sorin HerbanMatthew McCabe. Project was funded by COST Association – European Cooperation in Science and Technology

In Situ Observation-Constrained Global Surface Soil Moisture Using Random Forest Model

The inherent biases of different long-term gridded surface soil moisture (SSM) products, unconstrained by the in situ observations, implies different spatio-temporal patterns. In this study, the Random Forest (RF) model was trained to predict SSM from relevant land surface feature variables (i.e., land surface temperature, vegetation indices, soil texture, and geographical information) and precipitation, based on the in situ soil moisture data of the International Soil Moisture Network (ISMN). The results of the RF model show an RMSE of 0.05 m3 m−3 and a correlation coefficient of 0.9. The calculated impurity-based feature importance indicates that the Antecedent Precipitation Index affects most of the predicted soil moisture. The geographical coordinates also significantly influence the prediction (i.e., RMSE was reduced to 0.03 m3 m−3 after considering geographical coordinates), followed by land surface temperature, vegetation indices, and soil texture. The spatio-temporal pattern of RF predicted SSM was compared with the European Space Agency Climate Change Initiative (ESA-CCI) soil moisture product, using both time-longitude and latitude diagrams. The results indicate that the RF SSM captures the spatial distribution and the daily, seasonal, and annual variabilities globally.

How to cite: Zhang, L.; Zeng, Y.; Zhuang, R.; Szabó, B.; Manfreda, S.; Han, Q.; Su, Z. In Situ Observation-Constrained Global Surface Soil Moisture Using Random Forest Model. Remote Sens. 202113, 4893. [pdf]

Characterizing vegetation complexity with unmanned aerial systems (UAS) – A framework and synthesis

Ecosystem complexity is among the important drivers of biodiversity and ecosystem functioning, and unmanned aerial systems (UASs) are becoming an important tool for characterizing vegetation patterns and processes. The variety of UASs applications is immense, and so are the procedures to process UASs data described in the literature. Optimizing the workflow is still a matter of discussion. Here, we present a comprehensive synthesis aiming to identify common rules that shape workflows applied in UAS-based studies facing complexity in ecosystems. Analysing the studies, we found similarities irrespective of the ecosystem, according to the character of the property addressed, such as species composition (biodiversity), ecosystem structure (stand volume/complexity), plant status (phenology and stress levels), and dynamics (disturbances and regeneration). We propose a general framework allowing to design UAS-based vegetation surveys according to its purpose and the component of ecosystem complexity addressed. We support the framework by detailed schemes as well as examples of best practices of UAS studies covering each of the vegetation properties (i.e. composition, structure, status and dynamics) and related applications. For an efficient UAS survey, the following points are crucial: knowledge of the phenomenon, choice of platform, sensor, resolution (temporal, spatial and spectral), model and classification algorithm according to the phenomenon, as well as careful interpretation of the results. The simpler the procedure, the more robust, repeatable, applicable and cost effective it is. Therefore, the proper design can minimize the efforts while maximizing the quality of the results.

How to cite: Müllerová J. , X. Gago, M. Bučas,J. Company, J. Estrany, J. Fortesa, S. Manfreda, A. Michez, M. Mokroš, G. Paulus, E. Tiškus, M. A. Tsiafouli, R. Kent, Characterizing vegetation complexity with unmanned aerial systems (UAS) – A framework and synthesis, Ecological Indicators, Volume 131, November 2021, 108156. [pdf]

Special Issue “Global Gridded Soil Information Based on Machine Learning”

Dear Colleagues,


Recent technological advances in both remote sensing and soil mapping approaches and progress in establishing harmonized soil profile datasets have opened up the potential to derive global gridded soil information. This has been possible because worldwide researchers have gained a growing experience in building standardized soil profile datasets with measured physical, chemical data and taxonomical information; filling data gaps; using Earth observation data for soil mapping; optimizing soil sampling strategy; processing big data; applying machine learning algorithms; and assessing uncertainty; which support the preparation of global soil maps with increasing accuracy and spatiotemporal resolution.

Data-intensive computing solutions to process and analyze the exploding amount of environmental information are continuously updated. Machine learning algorithms are among the most frequently used tools for data preprocessing and describing the complex relationship between soil properties and environmental covariates with the ability to assess the uncertainty of the predictions. One of the greatest challenges in deriving global gridded soil information is to make the most of the predictive power of machine learning algorithms with the continuously increasing amount of environmental information. This Special Issue is dedicated to machine learning-based methods in:

  • proximal and digital global mapping of soil properties (e.g., basic, hydraulic, thermal, functional, ecosystem services);
  • computing systems/algorithms/approaches using Earth observation data to derive global gridded soil datasets;
  • preprocessing Earth observation data to feed into global soil mapping;
  • data-intensive computing methods for incorporating Earth observation data for predictive soil mapping;
  • optimizing temporal resolution to globally track the changes of soil properties,
  • uncertainty assessment of the derived gridded soil information;
  • specifying algorithms to local soil specificities in, e.g., proximal soil mapping;
  • the engagement of remote sensing data with digital soil mapping;
  • downscaling of large-scale soil feature;
  • other related topics.

Review contributions on the abovementioned topics are welcomed as well.Dr. Brigitta Szabó (Tóth)
Prof.Dr. Eyal Ben-Dor
Dr. Yijian Zeng
Prof.Dr. Salvatore Manfreda
Dr. Madlene Nussbaum
Guest Editors

Keywords

  • Global gridded soil information
  • Predictive soil mapping
  • Uncertainty assessment
  • Spectral data
  • Parallel distributive platforms
  • Machine learning
  • Digital soil mapping

An Integrative Information Aqueduct to Close the Gaps between Satellite Observation of Water Cycle and Local Sustainable Management of Water Resources

The past decades have seen rapid advancements in space-based monitoring of essential water cycle variables, providing products related to precipitation, evapotranspiration, and soil moisture, often at tens of kilometer scales. Whilst these data effectively characterize water cycle variability at regional to global scales, they are less suitable for sustainable management of local water resources, which needs detailed information to represent the spatial heterogeneity of soil and vegetation. The following questions are critical to effectively exploit information from remotely sensed and in situ Earth observations (EOs): How to downscale the global water cycle products to the local scale using multiple sources and scales of EO data? How to explore and apply the downscaled information at the management level for a better understanding of soil-water-vegetation-energy processes? How can such fine-scale information be used to improve the management of soil and water resources? An integrative information flow (i.e., iAqueduct theoretical framework) is developed to close the gaps between satellite water cycle products and local information necessary for sustainable management of water resources. The integrated iAqueduct framework aims to address the abovementioned scientific questions by combining medium-resolution (10 m–1 km) Copernicus satellite data with high-resolution (cm) unmanned aerial system (UAS) data, in situ observations, analytical- and physical-based models, as well as big-data analytics with machine learning algorithms. This paper provides a general overview of the iAqueduct theoretical framework and introduces some preliminary results.

Concept Diagram

How to cite: Su, Z.; Zeng, Y.; Romano, N.; Manfreda, S.; Francés, F.; Dor, E.B.; Szabó, B.; Vico, G.; Nasta, P.; Zhuang, R.; Francos, N.; Mészáros, J.; Sasso, S.F.D.; Bassiouni, M.; Zhang, L.; Rwasoka, D.T.; Retsios, B.; Yu, L.; Blatchford, M.L.; Mannaerts, C. An Integrative Information Aqueduct to Close the Gaps between Satellite Observation of Water Cycle and Local Sustainable Management of Water Resources. Water 202012, 1495. [pdf]

Quantifying Long-term Land Surface and Root Zone Soil Moisture over Tibetan Plateau

It is crucial to monitor the dynamics of soil moisture over the Tibetan Plateau, while considering its important role in understanding the land-atmosphere interactions and their influences on climate systems (e.g., Eastern Asian Summer Monsoon). However, it is very challenging to have both the surface and root zone soil moisture (SSM and RZSM) over this area, especially the study of feedbacks between soil moisture and climate systems requires long-term (e.g., decadal) datasets. In this study, the SSM data from different sources (satellites, land data assimilation, and in-situ measurements) were blended while using triple collocation and least squares method with the constraint of in-situ data climatology. A depth scaling was performed based on the blended SSM product, using Cumulative Distribution Function (CDF) matching approach and simulation with Soil Moisture Analytical Relationship (SMAR) model, to estimate the RZSM. The final product is a set of long-term (~10yr) consistent SSM and RZSM product. The inter-comparison with other existing SSM and RZSM products demonstrates the credibility of the data blending procedure used in this study and the reliability of the CDF matching method and SMAR model in deriving the RZSM.

How to cite: Zhuang, R.; Zeng, Y.; Manfreda, S.; Su, Z. Quantifying Long-term Land Surface and Root Zone Soil Moisture over Tibetan Plateau. Remote Sens. 202012, 509. [pdf]

Soil Moisture Monitoring in Iran by Implementing Satellite Data into the Root-Zone SMAR Model

Monitoring Surface Soil Moisture (SSM) and Root Zone Soil Moisture (RZSM) dynamics at the regional scale is of fundamental importance to many hydrological and ecological studies. This need becomes even more critical in arid and semi-arid regions, where there are a lack of in situ observations. In this regard, satellite-based Soil Moisture (SM) data is promising due to the temporal resolution of acquisitions and the spatial coverage of observations. Satellite-based SM products are only able to estimate moisture from the soil top layer; however, linking SSM with RZSM would provide valuable information on land surface-atmosphere interactions. In the present study, satellite-based SSM data from Soil Moisture and Ocean Salinity (SMOS), Advanced Microwave Scanning Radiometer 2 (AMSR2), and Soil Moisture Active Passive (SMAP) are first compared with the few available SM in situ observations, and are then coupled with the Soil Moisture Analytical Relationship (SMAR) model to estimate RZSM in Iran. The comparison between in situ SM observations and satellite data showed that the SMAP satellite products provide more accurate description of SSM with an average correlation coefficient (R) of 0.55, root-mean-square error (RMSE) of 0.078 m3 m-3 and a Bias of 0.033 m3 m-3. Thereafter, the SMAP satellite products were coupled with SMAR model, providing a description of the RZSM with performances that are strongly influenced by the misalignment between point and pixel processes measured in the preliminary comparison of SSM data.

How to cite: Gheybi, F., P. Paridad, F. Faridani, A. Farid, A. Pizarro, M. Fiorentino and S. Manfreda, Soil Moisture Monitoring in Iran by Implementing Satellite Data into the Root-Zone SMAR Model, Hydrology 2019, 6, 44 (doi: 10.3390/hydrology6020044), 2019. [pdf]

Emerging earth observing platforms offer new insights into hydrological processes

Data, and its timely delivery, presents one of the major constraints in advancing the hydrological sciences. Traditional monitoring techniques are time consuming, expensive, and discontinuous in space and time. Moreover, field observations are influenced by instrumental degradation and human errors. While providing the foundation upon which much of our hydrological knowledge is based, new observational strategies are required to drive further understanding and insights. Recent advances in earth observation (EO) technologies present a new frontier for hydrologic monitoring and process description.

Figure: UAS derived 3D dense point cloud, (B) mesh model, and (C) tiled model derived from a UAS based survey of an earthen dam next to the village of Pischia (Timisoara, Romania). Such data provide the framework for development of high-resolution flood modeling, urban watershed mapping and civil engineering design and map updating.

How to cite: Manfreda, S., M.F. McCabe. Emerging earth observing platforms offer new insights into hydrological processesHydrolink, 1, 8-9, 2019.  [pdf]

Data Fusion Through Bayesian Methods for Flood Monitoring from Remotely Sensed Data

Producing high-precision flood maps requires integrating and correctly classifying information coming from heterogeneous sources. Methods to perform such integration have to rely on different knowledge bases. A useful tool to perform this task consists in the use of Bayesian methods to assign probabilities to areas being subject to flood phenomena, fusing a priori information and modeling with data coming from radar or optical imagery. In this chapter we review the use of Bayesian networks, an elegant framework to cast probabilistic descriptions of complex systems, applied to flood monitoring from multi-sensor, multi-temporal remotely sensed and ancillary data.

How to cite: Annarita D’Addabbo, Alberto Refice, Domenico Capolongo, Guido Pasquariello and Salvatore Manfreda, Data Fusion Through Bayesian Methods for Flood Monitoring from Remotely Sensed Data, Springer International Publishing AG 2018, Pages 181 – 208, (doi: https://doi.org/10.1007/978-3-319-63959-8_8), 2018. [pdf]