Initiatives

Remote Sensing for All Sectors

Studying the interaction between human activities and natural processes in the urban environment requires many hyperspectral capabilities such as synthesis of disparate datasets, statistical tools, modelling approaches, and appropriate visualizations. The lab will serve as a catalyst for space sciences and associated remote sensing activities, and the benefits that could bring to our federal institutions, UAE Space Agency and Zayed University, in terms of:

1. Producing research outputs, on the fundamentals and applications of hyperspectral data unmixing, of the highest quality

2. Supporting and develop further excellent faculty, staff and students from the UAE Space Agency Science Group

3. Developing positive impact on the local and regional economy and society.

The main initiative of the of the laboratory is to enable and catalyze interdisciplinary research efforts that advance our understanding of hyperspectral processes and spatial relationships emphasizing earth and planetary environmental systems through improved theory, methods, technology, data, and community outreach. The laboratory will be focused initially on hyperspectral analysis of UAE environments. It builds on ZU’s existing strength in environmental sciences. Currently, scattered research groups pursue basic research in the science and application of hyper-spectroscopy, remote sensing algorithm and codes. We envision the lab that supports remote sensing research in four primary areas: scientific applications, theory and tool development, ground and airborne data, and education.

Planetary Observation with Emphasis on Mars

Environmental monitoring (e.g. Dust, NOx ) using satellite data: ZU have past studies on the terrestrial dust storms using remote sensing techniques. One related activity that the lab will emphasize on is to use the hyper spectral analyses (UV-VIS-NIR-FAR) of Mars dust to reveal its quality, mineralogy, and morphological characteristics, its patterns or even its grain size, and relate this to topography, and atmosphere, etc. An initial step toward this research element would be to produce a reference hyperspectral library for spectra and images of the Martian dust grain sizes and their mineralogy. The experiments, which will be performed at the lab, shall consider different scenarios and controlling variables similar to those on Mars (low pressure, low temperature, and substrate, etc.). This approach is necessary in order to have new guidelines for interpretation-correlation with expected data coming from the UAE Mars mission. For this purpose, we can use GER 3700, FMCW Radar Sensors and Visible and Near Infrared Cameras, low vacuum chamber, and other tools as well.

It will be useful to introduce some interference to the measurements, and test the behavior of spectra and signals before and after the data correction. ENVI Software could be suitable to do some information extraction, image cubes, and corrections. Even without this component, it will be very interesting to empirically determine and assign certain behavior and characteristics of the reflectance spectra and radar backscatter to the dust, its grain size, mineralogy and other related characteristics (especially under Martian conditions). To have a unique experimental setting, one needs to review the data from other missions before designing the ground-based experiment. This will help us to look at the research question from different angles. Once, we establish the likely scenarios from theoretical and practical point of view, it will becomes possible to match and compare our findings with the data coming from the mission. And, during the progress of this work and associated experiments, we could advise on the sensors calibration, etc. The new data from the mission and the insight from the experiments will help us assemble a new model about dust storms pattern, dust mineralogy and their sizes on mars, and give indirect insights on the atmospheric patterns on mars. The lab scientist can also give 2-3 hrs training/ lecture on findings and project progress for interested members from the UAE space community.

Develop UAE Space Agency Spectral Library

Much research has been devoted to spectral un-mixing of hyperspectral data, which is an efficient way to solve standard problems encountered in hyperspectral imagery. These problems include pixel classification, material quantification and subpixel detection. Spectral un-mixing aims at decomposing a pixel spectrum into a collection of elementary material spectra, usually referred to as end-members, and at estimating the corresponding proportions or abundances. To describe the mixture, the most frequently encountered model is the macroscopic model which assumes that the observed pixel spectrum is a weighted linear combination of the end-member spectra. Efficient supervised or unsupervised algorithms based on this linear model have been proposed in the literature to solve the un-mixing problem, which can be considered as a matrix factorization task. However, in nature many types of mixtures exists beside the linear types of mixing e.g. intimates, coating, molecular mixtures and their combinations. Hence, the laboratory main activates emphasizes on the possibility of spectral mixture specifically in different environment and suggests the incorporation of such spectral mixing in different proportions in the process of development of spectral libraries. As an alternative to the use of very limited hyperspectral satellite image, it seeks for the solution to spatial feature discrimination with a combination of laboratory-generated hyperspectral signatures. The spectroradiometer can be employed to atmospheric studies, aerosol estimation and vegetation fluorescence detection.

Theory and Tools

The capabilities of hyperspectral remote sensing to describe, explain, simulate, and optimize various spatial and temporal processes is hindered by the availability and reliability of some theories and tools which we will be working on to improve. Generalization and refinement of these tools by constructing new data models, developing sensor technologies, improving spatial cognition theory, applying advances in spatial statistics, strengthening decision support tools, bridging semantic gaps with empirical studies, and visualization are a comparative researchbased strength of the proposed lab. Hyperspectral data representing the human and natural environment are highly heterogeneous in format, quality, size, and reference frame. A goal of the lab will be to provide straightforward access to multi-purpose data sets in support of the research described above.

Education

In many fields, such as environmental investigation and urban planning, remote sensing are essential tools. Knowledge of their application is a minimum requirement. Students and professionals require education and training. To enhance continuing Education, this lab shall provide training to professionals from a wide variety of careers. Many of these professionals can in turn provide synergistic opportunities through student internships, and data exchange/access. We could also target a faculty hire in remote sensing for education with possible linkage with the lab on grant proposals. This lab can also be accessible for teaching and for graduate and undergraduate research projects from Zayed University and other universities in the UAE.

Students and Faculty Engagements

Engaging students and faculty in the lab activities will be through the cooperative research projects, and through involving them in the outreach programs of the lab. Involve students in related internship opportunities. Basically, most of our students are end users for the data coming from space. Few faculty have interests and experiences in hyperspectral remote sensing, geosciences, geomorphology and life in extreme environment (including Mars), among other lines about terrestrial environment. Involve students in studying the terrestrial dust storms using remote sensing techniques. One can use the hyper spectral analyses (UV-VIS-NIRFAR) of Mars dust to reveal its quality, mineralogy, and morphological characteristics, its patterns or even its grain size, and relate this to topography, atmosphere. etc.

The UAE Space Agency Hyperspectral Laboratory will provide a catalyst for; industry engagement; delivery of international quality research; publication of research papers in internationally respected journals; engagement of students in research; development of under/graduate and professional academic development education provision; development of consulting activities among others. The proposed lab will offer a significant level of expertise, innovation and technical excellence providing a high level of service to all clients. The proposed lab will ensure client communication and involvement is maintained through all stages of project delivery. Dedicated project managers will be assigned to each project with a view to holding regular progress meetings are site visits when required.

Main Other Activities

The proposed hyperspectral remote sensing lab will focus on interdisciplinary activities in remote sensing and related disciplines (including, but not limited to, remote sensing physics, imaging science, space and planetary sciences, and the application of remote sensing, particularly in environmental and ecosystem science). Through its wide range of activities, the center will have a significant impact on the society and quality of life in the UAE. The center will work and coordinate with key stakeholders from government, academia, financial institutions, the business community, industry and private foundations to focus on the key roles played by research and technology-based innovation in implementing the goals of the UAE’s new National Innovation Strategy especially in space sector; below are an outline of the center’s main activities:

The Lab will undertake the following activities:

1. Organize one conference annually on a theme relevant to the mission and vision of the lab, in collaboration with the Center’s principal governmental and/or non-governmental partners,

2. Organize one symposium every year promoting regional dialogue on key issues pertaining environmental problems and associated solutions

3. Organize one workshop each year, in collaboration with key collaborative partners, along with academics and students of the university, focusing upon specific aspect of the environmental dimensions of sustainable environment and related innovations.

4. Issue online bulletins on standardization of use of spectroradiometer. An important primary task of the project work is to use different spectroradiometers capable of performing hyperspectral measurements and to standardize its method of use. hyperspectral signatures will be generated in laboratory throughout the ultravioletvisible-near-infrared (UV-vis-NIR vs FIR) region (400-1000-3200 nm) with 1 nm resolution for their mixtures within a pixel; and will be posted on the lab website.

5. Continue to work on the mixed pixel analysis and unmixing; objects, specifically in most cases may have dimensions smaller than the spatial resolution of the satellite sensor so that a mixture of objects fills up the field of view. Consequently, a pixel in the satellite image or our own ground and airborne platforms consists of a mixture of spectral signatures. The present line of research emphasizes on the point that such situations of spectral mixing should be incorporated in the development of spectral libraries. Spectral signatures can be derived from hyperspectral satellite images or our own data, but those are very limited in availability. This work proposes a methodology of analyzing mixed pixel signatures using a combination of hyperspectral laboratory data and empirical data; and will be published and updated frequently.

6. Issue online bulletins on solar irradiance studies. The diurnal, seasonal and atmospheric variations of the solar irradiance spectra throughout UV-vis-NIR wavelengths will be studied for UAE environment. The change of the atmospheric oxygen absorption band depth, a hyperspectral signature at around 760 nm will be monitored; and the impact of aerosol and atmosphere will be studied. All these activities and initiatives will be posted on the lab interactive and dynamic webpage