HICO Image Processing System – Update

(14-Dec-2016) Please pardon the interruption. The HICO Image Processing System is currently being migrated to another hosting platform and will be back up and running soon. Stay tuned here for an announcement once the maintenance is complete. Thank you.

HICO IPS

A Year of Hyperspectral Image Processing in the Cloud – HICO IPS reaches a milestone

It has been a little more than one year since we first launched the HICO Image Processing System (HICO IPS), and its performance continues to be exceptional. In fact, as a prototype, HICO IPS has exceeded all expectations, working flawlessly since it was first launched in May 2015.

HICO IPS

HICO IPS is a web-application for on-demand remote sensing image analysis that allows users to interactively select images and algorithms, dynamically launch analysis routines, and then see results displayed directly in an online map interface. More details are as follows:

  • System developed to demonstrate capabilities for remote sensing image analysis in the cloud
  • Software stack utilizes a combination of commercial and open-source software
  • Core image processing and data management performed using ENVI Services Engine
  • Operational system hosted on Rackspace cloud server
  • Utilizes imagery from the Hyperspectral Imager for the Coastal Ocean (HICO), which was deployed on the International Space Station (ISS) from 2009-2014
  • Example algorithms are included for assessing coastal water quality and other nearshore environmental conditions
  • Application developed in collaboration between HySpeed Computing and Exelis Visual Information Solutions (now Harris Geospatial Solutions)
  • Project supported by the Center for the Advancement of Science in Space (CASIS)

And here’s a short overview of HICO IPS accomplishments and performance in the past year, including some infographics to help illustrate how the system has been utilized:

  • The application has received over 5000 visitors
  • Users represent over 100 different countries
  • System has processed a total of 1000 images
  • Equivalent area processed is 4.5 million square kilometers
  • The most popular scene selected for analysis was the Yellow River
  • The most popular algorithm was Chlorophyll followed closely by Land Mask
  • Application has run continuously without interruption since launch in May 2015

HICO IPS infographics

Try it out today for yourself: http://hyspeedgeo.com/HICO/

Remote Sensing in the Cloud – Characterizing shallow coastal environments using HICO IPS

This is the last in a series of posts devoted to reviewing the algorithms currently implemented in the cloud-based HICO Image Processing System (HICO IPS). Links to the other posts in this series are provided below. Here we provide an overview of the algorithm utilized for retrieving multiple layers of water and habitat information for shallow coastal environments.

HICO IPS

Objective – Demonstrate the capability to simultaneously derive information on water properties, water depth and seabed features in shallow coastal environments using a complex hyperspectral image processing algorithm.

Algorithm – Derives inherent optical properties for absorption and backscattering (phytoplankton absorption, detritus and gelbstoff absorption, particle backscattering, total absorption, and total backscattering [m-1]), bottom albedo (reflectance at 550 nm), and water depth (depth [m]) using a semi-analytical inversion model (Goodman and Ustin 2007; Lee et al. 1999, 1998).

Note that this algorithm uses a computationally intensive optimization scheme, which can result in long processing times for large areas. As a reference, the example shown here required approximately 15 minutes to generate output. It is suggested that users select relatively small regions-of-interest when implementing this algorithm.

Inputs – User specified HICO scene, with optional region-of-interest; optional NDWI land/water mask, with user adjustable NDWI threshold; and specification of desired output parameter.

HICO IPS Shark Bay subset

Output – Selected output parameter depicted using a blue-red color ramp, where blue represents low values and red represents high values. If the NDWI land/water mask was selected, then the selected optical property is only calculated and mapped for the water pixels.

HICO IPS Shark Bay aquacor

Try it out today for yourself: http://hyspeedgeo.com/HICO/

 

Related posts

Introducing the HICO Image Processing System

Calculating a land/water mask using HICO IPS

Deriving chlorophyll concentration using HICO IPS

Evaluating water optical properties using HICO IPS

 

References

Goodman J, Ustin SL (2007) Classification of benthic composition in a coral reef environment using spectral unmixing, Journal of Applied Remote Sensing, vol. 1(1), 011501-011501.

Lee Z, Carder KL, Mobley CD, Steward RG, Patch JS (1999) Hyperspectral remote sensing for shallow waters. 2. Deriving bottom depths and water properties by optimization, Applied Optics, vol. 38(18), 3831-3843.

Lee Z, Carder KL, Mobley CD, Steward RG, Patch JS (1998) Hyperspectral remote sensing for shallow waters. I. A semianalytical model, Applied Optics, vol. 37(27), 6329-6338.

Remote Sensing in the Cloud – Evaluating water optical properties using HICO IPS

This is part of an ongoing series dedicated to reviewing the algorithms currently implemented in the cloud-based HICO Image Processing System (HICO IPS). Links to additional posts in this series describing the other algorithms are provided below. Here we provide an overview of the algorithm utilized for evaluating water optical properties in coastal and oceanic water.

HICO IPS

Objective – Retrieve water optical properties for coastal and oceanic water from hyperspectral imagery using a generalized multi-band algorithm.

Algorithm – Estimate water optical properties for absorption and backscattering (specifically, total absorption, phytoplankton absorption, detritus and gelbstoff absorption, total backscattering, and particle backscattering) using the Quasi-Analytical Algorithm (QAA v5; Lee et al. 2009, 2002).

Inputs – User specified HICO scene, with optional region-of-interest; optional NDWI land/water mask, with user adjustable NDWI threshold; and specification of desired optical property.

HICO IPS Turkish Straits

Output – Selected water optical property at 438 nm (m-1) depicted using a blue-red color ramp where blue represents low values and red represents high values. If the NDWI land/water mask was selected, then the selected optical property is only calculated and mapped for the water pixels.

HICO IPS Turkish Straits QAA

Try it out today for yourself: http://hyspeedgeo.com/HICO/

 

Related posts

Introducing the HICO Image Processing System

Calculating a land/water mask using HICO IPS

Deriving chlorophyll concentration using HICO IPS

Characterizing shallow coastal environments using HICO IPS

 

References

Lee Z, Lubac B, Werdell J, Arnone R (2009) An update of the quasi-analytical algorithm (QAA_v5), International Ocean Color Group Software Report, 9 pp.

Lee Z, Carder KL, Arnone RA (2002) Deriving inherent optical properties from water color: a multiband quasi-analytical algorithm for optically deep waters, Applied optics, vol. 41(27), 5755-5772.

Remote Sensing in the Cloud – Deriving chlorophyll concentration using HICO IPS

Continuing our review of the algorithms currently implemented in the cloud-based HICO Image Processing System (HICO IPS), here we provide an overview of the two algorithms utilized for deriving estimates of chlorophyll concentration in oceanic water.

HICO IPS

Objective – Implement multiple algorithms for estimating chlorophyll concentration, as well as a methodology for evaluating the difference between these algorithms.

Algorithms – Derive estimates of surface chlorophyll-a concentration using one of two ocean color algorithms, OC4 (O’Reilly et al. 2000) or OCI (Hu et al. 2012); or for comparison, the difference between these two algorithms (OC4 – OCI). These algorithms first perform spectral resampling of the HICO hyperspectral data to the multispectral SeaWiFS bands on which the algorithms are based.

Inputs – User specified HICO scene, with optional region-of-interest; optional NDWI land/water mask, with user adjustable NDWI threshold; and specification of desired chlorophyll algorithm.

HICO IPS Chesapeake Bay

Output – Surface chlorophyll-a concentration (mg/m^3) depicted using a blue-red color ramp where blue represents low chlorophyll concentration and red represents high concentration. If the NDWI land/water mask was selected, then chlorophyll concentrations are only calculated and mapped for the water pixels (as is logical).

HICO IPS Chesapeake Bay Chlorophyll

Try it out today for yourself: http://hyspeedgeo.com/HICO/

 

Related posts

Introducing the HICO Image Processing System

Calculating a land/water mask using HICO IPS

Evaluating water optical properties using HICO IPS

Characterizing shallow coastal environments using HICO IPS

 

References

Hu C, Lee Z, Franz BA (2012) Chlorophyll-a algorithms for oligotrophic oceans: A novel approach based on three-band reflectance difference, Journal of Geophysical Research, vol. 117(C1), 25 pp.

O’Reilly JE, Maritorena S, O’Brien MC, et al. (2000) SeaWiFS postlaunch calibration and validation analyses, Part 3, NASA Technical Memorandum 2000-206892, vol. 11, 49 pp.

Remote Sensing in the Cloud – Calculating a land/water mask using HICO IPS

Last year we launched the HICO Image Processing System (HICO IPS) – a prototype web application for on-demand remote sensing data analysis in the cloud.

HICO IPS

To demonstrate the capabilities of this system, we implemented a collection of coastal remote sensing algorithms to produce information on water quality, water depth and benthic features using example imagery from the HICO instrument on the International Space Station.

As the HICO IPS approaches its one year anniversary, and continues its excellent performance, we’d like to take a moment to highlight each of the algorithms currently implemented in the system.

Here we begin with an overview of the land/water mask utilized in the HICO IPS.

Objective – Implement an automated algorithm for classifying land versus water, thereby masking land pixels from further analysis and allowing subsequent processing steps to focus on just water pixels.

Algorithm – Generates a binary mask differentiating land from water using the Normalized Difference Water Index (NDWI; McFeeters 1996). This algorithm can be implemented on its own, or as a pre-processing step in other algorithm workflows.

Inputs – User specified HICO scene, with optional region-of-interest; and user adjustable NDWI threshold, where -1.0 ≤ NDWI ≤ 1.0, land ≤ threshold < water, and default threshold = 0.0.

HICO IPS Christchurch

Output – Binary land/water mask (0 = land; 1 = water), where land is displayed in the online map using a black mask and water remains unchanged.

HICO IPS Christchurch mask

Reference – McFeeters SK (1996) The use of the Normalized Difference Water Index (NDWI) in the delineation of open water features, International Journal of Remote Sensing, vol. 17(7), 1425-1432.

Try it out today for yourself: http://hyspeedgeo.com/HICO/

 

Related posts

Introducing the HICO Image Processing System

Deriving chlorophyll concentration using HICO IPS

Evaluating water optical properties using HICO IPS

Characterizing shallow coastal environments using HICO IPS

From here to there – and everywhere – with Geospatial Cloud Computing

Reposted from Exelis VIS, Imagery Speaks, June 30, 2015, by James Goodman, CEO HySpeed Computing.

In a previous article we presented an overview of the advantages of cloud computing in remote sensing applications, and described an upcoming prototype web application for processing imagery from the HICO sensor on the International Space Station.

First, as a follow up, we’re excited to announce availability of the HICO Image Processing System – a cloud computing platform for on-demand remote sensing image analysis and data visualization.

HICO IPS - Chesapeake Bay - Chlorophyll

HICO IPS allows users to select specific images and algorithms, dynamically launch analysis routines in the cloud, and then see results displayed directly in an online map interface. System capabilities are demonstrated using imagery collected by the Hyperspectral Imager for the Coastal Ocean (HICO) on the International Space Station, and example algorithms are included for assessing coastal water quality and other nearshore environmental conditions.

This is an application-server, and not just a map-server. Thus, HICO IPS is delivering on-demand image processing of real physical parameters, such as chlorophyll concentration, inherent optical properties, and water depth.

The system was developed using a combination of commercial and open-source software, with core image processing performed using the recently released ENVI Services Engine. No specialized software is required to run HICO IPS. You just need an internet connection and a web browser to run the application (we suggest using Google Chrome).

Beyond HICO, and beyond the coastal ocean, the system can be configured for any number of different remote sensing instruments and applications, thus providing an adaptable cloud computing framework for rapidly implementing new algorithms and applications, as well as making these applications and their output readily available to the global user community.

However, this is but one application. Significantly greater work is needed throughout the remote sensing community to leverage these and other exciting new tools and processing capabilities. To participate in a discussion of how the future of geospatial image processing is evolving, and see a presentation of the HICO IPS, join us at the upcoming ENVI Analytics Symposium in Boulder, CO, August 25-26.

With this broader context in mind, and as a second follow-up, we ask the important question when envisioning this future of how we as an industry, and as a research community, are going to get from here to there?

The currently expanding diversity and volume of remote sensing data presents particular challenges for aggregating data relevant to specific research applications, developing analysis tools that can be extended to a variety of sensors, efficiently implementing data processing across a distributed storage network, and delivering value-added products to a broad range of stakeholders.

Based on lessons learned from developing the HICO IPS, here we identify three important requirements needed to meet these challenges:

  • Data and application interoperability need to continue evolving. This need speaks to the use of broadly accessible data formats, expansion of software binding libraries, and development of cross-platform applications.
  • Improved mechanisms are needed for transforming research achievements into functional software applications. Greater impact can be achieved, larger audiences reached, and application opportunities significantly enhanced, if more investment is made in remote sensing technology transfer.
  • Robust tools are required for decision support and information delivery. This requirement necessitates development of intuitive visualization and user interface tools that will assist users in understanding image analysis output products as well as contribute to more informed decision making.

These developments will not happen overnight, but the pace of the industry indicates that such transformations are already in process and that geospatial image processing will continue to evolve at a rapid rate. We encourage you to participate.

About HySpeed Computing: Our mission is to provide the most effective analysis tools for deriving and delivering information from geospatial imagery. Visit us at hyspeedcomputing.com.

To access the HICO Image Processing System: http://hyspeedgeo.com/HICO/

HICO Image Gallery – Looking beyond the data

This slideshow requires JavaScript.

What’s in an image? Beyond the visual impact, beyond the pixels, and beyond the data, there’s valuable information to be had. It just takes the right tools to extract that information.

With that thought in mind, HySpeed Computing created the HICO Image Processing System to make these tools readily available and thereby put image processing capabilities directly in your hands.

The HICO IPS is a prototype web application for on-demand remote sensing image analysis in the cloud. It’s available through your browser, so it doesn’t require any specialized software, and you don’t have to be a remote sensing expert to use the system.

HICO, the Hyperspectral Imager for the Coastal Ocean, operating on the International Space Station from 2009-2014, is the first space-based imaging spectrometer designed specifically to measure the coastal environment. And research shows that substantial amounts of information can be derived from this imagery.

To commemorate the recent launch of the HICO IPS and celebrate the beauty of our coastal environment, we’ve put together a gallery highlighting some of the stunning images acquired by HICO that are available in the system.

We hope you enjoy the images, and encourage you to explore the HICO IPS web application to try out your own remote sensing analysis.

HICO IPS: Chesapeake Bay Chla

To access the HICO Image Processing System: http://hyspeedgeo.com/HICO/

For more information on HICO: http://hico.coas.oregonstate.edu/

Sunglint Correction in Airborne Hyperspectral Images Over Inland Waters

Announcing recent publication in Revista Brasileira de Cartographia (RBC) – the Brazilian Journal of Cartography. The full text is available open-access online: Streher et al., 2014, RBC, International Issue 66/7, 1437-1449.

Title: Sunglint Correction in Airborne Hyperspectral Images Over Inland Waters

Authors: Annia Susin Streher, Cláudio Clemente Faria Barbosa, Lênio Soares Galvão, James A. Goodman, Evlyn Marcia Leão de Moraes Novo, Thiago Sanna Freire Silva

Abstract: This study assessed sunglint effects, also known as the specular reflection from the water surface, in high-spatial and high-spectral resolution, airborne images acquired by the SpecTIR sensor under different view-illumination geometries over the Brazilian Ibitinga reservoir (Case II waters). These effects were corrected using the Goodman et al. (2008) and the Kutser et al. (2009) methods, and a Kutser et al. (2009) variant based on the continuum removal technique to calculate the oxygen absorption band depth. The performance of each method for reducing sunglint effects was evaluated by a quantitative analysis of pre- and post-sunglint correction reflectance values (residual reflectance images). Furthermore, the analysis was supported by inspection of the reflectance differences along transects placed over homogeneous masses of waters and over specific portions of the scenes affected and non-affected by sunglint. Results showed that the algorithm of Goodman et al. (2008) produced better results than the other two methods, as it approached zero amplitude reflectance values between homogenous water masses affected and non-affected by sunglint. The Kutser et al. (2009) method also presented good performance, except for the most contaminated sunglint portions of the scenes. When the continuum removal technique was incorporated to the Kutser et al. (2009) method, results varied with the scene and were more sensitive to atmospheric correction artifacts and instrument signal-to noise ratio characteristics.

Keywords: coral reefs; remote sensing; field spectra; scale; ecology; biodiversity; conservation hyperspectral remote sensing, specular reflection, water optically active substances, SpecTIR sensor

Figure 5. Deglinted SpecTIR hyperspectral of Ibitinga reservoir (São Paulo, Brazil) images and resultant reflectance profiles after correction by the methods of: (a) Goodman et al. (2008); (b) Kutser et al. (2009); and (c) modified Kutser et al. (2009).

Streher et al. 2015 Fig 5 Deglint

Remote Sensing Analysis in the Cloud – Introducing the HICO Image Processing System

HySpeed Computing is pleased to announce release of the HICO Image Processing System – a prototype web application for on-demand remote sensing image analysis in the cloud.

HICO IPS: Chesapeake Bay Chla

What is the HICO Image Processing System?

The HICO IPS is an interactive web-application that allows users to specify image and algorithm selections, dynamically launch analysis routines in the cloud, and then see results displayed directly in the map interface.

The system capabilities are demonstrated using imagery collected by the Hyperspectral Imager for the Coastal Ocean (HICO) located on the International Space Station, and example algorithms are included for assessing coastal water quality and other nearshore environmental conditions.

What is needed to run the HICO IPS?

No specialized software is required. You just need an internet connection and a web browser to run the application (we suggest using Google Chrome).

How is this different than online map services?

This is an application-server, not a map-server, so all the results you see are dynamically generated on-demand at your request. It’s remote sensing image analysis in the cloud.

What software was used to create the HICO IPS?

The HICO IPS is a combination of commercial and open-source software; with core image processing performed using the recently released ENVI Services Engine.

What are some of the advantages of this system?

The system can be configured for any number of different remote sensing instruments and applications, thus providing an adaptable framework for rapidly implementing new algorithms and applications, as well as making these applications and their output readily available to the global user community.

Try it out today and let us know what you think: http://hyspeedgeo.com/HICO/

 

Related posts

Calculating a land/water mask using HICO IPS

Deriving chlorophyll concentration using HICO IPS

Evaluating water optical properties using HICO IPS

Characterizing shallow coastal environments using HICO IPS