Application Tips for ENVI 5 – Utilizing the new portal view for visualizing data layers

This is part of a series on tips for getting the most out of your geospatial applications. Check back regularly or follow HySpeed Computing to see the latest examples and demonstrations.

Objective: Demonstrate the use of ENVI’s new portal view for visualizing data in multiple layers.

Scenario: This tip utilizes a Landsat ETM+ scene from Kaneohe Bay, Oahu, Hawaii downloaded from USGS EarthExplorer. The example calculates the normalized difference water index (NDWI) proposed by McFeeters (1996, IJRS 17:1425-1432), displays the index results using the Raster Color Slice tool, and then utilizes the portal view to visually validate the capacity of NDWI to differentiate land from water.

ENVI Portal

The Tip: Below are steps to open the image in ENVI, calculate the NDWI index, display output using Raster Color Slice, and use the portal to visualize the NDWI output as a ‘sub-view’ within a standard RGB view:

  • Open the Landsat scene in ENVI. After un-compressing the download folder from USGS, simply use a standard method for opening files (e.g., File > Open…) to open the <*_MTL.txt> metadata file. This loads all of the Landsat bands into the Data Manager, and opens an RGB layer in the Layer Manager.
  • Calculate NDWI. Use Band Math (Toolbox > Band Ratio > Band Math) to implement the following equation (float(b2)-float(b4))/(float(b2)+float(b4)), where b2 is Band-2 (560 nm), b4 is Band-4 (835 nm), and the float() operation is used to transform integers to floating point values and avoid byte overflow.

NDWI

  • Examine the NDWI output. Using the Cursor Value tool (Display > Cursor Value…) to explore the resulting grayscale NDWI image, it becomes apparent that there is a threshold near zero where values above the threshold are water and those below are land and cloud. While this data is alone sufficient for analysis, let’s use something more colorful for visualizing the output.
  • Display NDWI output using raster color slices. Start the Raster Color Slice tool (right click the NDWI layer name in the Layer Manager and select Raster Color Slices…), select the NDWI data in the Select Input File dialog, and then accept the default settings in the Edit Raster Color Slices dialog.

Raster Color Slice

  • Examine the Raster Color Slice layer. The color slices reveal a visually clear distinction between water and land/cloud, where in this example warm colors (reds and yellows) indicate water and cool colors (blues and green) indicate land/cloud.
  • Display output using a portal. First, make sure to put the layers in the appropriate order. Drag the RGB layer to the top of the display in the Layer Manager, and make sure the color slice layer is second. Then start the Portal tool (select the Portal icon on the toolbar, or Display > Portal). This will open a new smaller ‘sub-view’ that reveals the color slice NDWI layer within the larger view of the RGB layer. Alternatively, you could also use the portal to similarly view the grayscale NDWI layer. To do so, hover over the top of the portal to make the View Portal toolbar visible, right click on the toolbar, and select Load New Layer.

ENVI Portal

  • Explore data layers with the portal. The portal itself is interactive, which means it can be easily moved and resized to examine different portions of the image. In this example the portal can be moved around the image (left click within the portal and drag using your mouse) to explore how well the NDWI index works in different areas.
  • Other visualization options. In addition to the default portal display, the portal can also be animated to alternate between the different layers. The three options are Blend (transitions layer transparency), Flicker (toggles between layers), and Swipe (moves a vertical divider that separates the layers). These animations can be started as follows: hover over the top of the portal to make the View Portal toolbar visible, right click the toolbar, and select the desired animation option. The speed of animation can then be changed using the faster and slower buttons, and the animation can be paused or restarted using the pause and play buttons, located on the portal toolbar.
Advertisements

Highlights from VISualize 2013 – Connecting the remote sensing community

VISualize 2013 is an annual conference hosted by Exelis Visual Information Solutions, and co-sponsored by HySpeed Computing, that brings together thought leaders in the geosciences to discuss the latest trends in remote sensing. The focus this year was on “Connecting the Community to Discuss Global Change and Environmental Monitoring.”

With more than 20 presentations and ample discussion throughout, it was an insightful and very informative conference. Some highlights from VISualize 2013 include:

  • Jim Irons (NASA Goddard Space Flight Center) presented a summary of the Landsat 8 mission, including details on data distribution, sensor specifications, measurement capabilities, and new band designations. He also noted that responsibility for the instrument was officially shifted from NASA to USGS on 30 May 2013, signifying completion of all on-orbit checkouts and the initiation of public data dissemination. Currently more than 20,000 images are already available for users to download.
  • Mark Braza (U.S. Government Accountability Office) described the use of propensity score analysis to estimate the effectiveness of land conservation programs. The approach utilizes statistical analysis to identify control groups from amongst land areas associated with, but not included in, established conservation projects, and then leverages these control areas as a means to assess the relative impact of land conservation efforts.
  • Nasser Olwero and Charles Huang (World Wildlife Fund) summarized objectives of the Global Impact Award that WWF recently received from Google. In this project WWF will be using state-of-the-art technology, specifically animal tracking tags, analytical software that optimizes ranger patrolling, and airborne remote sensing, to reduce the impact of animal poaching and protect valuable species like elephants, rhinos and tigers.
  • Matthew Ramspott (Frostburg State University) presented findings from a study using Landsat data to assess wetland change along the Louisiana coast. A key aspect of the analysis was the methodology used to automatically delineate the land/water interface. Results demonstrate the value of using remote sensing to monitor long-term change in coastal wetlands and assess impact from storm damage, flood management decisions and rising sea levels.
  • Robert Rose (Wildlife Conservation Society) outlined the top 10 conservation challenges that can be addressed using remote sensing. The list of challenges result from a NASA funded workshop in early 2013, and are defined according to 10 general themes: species distribution and abundance; species movement and life stages; ecosystem properties and processes; climate change; fast response; protected areas; ecosystem services; conservation effectiveness; land cover change and agricultural expansion; and degradation and disturbance regime. In each theme the objective is to focus on achievable conservation outcomes with clear pathways for putting technology into practice.
  • Robert Rose also spoke about the revitalization of the Conservation Remote Sensing Working Group (CRSWG), which aims to encourage discussion around four main topic areas: research and collaboration; capacity development; communications; and best practices. To join the conversation, just look for CRSWG on Google Groups and contact the group admin to get involved.

Interested in more information on these and other speakers? Exelis VIS will soon be posting copies of all the VISualize 2013 presentations to their website. We’re looking forward to seeing everyone again next year at VISualize 2014.

World Wildlife FundAs thanks to WWF for opening its doors to VISualize, Exelis VIS and HySpeed Computing proudly contributed donations to WWF on behalf of the speakers. Pictured from left to right: Matt Hallas (Exelis VIS), Nasser Olwero (WWF), Charles Huang (WWF) and James Goodman (HySpeed Computing).

VISualize 2013 – Global change and environmental monitoring

HySpeed Computing is honored to be co-sponsoring VISualize 2013, the annual conference hosted by Exelis Visual Information Solutions that brings together thought leaders in the geosciences to discuss the latest trends in remote sensing. The focus of this year’s conference is “Connecting the Community to Discuss Global Change and Environmental Monitoring.” The event is being held at the World Wildlife Fund conference center in Washington, DC from 11-13 June 2013.

HySpeed Computing president Dr. James Goodman will be attending VISualize and presenting a talk on the “Power of Community Data Sharing.” The presentation will explore the status of data sharing in the remote sensing community and discuss how you can benefit and get involved.

The conference’s keynote address will be presented by Dr. Jim Irons, Associate Deputy Director for Atmospheres in the Earth Sciences Division at NASA Goddard Space Flight Center (GSFC) and Project Scientist for the NASA Landsat Data Continuity Mission (LDCM). Dr. Irons will be discussing some of the history of the LDCM mission, and what we should expect now that Landsat 8 is operational and delivering data to the science community. We are looking forward to hearing what he has to say, as well as engaging with all the other speakers and attendees. Hope to see you there.

Here are some other upcoming remote sensing conferences later this year that may also be of interest:

WHISPERS: Workshop on Hyperspectral Image and Signal Processing: Evolution in Remote Sensing, 25-28 June 2013, Gainesville, FL, USA, http://www.ieee-whispers.com/

ESRI: International User Conference, 8-12 July 2013, San Diego, CA, USA, http://www.esri.com/events/user-conference

IGARSS: International Geoscience and Remote Sensing Symposium, 21-26 July 2013, Melbourne, Australia, http://www.igarss2013.org/

SPIE: Remote Sensing, 23-26 September 2013, Dresden, Germany, http://spie.org/x6262.xml