ArcUser Online

Search ArcUser

Unlocking the Power of Image Data




What is ArcUser?


Submit an Article

Current Issue

Previous issues

Jump Station

Visit these site to learn more about topics mentioned in ArcUser

Image data--It used to be a precious commodity, but now it's everywhere. New technologies have taken us from aerial photographs to satellite imagery. Now remote sensing systems can capture and process digital data within hours so it can be used almost immediately for display and analysis purposes.

As image data has proliferated, the need for desktop tools to manage images has grown. Image analysis software was previously reserved for geospatial professionals using high-performance workstations. Esri and ERDAS have developed ArcView Image Analysis, an extension to ArcView GIS that brings these capabilities to desktop users.

VESTRA Resources, a GIS services firm based in Redding, California, provides integrated solutions for land management, local government, agricultural, and transportation applications. Dean Angelides, senior GIS analyst at VESTRA, says imagery is used in nearly every project performed by the company. "I can't think of a project that we've done in the last five years that does not involve some type of imagery. For our clients to get the most from the data sets that have been put together, it's really important to have image handling capabilities beyond just displaying them as a backdrop to vector data."

Angelides says that though VESTRA and many of their clients make extensive use of IMAGINE from ERDAS and Esri's ArcInfo software, applications are primarily deployed at the desktop level using ArcView GIS. "Image Analysis is a great package of tools for people who use imagery, which is nearly everybody, to make it easy to exploit the imagery," says Angelides. "It unlocks things that were done in high-end research facilities and puts them on common people's desktops."

Multispectral Image A multispectral image enhanced using the ArcView Image Anaysis legend editor compared to the lower view that shows the same image as depicted using standard ArcView GIS display capabilities.

ArcView Image Analysis creates Reduced Resolution Data Sets (RRDS) or "pyramid layers" that make images display more rapidly. Images often exceed the screen resolution of most computer monitors and graphics cards. On-the-fly reduction of image data is computationally intensive and time consuming.

ArcView Image Analysis avoids this performance degradation by preemptively creating pyramid layers when an ArcView Image Analysis theme is added to a view. These layers represent reduced views of the original data that has been resampled reduced in resolution by a power of two. The original data is not changed in the process.

"If you deal with imagery in large data sets, even as a backdrop, the speed advantages are very, very significant," says Angelides. "It may be worthwhile to purchase this extension just for the drawing speed enhancement."

Putting Image Data to Work
Land cover classification and change detection analysis are important techniques for industries, such as timber or agriculture, that work with large geographic areas. ArcView Image Analysis provides built-in functionality to perform both types of analyses.

ArcView Image Analysis uses the Iterative Self-Organizing Data Analysis (ISODATA) technique to categorize image data based on its spectral values. In land cover classification, the known spectral qualities of various types of land features, such as open land, forests, agriculture, water bodies, and structures, are used to identify these areas and create a theme. Categorization of resulting land cover themes can be refined to provide information on the types of trees or other vegetation growing in a forest.

multispectral images in Image Analysis A multispectral false color infrared image displays vigorous vegetation in bright red tones. The lower view displays the vegetation index image as a backdrop to citrus grove vigor vegetation index values summarized using the ArcView Spatial Analyst extension.

ADAR 5500 image courtesy of Crop Image, Salinas, CA.

Change detection analysis is another important and growing use of image data. Images captured on different dates can be analyzed to determine differences in spectral response. ArcView Image Analysis can work with both continuous and thematic data. The Image Difference function uses continuous data to determine the amount of change over time as measured in brightness or as a direct value. The Thematic Change function, used on categorized or thematic data, is used to learn both the type and amount of change.

Both commands compare before and after themes and create a third theme that shows areas of change. An additional tool is available when using Thematic Change on image data. The Summarize Areas function can be used to compile information about the area in tabular form. Using these methods, foresters can track change to monitor growth, identify pest infestation, measure forest fire damage, and schedule management activities.

Visualizing the Path of Growth
Regional transportation planning is a prime example of an application that benefits from the use of imagery. Transportation planners are often tasked with determining the future demand for transportation systems including light rail, highways, and streets. Often, regional databases are difficult to develop because they require data from multiple jurisdictions.

Using imagery can help tell a story a lot faster, and at a lower cost, than trying to merge and collect data sets from several government agencies. Angelides said, "You can very quickly derive different types of land uses and see land use patterns changing through time. It gives you a sense of where transportation systems are adequate, where they may need to be improved."

Vegetation Index Analysis Dialog boxes make powerful image processing procedures, such as Vegetation Index analysis, easy to perform. The image shows the Normalized Difference Vegetation Index (NDVI) image computed from the ADAR multispectral image shown in the upper view.

Angelides further notes that imagery is a valuable decision support tool in planning new transportation systems. "Traditionally, when transportation corridors are built, it encourages growth along those routes. Growth is often controversial, so new routes must be carefully planned. Imagery can help people visualize very large areas very quickly, which helps in decision making. They can look at proposed facilities and determine where capacities ought to be increased or left alone."

Assessing Farming Practices
Of all the industries that use imagery, agriculture is beginning to rely on it the most. To stay competitive in agriculture, farmers must use new technologies to get the highest possible crop yield while paying close attention to the return per acre of cultivated land. Imagery provides a cost-effective way to track variation in crop growth patterns through time and across acreage.

In California, VESTRA Resources has developed precision farming applications and has particular expertise in winery management applications. Angelides explains, "A lot of factors influence how grapes grow. To understand how those factors interact and relate them to the grape quality takes a fair amount of information and tools to organize that information. ArcView GIS, along with Spatial Analyst, ArcView 3D Analyst, and Image Analysis, is an excellent set of tools for being able to sort different factors and zero in on those that most influence the quality at a particular vineyard."

Vegetation classification A vegetation vigor classification was created from the NDVI image and vineyard blocks using ArcView Spatial Analyst and draped over the terrain model created with 3D Analyst.

Research has shown that if the vines are under a certain amount of water stress, they will produce higher quality grapes. The challenge for vineyard managers and viticulturists is to maintain the appropriate level of water stress without damaging the vines, which would make them more susceptible to disease.

Vineyard managers can evaluate the relative vigor of the vines over the course of the growth by capturing infrared snapshots or multispectral snapshots of the vineyard at various times of the year and applying the change analysis techniques. ArcView Image Analysis calculates the Normalized Difference Vegetation Index (NDVI) and creates a new grayscale single-band theme that highlights areas of vegetation and measures the biomass or health of these areas. The Vegetative Index function takes a multispectral image and computes an NDVI based on the relationship between near infrared and red reflectance values.

Angelides says that ArcView Image Analysis is also useful for vineyard operations. "Some of the vineyards that have been capturing multispectral imagery for a year or more now have a much better understanding of their vine vigor, cover, and other factors. In fact, the imagery actually helps them schedule their harvesting operations.

What they used to do is bring in a huge crew that worked day and night, monitoring sugars and other factors as they harvested. Sometimes they would keep harvesting from an operational perspective, even if the grapes weren't ready for harvesting. With imagery, vineyards can identify indicators long before it's obvious to the naked eye. It allows the vineyard managers to break out different areas of the vineyard that ought to be harvested at different times from others."

VESTRA often uses ArcView Image Analysis in conjunction with ArcView Spatial Analyst and ArcView 3D Analyst. Angelides cites a situation where a vineyard was trying to determine the cause of variations in plant vigor within a vineyard block. Evaluating soil nutrient information at different rooting depths as well as at the surface provided a clear picture of where the problems were so that remediation could be applied.

First, ArcView Spatial Analyst was used to interpolate soil nutrient values and combinations of nutrient values, water holding capacity, and other factors for each rooting depth. Then, ArcView 3D Analyst was used to "stack" the soil analysis results for each rooting depth along with a vine vigor map created using ArcView Image Analysis. Graphically displaying this data let the vineyard manager quickly identify which factors were causing differences in vine growth. Through the knowledge gained in this analysis, the vineyard manager was also able to select rootstock for new plantings that was better suited to the soil conditions.

Valuable Source of GIS Data
With the development of the ArcView Image Analysis extension, a very valuable data source--image data--becomes available to GIS users at the desktop level. Foresters, farmers, regional planners, and others who work with large eographic areas can benefit from this powerful data source in developing applications that track change over time, categorize land cover, or monitor vegetation.

Learn more about ArcView Image Analysis.

VESTRA Resources, Inc., an Esri business partner, provides solutions in the area of agriculture, forestry, sustained long-range planning of land resources, and natural resource management. VESTRA offers a range of products and services from project-oriented GIS database construction, map production, and image analysis to design, development, and implementation of integrated GIS applications.


Contact Us | Privacy | Legal | Site Map