Report a bug, broken link, or incorrect content

Image Interpretation

Quick Look

Image interpretation is the most basic form of remote sensing analysis, consisting of manual identification of features in a remote sensing image through visual interpretation.

(Ratings are given on a 1-5 scale. Click on any rating for an explanation)

Method Overview

Image interpretation is the process of examining an aerial photo or digital remote sensing image and manually identifying the features in that image. This method can be highly reliable and a wide variety of features can be identified, such as riparian vegetation type and condition, and anthropogenic features such as roads and mineral extraction activity. However, the process is time consuming and requires a skilled analyst who has a ground-level familiarity with the study area.

Image interpretation is based on elements that are inherent in imagery. These image characteristics (also called image attributes) are comprised of seven elements that we use to derive information about objects in an image. These image characteristics are: size, shape, tone/color, texture, shadow, association, and pattern.

Figure 1: The shape of the tree crowns in the lower-left image is indicative of the type (deciduous/conifer) or even species of tree. The image key graphics (upper right) indicate the importance of crown shape on large scale photos. The shape of the area (rectangle) in the upper left photo is indicative of the land use (harvesting).

Figure 2: Size is usually evaluated by looking at objects that the interpreter may be familiar with and comparing their relative size to less familiar objects. For example: an interpreter might look at the discrete vegetation dots in the red rectangle and interpret those as trees. However, by comparing the relative size of the vehicles and the one-lane road with the vegetation dots, it becomes obvious that the trees are actually shrubs.

Figure 3: Tone (in the B&W image) allows for easy distinctions between roads, forests, harvest areas, water, and other elements. Color (upper right image) allows for easy distinction between coniferous trees, deciduous trees (in yellow fall colors), senesced grasses, and road surface types.

Figure 4: Texture can offer the interpreter clues about the density, age, and type of vegetation present. Note: texture interpretation becomes increasingly important as the scale of imagery becomes smaller.

Figure 5: Shadows cast by objects in the image can give the interpreter information about the shape and size of certain features in the image.

Figure 6: By observing objects in the image and the features surrounding them, the interpreter can make inferences about what the objects really are. In this case, the presence of a dam suggests that this waterbody is a reservoir. Without that association, the water body could not have been further classified as a reservoir.

Figure 7: Pattern (in remote sensing) is a recognizable repetition of particular shapes. In this example, the checkerboard pattern of rectangular shapes suggests that this once forested area has been clear-cut along regular boundaries.

When two sequential sets of photo coverage are available, we can also consider these additional image characteristics: 1) discernible differences in any of the above characteristics between photo sets acquired at different times; 2) the presence or absence of features in successive photos; and 3) change of location, position, or extent of features.

The relative importance of each of the seven image characteristics is not constant. Their relative importance can depend significantly on the scale of the imagery and the properties of the feature of interest. For example, the shape of a tree crown can be a very important indicator of tree species on large scale (high resolution) imagery but, on small scale imagery, individual crowns may not be easily distinguishable. In this case, stand texture may become a much more important image characteristic.

Image interpretation is an important supporting component of any other remote sensing applications. For example, detailed medium-resolution (e.g. Landsat TM) satellite mapping projects frequently incorporate image interpretation of aerial photography. Typically, photography plays the following roles in a satellite-mapping project:

  1. High-resolution photos provide an efficient means of providing training site information for image classification.
  2. During intermediate assessment phases, they are the good tool for verification of a classification.
  3. Following field verification, photos are invaluable for reclassification and editing.
  4. During the final accuracy assessment of image classification, aerial photography provides an efficient means, in some cases the only means, of determining classification accuracy.

Thus, even when a project focuses on using relatively high- or medium- resolution satellite imagery, skills in using photography (which includes having ground-based knowledge) remain essential.

Digitizing Tools

A fundamental part of image interpretation projects is the process of digitizing (delineating the outlines of) the features identified on the image. Most digitizing is performed using “on-screen” or “heads-up” displays. In heads up digitizing, the image is displayed on the computer screen using a GIS software package, such as ESRI’s ArcGIS, and the analyst draws polygons with a mouse delineating features of interest. The term “heads-up” is used because the analyst’s attention is focused on the screen displaying the image, while the drawing is performed with the mouse.

Stereo Displays

A variation on heads-up display is the use of stereo displays that create the illusion of a three dimensional perspective of the scene. Stereo displays allow analysts to incorporate topographic variation into the analysis, greatly enhancing their ability to distinguish between vegetation types, and to calculate information such as tree height. Stereo displays utilize stereo image pairs, consisting of two images of the same location taken from slightly different angles. These two images mimic the different perspective of human stereoscopic vision; however, the separation between images is much greater than the separation of human eyes. Using a 3-D display, each image is displayed separately to each eye. The analyst’s brain combines the two images, creating the illusion of three dimensions. Because the separation between the images is greater, the vertical perspective of the scene is exaggerated. The 3-D display utilizes a special monitor, 3-D glasses, or both.

Interactive Displays

Advancements in screen technology have led to the development of interactive displays consisting of a touch screen that displays the digital image and a stylus that is used to “draw” directly onto the screen. This method is reminiscent of the pre-digital era when an analyst would draw directly onto hard copy photographs, and can be more efficient than heads-up digitizing.

RSAC performed an investigation comparing the use of an interactive display to traditional heads-up, mouse digitizing. This comparison found that digitizing with the interactive pen was about twice as fast as traditional mouse digitizing (McClarin and others, 2011). The interactive pen required fewer clicks to complete procedures and the pen grip was more ergonomic than the mouse.

Other Classification Methods

Data Inputs

Image interpretation can be performed on digital images, as well as hard copy photographs. Image interpretation is usually performed on high-resolution/large-scale aerial imagery.

Method Products

Image interpretation produces datasets delineating features identified in the imagery. These datasets can be the primary mapping products or can be used an ancillary data for further analysis, such as training data for supervised classification.

Riparian Example

Evans, D.; Vanderzanden, D.; Lachowski, H. 2002. Stream Geomorphic Classification, Riparian Area Delineation, and Riparian Vegetation Cover Mapping on the Upper Middle Fork of the John Day River, Oregon. Rep. No. RSAC-31-RPT2. Salt Lake City, UT: U.S. Department of Agriculture Forest Service, Remote Sensing Applications Center. 21 p.

This study mapped stream geomorphology and riparian vegetation on the Upper Middle Fork of the John Day River in Oregon. The study combined image interpretation and a smart buffering technique based on Rosgen level and Strahler stream order to map riparian areas and vegetation.

Technical References

McClarin, S.; Hamilton, R.; Fisk, H.; Lewis, B. 2011. Assembling, Updating, Organizing, and Packaging Geospatial Data for Prefire Planning. RSAC-10020-RPT1. Salt Lake City, UT: U.S. Department of Agriculture, Forest Service, Remote Sensing Applications Center. 15p.


Image interpretation requires a skilled analyst who is familiar with ground conditions. This method is time consuming when compared to automated methods, resulting in high costs per acre.

Software/Hardware Requirements

  • GIS software such as ESRI’s ArcDesktop which is capable of displaying images during digitizing
  • Stereo display if using stereo image pairs
  • Interactive displays can reduce production times

Additional Information

Interactive Displays:

Web Search Results



You must have an account and be logged in to post or reply to the discussion topics below. Click here to login or register for the site.


Home | Toolbox Wiki | Rangeland Methods Guide | Framework | Training | About | Who We Are | Visualization | Contact

A joint project of The Nature Conservancy and the USDA Agricultural Research Service

Terms of Use | Privacy Policy