Print
From a technology perspective, the simplest way to extract information from remotely sensed data is human interpretation. However, significant training and experience are needed to produce a skilled image interpreter. (Campbell, 2007) Eight elements of image interpretation employed by human image interpreters are:
- Image tone: the lightness or darkness of a region within an image.
- Image texture: the apparent roughness or smoothness of a region within an image.
- Shadow: may reveal information about the size and shape of an object which cannot be discerned from an overhead view alone.
- Pattern: the arrangement of individual objects in distinctive recurring patterns, such as buildings in an industrial complex or fruit trees in an orchard.
- Association: the occurrence of one type of object may infer the presence of another commonly associated object nearby.
- Shape: man made and natural features often have shapes so distinctive that this characteristic alone provides clear identification.
- Size: the relative size of an object related to other familiar objects gives the interpreter a sense of scale, which can aid in the recognition of objects less easily recognized.
- Site: refers to topographic position. For example, certain crops are commonly grown on hillsides or near large water bodies.
Tasks common to image interpretation are:
- Classification: assigning objects, features, or areas to classes. This occurs at three levels of confidence.
- Detection: determining the presence or absence of a feature.
- Recognition: assigning an object or feature to a general class or category.
- Identification: specifying the identity of an object with enough confidence to assign it to a very specific class.
- Enumeration: listing or counting discrete items visible on an image.
- Mensuration: measurement of objects and features in terms of distance, height, volume, or area.
- Delineation: drawing boundaries around distinct regions of the image characterized by specific tones or textures.
The results of image interpretation are most often delivered as a set of attributed points, lines, and/or polygons in any one of a variety of CAD or GIS data formats. The classification scheme or interpretation criteria must be agreed upon with the end user before the analysis begins.