The field of digital image processing has experienced continuous and significant expansion
in recent years. The usefulness of this technology is apparent in many different disciplines
covering medicine through remote sensing【遥感】. The advances and wide availability of image
processing hardware has further enhanced the usefulness of image processing.
Remote sensing is the process of collecting data about objects or landscape features without
coming into direct physical contact with them.
Digital Image Processing is not only a step in the remote sensing process, but is itself a
process that consists of several steps. It is important to remember that the ultimate goal of this
process is to extract information from an image that is not readily apparent or is not available in
its original form. The steps taken in processing an image will vary from image to image for
multiple reasons, including the format and initial condition of the image, the information of
interest (i.e., geologic formations vs. land cover), the composition of scene elements. There are
three general steps in processing a digital image; preprocessing, display and enhancement, and
PreprocessingBefore digital images can be analyzed, they usually require some degree of
preprocessing. This may involve radiometric corrections, which attempt to remove the effects of
sensor errors and/or environmental factors. A common method of determining what errors have
been introduced into an image is by modeling the scene at the time of data acquisition using
ancillary data collected.
Geometric corrections are also very common prior to any image analysis. If any types of
area, direction or distance measurements are to be made using an image, it must be rectified if
they are to be accurate. Geometric rectification is a process by which points in an image are
registered to corresponding points on a map or another image that has already been rectified. The
goal of geometric rectification is to put image elements in their proper planimetric (x and y)
Information EnhancementThere are numerous procedures that can be performed to
enhance an image. However, they can be classified into two major categories: point operations
and local operations. Point operations change the value of each individual pixel independent of
all other pixel, while local operations change the value of individual pixels in the context of the
values of neighboring pixels. Common enhancements include image reduction, image
magnification, transect extraction, contrast adjustments (linear and non-linear), band ratioing,
spatial filtering, fourier transformations, principle components analysis, and texture
Information ExtractionUnlike analog image processing, digital image processing
presently relies almost wholly on the primary elements of tone and color of image pixels.
There has been some success with expert systems and neural networks which attempt to
enable the computer to mimic the ways in which humans interpret images. Expert systems
accomplish this through the compilation of a large database of human knowledge gained from
analog image interpretation which the computer draws upon in its interpretations. Neural
networks attempt to "teach" the computer what decisions to make based upon a training data set.
Once it has "learned" how to classify the training data successfully, it is used to interpret and
classify new data sets.
Once the remotely sensed data has been processed, it must be placed into a format that can
effectively transmit the information it was intended to. This can be done in a variety of ways
including a printout of the enhanced image itself, and image map, a thematic map, a spatial
database, summary statistics and/or graphs. Because there are a variety of ways in which the
output can be displayed, a knowledge not only of remote sensing, but of such fields GIS,
cartography, and spatial statistics are a necessity. With an understanding of these areas and how
they interact one with another, it is possible to produce output that give the user the information
needed without confusion. However, without such knowledge it is more probable that output will
be poor and difficult to use properly, thus wasting the time and effort expended in processing the
remotely sensed data.
1, ancillary [æn'siləri]
2, rectify ['rektifai]
3, planimetric [,pleinə'metric]
4, transect [træn'sekt, -'zekt-, trɑ:n-]
5, mimic ['mimik]