285.

[Cover] [Contents] [Index]

Page 55

the type and the number of information classes that are represented in the study area to allow him or her to collect training samples of pixels from the image that are representative of the information classes. In contrast, unsupervised pattern recognition methods are less dependent on user interaction. Normally, unsupervised classifiers ‘learn’ the characteristics of each class (and, possibly, even the number of classes) directly from the input data. For instance, if the criterion used to label an object is the minimum distance between the object and the class means, this distance being measured in feature space, the unsupervised procedure will estimate the mean for each class and will refine this mean estimate iteratively (most unsupervised classifiers are iterative in operation). At each iteration, the previous set of estimates of the class means is refined until the process converges, usually when the means remain in the same place in feature space over successive iterations. The results output by unsupervised methods are called clusters or, sometimes, data classes. The pattern recognition process is complete when each cluster is identified, that is, linked to a specific information class by the user.

Although the unsupervised approach appears to be more elegant and automatic than the supervised procedures, the accuracy of unsupervised methods is generally lower than that achieved by supervised methods. In complex classification experiments, information classes often overlap. In the spectral domain, this implies that the reflectance, emittence, or backscatter characteristics of different classes may be similar. In the spatial domain, the implication is that any one object (a pixel or a field, for example) may contain areas representative of more than one information class. This is the mixed pixel problem. Spectral and spatial overlap of classes is the main barrier to the achievement of high classification accuracy. Even so, some interesting unsupervised algorithms are worthy of investigation as they may reveal useful information concerning the structure of the data set. Such methods can be thought of as exploratory data analyses or even data mining. A further problem with pixel-based classifiers is that radiance (carrying information) that apparently reaches the sensor from a given pixel actually includes contributions from neighbouring pixels, due to atmospheric effects and the properties of the instrument optics (Chapter 1). Townshend et al. (2000) show that, by considering this latter effect, improvements in accuracy can be achieved. They note that only where pixel size is small relative to the area of land cover units will this effect be unimportant.

For more than a decade, pattern recognition methods applied to remotely sensed imagery have mainly been based on conventional statistical techniques, such as the maximum likelihood or minimum distance procedures, using a pixel-based approach. Although these traditional approaches can perform well, their general ability for resolving inter-class confusion is limited. As a result, in recent years, and following advances in computer tech

[Cover] [Contents] [Index]