Two parallel and overlapping approaches to classification of remotely sensed data with the aid of spatial information are underway at the present time. The image processing approach attempts to model after the human visual system, while the second approach is primarily numerical. The technique of texture features, representing the image processing approach, and the sample classifier ECHO, representing the numerical approach are compared. The numerical approach is demonstrated to be superior in classification accuracy as well as being more efficient computationally.
Date of this Version