Object-Based Image Analysis
In this section, we will briefly discuss various image analysis techniques commonly used in the remote sensing industry. I will cover various techniques, including unsupervised and supervised classification, rule-based classification approaches, neural net and fuzzy logic classification and, finally, classification and regression trees (CART) and decision trees. Knowledge of these techniques is useful in developing rule bases to extract features of interest from the image. Although these techniques have been employed for pixel-based analysis, I will demonstrate that these techniques can be easily extended to objects.