Development of Knowledge-based Classification System

While the spatial resolution of remotely sensed data has improved, multispectral imagery is still not sufficient for urban classification. Problems include the difficulty in discriminating between trees and grass, the misclassification of buildings due to diverse roof compositions and shadow effects. Recently, aerial LiDAR data have been integrated with imagery to obtain better classification results. The LiDAR-derived nDSMs become an important factor for urban classification based on the experimental results. To generate better DEMs, we presents an adaptive LiDAR data-filtering algorithm that effectively filters out ground objects. In this book, we also propose a knowledge-based classification system (KBCS) which can improve overall accuracy by 12 and 7% compared to maximum likelihood and object- based classification, respectively. The implementation details of KBCS are discussed using Matlab, ERDAS IMAGINE Expert Classifier, and eCognition. All the programs presented in this book can be downloaded from author s web-site. For a urban-feature extraction researcher, this book is not only for studying theory but also for practical guidance.

Autorentext

Dr. Ming-Jer Huang: Studied marine environment and engineering at National Sun Yat-sen University in Taiwan. Adjoint assistant professor, department of Marine Science, Naval Academy (Taiwan). Research interests: image processing & classification, lidar data processing, computer vision, GIS & Virtual Reality, and numerical ocean model.