Detecting and managing change in spatial data-land use and infrastructure change analysis and detection

Abstract

In an effort to improve the accuracy, currency, and maintainability of spatial databases, technologies are needed to provide efficient, cost-effective land use change detection and feature extraction capabilities. A hybrid change detection methodology was explored as an initial process in a systematic multiresolution approach that separates areas of change from areas of no change. Change detection methods employed in the exploration of a hybrid change detection methodology were univariate image differencing, image ratioing, tasseled cap analysis, vegetation indexing, change vector analysis and post-classification thematic change detection. Textural analysis was then employed as a method for further refinement of change detection outputs to eliminate unwanted errors of commission. The results of the proposed hybrid change detection method show promise, and continued research is proposed to improve results