Analysis of Land Use Extraction through Morphological Analysis Using Geographic and Remotely Sensed Data

Lack of detailed land use (LU) information and inefficient data gathering methods have made modeling of urban systems difficult. This study aims to develop a hybrid remote sensing (RS)/geographic information (GI) system in order to extract residential LU information from very high resolution (VHR), remotely sensed imagery. Land cover information extracted from remote sensing and several types of geographic data from the study area (City of Fredericton, Canada) are fused into the residential LU extraction expert system to examine correlation/association rules at the building level. Morphological analysis at the building level is used through a step-wise binary logistic regression model to provide a set of multi-dimensional indicators for extracting the residential buildings. In this regard, sets of morphological properties derived from geographic vector and remotely sensed data are used in a binary regression model. LU classification from the morphological analysis results in an overall accuracy of 93.2% for extracting residential buildings. It should be noted that equipped with such a powerful LU data collection tool and detailed LU data, urban planners/modellers can more reliably and precisely predict economic interactions, activity locations, space and housing developments, business expansion, and trip patterns.