An object-based temporal inversion approach to urban land use change analysis

Rates of population growth and urbanization currently taking place in sub-Saharan Africa are and will be the highest in the world, providing impetus for remote-sensing-based monitoring of land cover and land use change (LCLUC). The objective of this study is to assess the utility of using segmentation and classification information derived from an object-based classification of an image at a later date (t=2) to aid in the segmentation and classification of an image at an earlier date (t=1) for the same area, in the context of a post-classification comparison for LCLUC analysis within Kumasi, Ghana. This object-based temporal inversion approach was found to increase the accuracy of the t=1 classification and land use change classifications by 6.2% and 6.6%, respectively, compared to a traditional, non-constrained object-based image analysis (OBIA) approach.