PCA-based land-use change detection and analysis using multitemporal and multisensor satellite data

摘要: Remote-sensing change detection based on multitemporal, multispectral, and multisensor imagery has been developed over several decades provided timely comprehensive information for planning decision-making. In practice, however, it is still difficult to select a suitable change-detection method, especially in urban areas, because of the impacts complex factors. This paper presents new method using multitemporal data (SPOT-5 Landsat data) detect land-use changes an environment principal-component analysis (PCA) hybrid classification methods. After geometric correction radiometric normalization, PCA was used enhance from stacked data. Then, classifier combining unsupervised supervised performed identify quantify changes. Finally, stratified random user-defined plots sampling methods were synthetically obtain total 966 reference points accuracy assessment. Although errors confusion exist, this shows satisfying results with overall be 89.54% 0.88 kappa coefficient. When compared post-classification PCA-based also showed better terms overall, producer's, user's index. The suggested that significant have occurred Hangzhou City 2000 2003, which may related rapid economy development expansion. It further indicated most cropland areas due encroachment.