Land use and land cover change detection by using principal component analysis and morphological operations in remote sensing applications

In remote sensing, the land use and land cover (LULC) play a vital role in analyzing Earth information for human development. These objects of change detections are helpful for finding the increased or decreased areas due to limitations of Landsat panchromatic images (gray scale) are not appearing identification of land object areas. To overcome this problem, multispectral bands are applied to individual channels by using principal component analysis (PCA) and morphological operations. Classification can then be performed for the identification of objects. This study focused on separating forests, land, and the vegetation area, including areas under crops and urban development. Regions of localized change in a multitemporal data set were enhanced using dimensional reduction and edge sharp techniques on a simple data set. This technique is feasible for extracting LULC information through remote sensing applications. The performance of the proposed method is superior to that of conventional methods, such as PCA, the normalized difference vegetation index (NDVI) method, and the band combination method. The NDVI and normalized difference water index are used to monitor the changes in the vegetation and water level, respectively, over an interval of time. After image processing, the changes in the area under forests, agriculture land, urban development, and water were determined using unsupervised change detection. The accuracy of the proposed technique was calculated using k-means clustering. The calculated accuracy indicates that the proposed method provides superior results compared with standard methods.