The paper aims to assess agricultural land infringement using satellite images by applying three methods of classification: supervised (maximum likelihood), unsupervised and normalized difference vegetation index. To determine which sets of remote sensing satellite images were the best, they were compared. During the monitoring periods (2010–2011 and 2020–2021), Beban village is used as a study area. Landsat 8, Sentinel 2, Aster, and Modes satellite images are used to generate the remote sensing data. This period has been selected to classify images in order to assess land cover changes and the infringement of agricultural lands within Beban village and Kom Hamada center. The proposed methods employ the multi-spectral remote sensing data technique for land cover classification, with the selection of a satellite image dependent on the comparisons between the data quality of each satellite image downloaded for the study area. For land cover classification, some band combinations of the remotely sensed data are exploited, and the spatial distributions such as urban areas, agricultural land, and water resources are interpreted. The results give two important points: the Landsat 8 OLI/TIRS sensor is the best when compared with the other satellite, and the second point for the percentage of agricultural land in the study area in 2020, 2015, and 2010 was estimated to be 77.76%, 78.88%, and 84.04%, respectively. That is, agricultural land infringement accounted for 6.28% of Beban Village's total area.
Read full abstract7-days of FREE Audio papers, translation & more with Prime
7-days of FREE Prime access