On Earth, the surface is referred to as land cover, whereas land use refers to the activities that occur on the land. Any type of terrain cover is possible, including bare soil, grassland, snow, and deciduous forests. Developing a technique for categorizing land cover employing ANN is the aim of this study. First, as a preliminary process for dimensionality reduction, principle component analysis (PCA) is carried out to the input image. After that, the preprocessed image is brought to the extraction of features phase. A neural filter is utilized in the feature extraction process. Statistical features such as minimum, maximum, as well as standard are derived from this collection of characteristics. The retrieved features are transmitted in order to train along with assessing these characteristics. Examine the image of the ground reality following. This ground truth image displays regions with fields, lakes, as well as buildings. Separating these images into sets to be trained and tested is the next stage in creating a classification model. An ANN classifier is employed in order to categorize the model. Finally, the ground truth image’s training label is used to test the attributes of the input picture. When an ANN classifier successfully classifies the retrieved picture, the result is precise forecasts of land cover regions. This investigation is being carried out using MATLAB, a simulation software.