Semantic segmentation, which has numerous applications such as remote sensing, augmented or virtual reality systems, autonomous driving, and many more, is one of the most exciting fields of machine learning and deep learning. Land cover segmentation, which divides the land surface into categories like vegetation, water, bare soil, and other elements, is one of its most well-liked uses. Many deep learning-based systems have been created to precisely extract this data. Therefore, the purpose of this paper is to give a thorough comparison of various deep learning models devoted to semantic segmentation. Two different data sets were used in the experiments for this. The outcomes clearly demonstrated that Segmenter is the most effective model in terms of overall performance, while the STDC-Seg model stands out for its efficiency in terms of execution time. Thus, this study helps to clarify choices for semantic segmentation in the particular context of land cover mapping.