Land use and land cover (LULC) analysis is a fundamental component of environmental monitoring and land management, offering valuable insights for urban planning and sustainable development. With the advancements in machine learning, new avenues have opened for gaining insights about urbanization trends, deforestation and climate change. There has been works done on LULC but they have used aerial images directly without applying image segmentation resulting in limited insights, inaccuracies, lack of localization and inability to distinguish between complex land cover types. To solve the drawbacks, we have applied image segmentation techniques to aerial images of LULC in this study. Five types of image segmentation techniques are used: 1) Threshold based, 2) Edge-based, 3) K-means clustering based, 4) Otsu’s segmentation, 5) Unet with ResUnet. The aerial images are divided into ten types of land uses. Dataset was prepared by acquiring aerial images of LULC, the images were pre-processed to enhance its quality and suitability for segmentation, and finally then one by one each segmentation technique was applied to all the images. Through experimentation and validation the most suitable image segmentation technique for LULC has been determined. The Accuracy, Jaccard Similarity Coefficient (JSC), Dice Similarity Coefficient (DSC), Intersection Over Union (IoU), Mean Intersection Over Union (MIOU) have been used to assess the effectiveness on LULC segmentation.