Dear Roshan,

You can utilize the Landsat satellite images (30 meter pixel resolution) to obtain LU/LC data. This link would help in getting an idea on howto proceed [https://www.e-education.psu.edu/geog160/node/1960]. In addition, you can freely download the MODIS LU/LC data (MCD12Q1) which is a yearly LU/LC data with 500m pixel resolution [https://lpdaac.usgs.gov/products/mcd12q1v006/]. There is another dateset from European Space Agency (ESA) that available from 1992 to present with pixel resolution around 300m [https://cds.climate.copernicus.eu/cdsapp#!/dataset/satellite-land-cover?tab=form]. Good Luck and Please let me know if you need anything else : )

Islam

The link of code may be helpful to you

https://code.earthengine.google.com/649d57ad1c18aa897aa671764b05e602

https://code.earthengine.google.com/808c4afbce04d2f70347f1dd40ff04a7

You can also use Unsupervised and supervised methods to generate LULC

Another GEE code can help you

https://code.earthengine.google.com/339bb75e85836312420aaccee7da2cf2

NDVI is often not a good variable for LUCL (change) mapping from a single or a few Landsat images. For a better method please see:

Article Forecasting the pattern and pace of Fagus forest expansion i...

However hyper-temporal NDVI (SPOT, MODIS) is excellent, please see:

Article A novel approach to optimize hierarchical vegetation mapping...

Dear Roshan Singh Thagunna

I think Sentinel and ML are valuable dataset and techniques to get the LULC for your study area. Good luck.