The very first step would be to convert the images from Digital Number (DN) formats to Top of Atmosphere (TOA) reflectance, which is then considered suitable for classification, specially if the image is of Tier 1 and Collection 1. Google Earth Engine (Gorelick et al., 2017) provides orthorectified TOA reflectance of Landsat-8 satellite images which are calculated according to Chander et al. (2009)'s methodology.

References

• Gorelick, N., Hancher, M., Dixon, M., Ilyushchenko, S., Thau, D., & Moore, R. (2017). Google Earth Engine: Planetary-scale geospatial analysis for everyone. Remote Sensing of Environment, 202, 18-27.
• Chander, G., Markham, B. L., & Helder, D. L. (2009). Summary of current radiometric calibration coefficients for Landsat MSS, TM, ETM+, and EO-1 ALI sensors. Remote sensing of environment, 113(5), 893-903.

Dear Aysar,

Please see the paper and the related toolbox (Supplementary Materials) at:

https://www.mdpi.com/2072-4292/12/2/294