Performing data normalization on satellite imagery (Landsat , ASTER, Sentinel) is necessary for correcting satellite image observations to a uniform nadir view (primarily for visual consistency, vegetation monitoring, or detection of subtle surface changes), and terrestrial monitoring applications. The process requires the conversion of the top of atmosphere (TOA) reflectance and then atmospherically corrected to surface reflectance using the Landsat Ecosystem Disturbance Adaptive Processing System (LEDAPS) codes.

For further readings, see attached related articles

Alya Nasir you may want to look at the following discussion. https://www.researchgate.net/post/What_is_normalization_of_Landsat_image_bands_How_to_do_that

Alya Nasir The goal of image normalization is to reduce the radiometric influence of non-surface factors, so that differences in Bright- ness Values (BVS) between satellite images from different dates will reflect actual changes on the surface of the Earth.

Therefore, if you are trying to extract the shoreline data from different epochs of satellite imageries, you will need it.

I recommend you to use landsat 8 handbook instead of videos. There is a very simple and straight forward method in which you can calibrate and atmospherically correct your data. Another solution is to first use band coefficients to change your DN values to top of atmosphere radiance and then use online atmospheric correction websites such as https://atmcorr.gsfc.nasa.gov/ to change them to bottom of atmosphere reflectance. Once you got the reflectances, you can use spectral reflectance to detect shorelines