The present research was designed to focus on the utility of Landsat 8-OLI multispectral data for identifying and classifying benthic habitats mapping of the Red Sea after applying atmospheric and water-column corrections at Hurghada city. Atmospheric and water column corrections were applied to the imagery, making it an effective method for mapping benthic habitats. Water column correction was achieved by deriving absorption and backscattering coefficients for each band of the image of clear water pixels. An unsupervised classification (ISODATA) algorithm was applied to generating 22 class habitats. The supervised classification was performed using machine-learning algorithm a maximum likehood and reference points to produce 7 classes of benthic habitat as the following, coral reefs (dense and patch), sea weeds (macro-algae), sea grass (dense and patch), deep water (more than 20 m), shallow water (less than 20 m), sandy bottom (mainly consist of calcium carbonates and silicates) and rocky bottom. Sea weeds (Macroalgae) and deep water areas showed the highest producer's and user's accuracies, when compared to dense seagrass, mixed: seagrass/sand, and mixed: coral/sand areas. Based on 1050 reference points overall accuracy of the benthic habitat assessment is 66.7 percent, with an overall Kappa coefficient value of 0.611.