Land Surface Emissivity (LSE) is a significant parameter in many different land surface studies. It can be used as an index for analyzing the structure of the material. Furthermore, LSE estimation is a significant factor in the land surface temperature estimation from remotely sensed data. In this study we presented a novel operational algorithm for retrieving LSE from Landsat-8 thermal bands (i.e.: band 10 and 11) based on vegetation index (VI). The study includes three steps: I) building up simulated dataset for Landsat-8 bands II) threshold determination for VIs and correlation analysis between VIs and LSE III) derivation regression between LSE and Vis. First, the simulated dataset has been built up based on spectral library and spectral response function of Landsat-8. ASTER Spectral Library (ASL, http://speclib.jpl.nasa.gov) and Vegetation Spectral Library (VSL), which is published by system ecology laboratory at the University of Texas at EL Paso in cooperation with the colleagues in University of Alberta (http://spectrallibrary.utep.edu/SL_browseData), were used to build up simulated dataset. These Library contain directional hemispherical reflectance of the different type’s area.  Then the threshold has been determined for each VIs and correlation analysis has been done between each VIs and LSE. The correlation between the vegetation indices and emissivity values was analyzed. The vegetation indices that were tested include: the Simple Ratio, SR, the Normalized Difference Vegetation Index, NDVI, the Enhanced Vegetation Index, EVI, Transformed Vegetation Index, TVI, Soil-Adjusted Vegetation Index, SAVI, Leaf Area Index, LAI and the Modified Soil-Adjusted Vegetation Index, MSAVI. The results of this analysis show that the correlation between VIs and emissivity is acceptable therefore these indices are used for retrieving LSE. The results show that the maximum correlation occurred between NDVI and Emissivity, also the minimum occurred for MSAVI. For determining the threshold in this study we assumed that the area can be separated into three categories, including bare soil area, vegetated area and partially vegetated (mixed area). Then the statistical parameters (max, min, mean and standard deviation) for each category (bare soil, vegetation and mixed area) were calculated and based on these parameters, threshold values were determined for each category. Finally, regression relations have been derived to estimate LSE based on VIs. Support Vector Regression, SVR, and least square method were used for this regression. The RMSE of regression is different for each VIs. However, this value is less than 0.0035 for all VIs. The minimum of them occurred for NDVI and TVI also the maximum is for MSAVI. The presented method was evaluated by using an independent dataset. The result shows that the RMSE of LSE for band 10 and 11 is less than 0.007 and 0.009 respectively. The presented method is robust for estimating LSE from Landsat-8 satellite imagery and also is simple and do not need any auxiliary data. For further study, local comprehensive dataset can be built up and also the effect of atmospheric parameters or dust on regression coefficients can be analyzed.