A novel method for Sentinel-2 satellite images radiometric calibration

Abstract

Background and objective: Absolute and Relative Models of Radiometric Corrections Due to the relative information provided from different parts of the globe, there are always errors when correcting images, which has led many researchers to develop local models or design tools to obtain information about the situation of Atmospheric of the area under study when the satellite passes. The present study aims to present a new method for radiometric correction of satellite images.Material and methods: To apply the proposed method for correcting the calibration of satellite images, a sun-photometer was first designed and developed that can obtain atmospheric data and attenuate sunlight in the range of 400 to 950 nm in a hyperspectral mode. Sentinel-2 satellite images related to the city of Tehran were selected as a sample and in 2 days with clean and high pollution, data were obtained and the coefficients obtained were applied through a sun-photometer.Results and conclusions: Based on made assessments, the importance of radiometric corrections based on local information for satellite images become more apparent. Evaluation of the research results shows the high capability of the designed sun-photometer in accurately assessing the attenuation of sunlight in different spectral. The amount of sunlight attenuation in the green band is more than the blue and red bands, which indicates the type of urban pollution in the region, which had the greatest impact on the green band and the least impact on the red band.