21 / 2021-07-13 15:35:59

Bias impact factor analysis based on satellite radiation data

remote sensing data,feature selection,bias impact factor,XGBoost algorithm

    The quality evaluation of remote sensing data plays an important role in the application and development of remote sensing technology. It is also an important hub for the development of remote sensing instruments and the application of remote sensing data. It can not only make a reasonable analysis of the results of previous work, but also provide a scientific basis for the launch of subsequent satellites. With the continuous development of satellite remote sensing technology in China, remote sensing data have been more widely used in many fields, so there are higher requirements for the quality of satellite data. Therefore, the selection of key impact factors is of great importance to the simulation of rapid radiation transmission, and is also the key to the prediction effect of the model. Based on Aqua MODIS satellite data, aiming at the existing problems of traditional feature selection methods and the advantages of XGBoost method, the importance of features of XGBoost algorithm was calculated and ranked, and then the key influencing factors of satellite radiation data deviation were screened out. Experiments show that the proposed method improves the accuracy of remote sensing data quality analysis, and the accuracy and stability of on-orbit calibration are significantly improved.