A multi-frequency one-dimensional synthetic aperture microwave radiometer sst deep learning inversion method

Abstract

The invention discloses a multi-frequency one-dimensional comprehensive aperture microwave radiometer SST deep learning inversion method which comprises the following steps of constructing initial background field data, and supplying a data support for SST inversion; according to a multi-incidence-angle observation characteristic of the one-dimensional comprehensive aperture microwave radiometer,by means of a microwave radiation transmission model, respectively calculating mode brightness temperatures of an atmosphere top screen in five frequencies, adding a random error into the mode brightness temperature, and simulating the observation brightness temperature of the one-dimensional comprehensive aperture microwave radiometer; constructing a deep learning model in which an auto-encoder is coupled with a full connecting layer, training the self-encoder through the mode brightness temperature and the observation brightness temperature so that the auto-encoder realizes a data error reduction effect; using output of the auto-encoder and incidence angle data as an input of the full connecting layer, training the full connecting layer through minimizing a loss function, and obtaining asea level temperature through inversion. According to the method of the invention, a method for realizing SST inversion is realized through multi-frequency brightness temperature data, thereby settling a problem of high difficulty in sea level temperature inversion in a multi-incidence-angle condition.

Description

technical field [0001] The invention relates to the technical field of remote sensing, in particular to an SST deep learning inversion method of a multi-frequency one-dimensional synthetic aperture microwave radiometer. Background technique: [0002] Sea surface temperature (SST) plays an important role in global climate change and long-term weather processes. According to the law of radiology, any object with a temperature greater than absolute zero will radiate electromagnetic waves, and microwaves have longer wavelengths and are less affected by the atmosphere. Therefore, passive microwave remote sensing has an absolute advantage, and it can conduct uninterrupted observations around the clock . One of the representative instruments of passive microwave remote sensing is the real-aperture microwave radiometer, which can provide products of various marine environmental elements including sea surface temperature. However, because the spatial resolution of the real-aperture...

AI Technical Summary

Problems solved by technology

However, due to the difference in imaging methods, the existing real-aperture microwave radiometer sea surface temperature retrieval algorithm cannot be applied to the one-dimensional synthetic aperture microwave radiometer

Traditional real-aperture microwave radiometers generally scan and image at a fixed incident angle, while synthetic aperture microwave radiometers are staring images of scenes, and their incident angles generally change within a certain range, which is more complicated than real-aperture microwave radiometers.

Images