Hot spot grid pollutant data capture method based on the study of gridding depths of features

Hot spot grid pollutant data capture method based on the study of gridding depths of features

CN
  • CN 109,213,839 A
  • Filed: 09/12/2018
  • Published: 01/15/2019
  • Est. Priority Date: 09/12/2018
  • Status: Active Application
First Claim
Patent Images

1. a kind of hot spot grid pollutant data capture method based on the study of gridding depths of features, which is characterized in that describedMethod includes:

  • Obtain multi-source Satellite Observations, terrain surface specifications data, space-time characteristic data, the atmosphere pollution number in pollution monitoring regionAccording to, meteorological data;

    It is multiple grid cells by the pollution monitoring region division, each corresponding monitoring subregion of the grid cell;

    The aerosol optical depth AOD characteristic parameter of each monitoring subregion is obtained according to the multi-source Satellite Observations;

    According to multiple terrain surface specifications parameters of each monitoring subregion of the terrain surface specifications data acquisition;

    According to multiple space-time characteristic parameters of each monitoring subregion of the space-time characteristic data acquisition;

    According to the pollutant concentration characteristic parameter of each monitoring subregion of the atmosphere pollution data acquisition;

    Multiple Meteorological Characteristics parameters of each monitoring subregion are obtained according to the meteorological data;

    It is special according to the AOD characteristic parameter, terrain surface specifications parameter, space-time characteristic parameter, pollutant concentration characteristic parameter and meteorologyLevy the multidimensional characteristic vectors of each monitoring subregion of parametric configuration;

    All multidimensional characteristic vectors are generated into multidimensional characteristic sample set;

    The multidimensional characteristic sample set is trained using deep learning model, obtains relational model;

    The multi-source Satellite Observations of target area are obtained, and extract gridding multidimensional characteristic information;

    According to the gridding multidimensional characteristic information and the relational model, the gridding pollutant for obtaining the target area is denseDegree evidence.

View all claims
    ×
    ×

    Thank you for your feedback

    ×
    ×