Night Light Data Reconstruction via Inverse Hyperbolic Sine Transform
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods struggle to establish continuous and consistent multi-source long-time-series night light data sets from DMSP/OLS and NPP/VIIRS due to data faults, limiting their applicability for research purposes.
Innovation Solution
A method and system for reconstructing multi-source long-time-series night light data by acquiring and preprocessing DMSP/OLS and NPP/VIIRS data sets, performing mutual and continuity corrections, applying an inverse hyperbolic sine transform for data fitting, and integrating the reconstructed data sets to achieve high accuracy and consistency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If existing integration methods (power function/Gaussian filter or logical function model) are used to combine DMSP/OLS and NPP/VIIRS data, then data consistency can be improved in specific regions, but the method lacks universal applicability and reliability across different regions
Solution Approach 1:
The patent applies parameter changes by transforming the NPP/VIIRS data through an inverse hyperbolic sine transform to adjust its distribution characteristics. This transformation modifies the parameters of the data (making it more normally distributed and reducing skewness), which enables the construction of a universal fitting model that can accurately represent both DMSP/OLS and NPP/VIIRS data across different regions, thereby resolving the contradiction between data consistency and universal applicability
Solution Approach 2:
The patent achieves universality by constructing a fitting model that can simultaneously fit both DMSP/OLS and NPP/VIIRS data using the inverse hyperbolic sine transform. This single model structure serves multiple purposes: it can process different data sources, different time periods, and different spatial regions, making it universally applicable across all scenarios rather than region-specific
2Quantity of substance
If multi-source night light data from DMSP/OLS and NPP/VIIRS are integrated without systematic correction, then data coverage is expanded, but data faults and inconsistencies increase
Solution Approach 1:
The patent applies preliminary action by performing systematic preprocessing on the raw data before integration. This includes correcting sensor-specific biases, adjusting for temporal variations, and transforming the data distribution through inverse hyperbolic sine transform. These preliminary corrections are performed on each data source independently before merging, ensuring that data faults and inconsistencies are addressed in advance, thereby maintaining high data quality while expanding data coverage
Solution Approach 2:
The patent uses the inverse hyperbolic sine transform as an intermediary that mediates between the DMSP/OLS and NPP/VIIRS data. This transformation acts as a common language or bridge that allows both data sources to be consistently represented and compared, enabling reliable integration while preserving the advantages of both sources and eliminating their respective faults
3Productivity
If simple data merging is used to combine night light data sets, then processing time is reduced, but pixel-by-pixel fitting precision decreases
Solution Approach 1:
The patent applies parameter changes through the inverse hyperbolic sine transform, which fundamentally alters the parameter space of the night light data. This transformation recodes the data values to create a new parameter space where the fitting relationship between different data sources becomes linear and more straightforward, enabling high pixel-by-pixel fitting precision while maintaining computational efficiency through simpler mathematical operations
Data Source
AI summary
Provided is a method of reconstructing multi-source long-time-series night light data and a system thereof, relating to the field of reconstructing ecological remote sensing data. Based on night light images of a first-generation satellite DMSP/OLS (Defense Meteorological Satellite Program/Operational Line-scan System) and a second-generation satellite NPP/VIIRS (National Polar-orbiting Partnership/Visible Infrared Imaging Radiometer Suite), a method of reconstructing a set of long-time-series night light data products is developed. Since satellite images that two generations of satellites have at the same time in a certain year use an inverse hyperbolic sine transform to fit NPP/VIIRS in 2013 into a data form of DMSP/OLS, and obtain an optimal fitting equation, thus producing a set of long-time-series data products from 1992 to 2021. The method can solve a fault problem between two generations of light data of DMSP/OLS and NPP/VIRS, and improve the accuracy of data reconstruction.