Data compression method suitable for marine meteorological visualization

Abstract

The invention discloses a data compression method suitable for marine meteorological visualization, which comprises the following steps of: step 1, reading original data, namely reading the original marine meteorological data, and extracting a time slice and meteorological element data; step 2, meteorological data preprocessing: performing integer processing on the original data; step 3, removingcorrelation of the meteorological data; and step 4, compressing code stream data and performing entropy coding on the meteorological data subjected to decorrelation. Compared with an existing data compression mode, the method is more suitable for compressing the data visually displayed by the marine meteorology, the characteristics of high real-time requirement and large data volume of traditionalmeteorological data are fully considered, the marine meteorological data can be compressed, the data repeatability is reduced, redundant labels are removed, and the data size is reduced to the maximum extent. The method specifically comprises the steps that 1, original data can be read; 2, truncating is performed according to the observation precision of the elements; and 3, the correlation of the meteorological data is removed; and 4, the marine meteorological data compression ratio is increased.

Description

technical field [0001] The invention relates to a data compression method suitable for marine meteorological visualization, which belongs to the technical field of marine meteorological prediction. Background technique [0002] Due to the relatively large time span involved, dozens of types of meteorological elements, dense time-series data sets that may have data every minute, and meteorological element data at different levels and heights, the amount of data is often It is very large, which brings great pressure to the storage and transmission of data. Taking the product data of a medium-term forecast model as an example, based on the output of only 5 basic meteorological elements with a resolution of 360*181 per day, 17 layers, a forecast every 6 hours, and a maximum time limit of 168 hours, the data volume is 1.6 billion One, stored in the form of conventional 32bit floating point, will take up 612.71MB of space, up to 223.8GB a year. Although the capacity of storage m...

AI Technical Summary

Problems solved by technology

However, the compression of marine meteorological data with a large amount of data has little effect, and there is no targeted compression for the correlation and characteristics of meteorological data in time and space. The compression rate is relatively low, and it is not suitable for the compression of marine meteorological data.

Images