Foundation cloud form identification method based on multiscale structure characteristics

Abstract

The present invention provides a foundation cloud form identification method based on multiscale structure characteristics. The method has the following concrete steps of (1) utilizing a foundation infrared cloud measurement instrument to obtain images, screening the images in the artificial contrast and artificial correction manners, selecting the typical samples to combine into a typical sample base as a training standard set, wherein the rest is used as a test sample set; (2) decomposing the training standard set and the test sample set by a layering method to obtain a multiscale image of each cloud picture; (3) extracting the multiscale structure characteristics of the cloud pictures; (4) by a support vector machine classifier and according to the final character representation of the training foundation cloud picture samples in the training standard set and the final character representation of the test foundation cloud pictures in the test sample set, classifying the test foundation cloud pictures to obtain a classification result of the test foundation cloud pictures, and optimizing the parameters of a support vector machine according to the classification result.

Description

technical field [0001] The invention relates to a ground-based cloud shape recognition method based on multi-scale structural features, which is mainly used in automatic cloud shape observation. Background technique [0002] The formation and evolution of clouds is one of the concrete manifestations of the intricate physical processes taking place in the atmosphere. The shape, distribution and changes of clouds indicate the state of atmospheric movement and can be used as a sign of weather changes. The automatic observation of clouds is a concern of the World Meteorological Organization and the China Meteorological Administration, and it has been listed as a priority research topic in the "Comprehensive Meteorological Observation Research Program" of the China Meteorological Administration. [0003] Cloud shape is an important meteorological element and plays an important role in weather forecasting, on-site support, aerospace and climate research. An experienced observer ...

AI Technical Summary

Problems solved by technology

The above methods mainly use the texture features developed in digital image processing technology to express the characteristics of cloud images, and study the distribution and combination information of pixel gray values, but do not combine the characteristics of ground-based cloud images and the experience of artificial visual inspection to study the structural information of clouds.

Judging from the current research results, the existing expression factors of ground-based cloud features are not effective enough, and cannot well represent the information of ground-based cloud images

Images