Improved AdaBoost algorithm-based foundation cloud picture identification method

Abstract

The invention discloses an improved AdaBoost algorithm-based foundation cloud picture identification method. A clustering algorithm is used for separating a cloud target from a background, only the characteristics of the cloud target are extracted, and characteristic values for cloud identification are calculated, so that the identification accuracy is improved; and an AdaBoost integration algorithm is used for integrating a plurality of individual classifiers trained by a support vector machine (SVM) learning algorithm, and parameters in the SVM learning algorithm are rationally regulated in a data training process to diversify the trained classifiers, so that the cloud picture identification accuracy is improved, and the generalization performance is greatly improved. Due to the adoption of the integration algorithm, requirements on the design of the individual classifiers are low, and difficulty in the design of the individual classifiers is effectively reduced; the method is simple in structure; and the method can be implemented by the conventional picture acquisition equipment and an ordinary computer, so that the practicability and the applicability are improved.

Description

technical field [0001] The invention discloses a ground-based cloud image recognition method based on an improved AdaBoost algorithm, and relates to the application of digital image processing technology in the field of meteorological observation. Background technique [0002] Clouds play an important role in atmospheric radiative transfer, and the shape, distribution, quantity and changes of clouds indicate the state of atmospheric motion. Different clouds have different radiation characteristics and distributions, so they are of great significance to service industries such as weather forecasting and flight support. At present, the general meteorological elements have basically achieved automatic observation, but the observation of ground-based cloud images cannot be fully automated, and still relies on manual observation. Due to the relatively small range of ground-based cloud observation, the contained texture information is relatively rich, and it has strong practical ...

AI Technical Summary

Problems solved by technology

Among them, the Bayesian rule needs to know the prior probability and other factors of each category, but in actual situations, the prior probability and other conditions are often difficult to obtain; the neural network rule has a slow training speed, and the classification accuracy is low when there are many sample categories and other shortcomings; although the support vector machine has become a model for small-sample learning problems, it has certain shortcomings, such as it is difficult to implement large-scale training samples, and it is difficult to solve multi-classification problems.

Images