Method, device and system for automatically adding new labeled time sequence samples of remote sensing images

Abstract

The invention relates to a method, a device and a system for automatically adding new labeled time sequence samples of remote sensing images. According to the method, firstly, every unclassified and labeled sample in an unclassified and labeled sample set is sorted into a candidate set of a corresponding category according to the dynamic time warping distance between the corresponding unclassified and labeled time sequence sample and category centers of the classified and labeled time sequence sample set; secondarily, samples which are most distinguished from the category centers of different categories in the classified and labeled sample set are selected from the candidate sets in different categories and labeled. In the iterative new sample adding process, the diversity of the sample sets can be sufficiently guaranteed, high-quality training data can be provided for remote sensing big-data monitoring and scene recognition and other application scenes, and the implementation process is completely automatic, efficient and stable.

Description

technical field [0001] The invention relates to the technical field of remote sensing image processing, in particular to active learning technology and remote sensing image classification technology. More specifically, the present invention describes an automatic sample labeling technology, which draws more samples of the corresponding category from a small number of labeled samples, so as to realize the automatic update of labeled samples. Background technique [0002] In the remote sensing industry, image classification has always been one of the most basic and important tasks. Whether it is low-to-medium resolution land cover classification or high-resolution scene classification, it is the starting point for the development of related applications. The classification based on the time series of remote sensing images can use the information in multiple images at the same time to effectively improve the classification accuracy. In today's era of big data, the sources of r...

AI Technical Summary

Problems solved by technology

A type of method closely related to this idea is called active learning in machine learning, but active learning is more concerned with selecting data, and does not solve automatic labeling, and there is no selection criterion for the time series of remote sensing images

The scheme proposed by the present invention on the basis of this idea not only solves the problem of automatic selection of time series data, but also solves the problem of automatic labeling, and forms a set of fully automatic methods for classifying and labeling samples.

[0005] The present invention proposes a method, device, system and storage device for automatically updating time series classification and labeling samples of remote sensing images, so as to solve the serious shortage of classified and labeling samples in the big data analysis of time series of remote sensing images

Images