SAR image recognition method based on multi-scale fuzzy measure and semi-supervised learning

Abstract

The invention discloses a SAR image recognition method based on multi-scale fuzzy measurement and semi-supervised learning, which solves the problem of low accuracy of SAR image recognition in the prior art. The implementation steps are: establish an image library by segmenting the original SAR image, and select a single target image block from it; extract the feature vector of the image block in the library; divide the selected image blocks into several categories, and use the corresponding feature vector as training Sample, train a semi-supervised classifier, use this classifier to classify the image library; use the trained classifier to obtain the category of the query image block input by the user; calculate the category set of the image block according to the confusion matrix, and calculate the query image block The multi-scale area fuzzy similarity with the image blocks belonging to the set in the image library, and return the number of image blocks required by the user in order from large to small. The invention can correct classification errors, has high information identification accuracy, and can be used to simultaneously interpret multiple SAR images.

Description

technical field [0001] The invention belongs to the technical field of image processing, and relates to a method for identifying SAR image information, which can be applied to simultaneous interpretation of multiple SAR images. Background technique [0002] Because SAR images have all-day and all-weather detection capabilities, especially the characteristics that optical images are completely independent of weather factors, the application fields of SAR images are gradually expanding, including agriculture, geographic surveillance, navigation, military, etc. SAR image fusion, segmentation, denoising, and change detection are all research hotspots, and SAR image recognition is an important basis for these research fields. The traditional recognition technology is mainly aimed at the problem of recognition accuracy, and most of them are applied to the small-scale area recognition problem of a single SAR image, such as the SAR segmentation method of spectral clustering integrat...

AI Technical Summary

Problems solved by technology

The above two retrieval methods have been successfully applied to massive natural image retrieval problems, but due to technical limitations and the characteristics of SAR images, the effect of directly applying them to SAR image recognition is not ideal

In 2009, a SAR image retrieval system combined with Gaussian mixture model classification was proposed, that is, the GMM retrieval system, see Hou, B., Tang, X., Jiao, L., & Wang, S. (2009, October). SAR image retrieval based on Gaussian MixtureModel classification.In Synthetic Aperture Radar,2009.APSAR 2009.2nd Asian-Pacific Conference on(pp.796-799).IEEE, this method is oriented to SAR images, effectively using texture features in the retrieval process, but Due to the use of supervised classification methods, its generalization ability in real problems is low, and because the similarity matching technology of this method does not consider the characteristics of SAR images, the retrieval effect is not ideal

Although the excellent experimental results are given in this article, these results rely on overlapping cutting original SAR images to build a gallery. The image blocks obtained by this strategy have a high degree of clustering characteristics, that is, the distance between samples in the same class is small, The distance between samples of different classes is very large. Such data distribution is often very different from the data distribution in practical applications. The experimental results cannot fully verify the effectiveness of the method.

Images