SAR image automatic classification system and method

Abstract

The present invention provides an SAR (Synthetic Aperture Radar) image automatic classification system. The system comprises an original image warehousing processing unit, which analyzes original SAR data to obtain basic information, generates an SAR image on the basis of a polarized mode included in the basic information, and enables the basic information and the SAR image to be associated and to be stored in an original image library; the original image library, which stores the SAR image and the basic information; a sample database, which stores a sample image and a sample configuration file; a sample database construction unit, which selects the SAR image from the original image library and makes the SAR image serve as a sample image, sets parameters of the sample image to generate a sample configuration file, and enables the sample image and the corresponding configuration file to be associated and stored in the sample database; and a classification unit, which obtains, from the sample database, a matched sample image and a corresponding configuration file according to the basic information of images to be classified, generates an automatic classification configuration file according to the sample configuration file, and classifies the images to be classified according to the automatic classification configuration file.

Description

technical field [0001] The invention belongs to the technical field of remote sensing image processing, in particular to a SAR image automatic classification method based on sample library management and multiple classification methods. Background technique [0002] Synthetic Aperture Radar (SAR) is an active imaging radar system, which has the characteristics of all-weather, day and night imaging. Among them, the polarimetric SAR system can work in different polarization channel combination modes, and can obtain more ground objects. target information. In recent years, a variety of SAR systems have emerged, and the interpretation and application of SAR information has shown a wider application prospect, playing an increasingly important role in land use, land cover, and target detection. [0003] In the application process of SAR images, image classification is an important content of polarimetric SAR image interpretation. At present, there is no special classification sys...

AI Technical Summary

Problems solved by technology

[0003] In the application process of SAR images, image classification is an important content of polarimetric SAR image interpretation. At present, there is no special classification system for SAR scenes, which cannot effectively manage massive SAR data. SAR processing mainly depends on specific data types. and algorithms, mainly rely on manual visual interpretation, unable to achieve true automatic interpretation

These problems greatly hinder the application potential of SAR images and reduce the interpretation efficiency of SAR images

Images