SAR image classification method based on shrinkage autoencoder

Abstract

The present invention disclosed a SAR image classification method based on the shrinkable automatic encoder.The steps of realizing the present invention are: (1) Enter the image; (2) the smooth wave decomposition; (3) select the training sample;Contraction self -compilation; (6) Construct sample feature collection; (7) training SoftMax classifier; (8) classification.Compared with the existing technology multi -level local mode, the characteristic extraction method of the multi -level local model has the advantages of high classification accuracy, good regional consistency, and accurate edge classification.EssenceThe invention can be applied to the synthetic aperture radar SAR image target detection and target recognition.

Description

technical field [0001] The invention belongs to the technical field of image processing, and further relates to a SAR (Synthetic Aperture Radar) image classification method based on a contraction autoencoder in the technical field of target recognition. The invention can be applied to the acquisition of target recognition image information and SAR image target recognition, and can accurately classify different regions of the image. Background technique [0002] Synthetic Aperture Radar (SAR) is an active earth observation system. Compared with other sensors such as optical and infrared sensors, SAR imaging is less limited by conditions such as atmosphere and illumination. Ground reconnaissance, with high resolution, large width and other characteristics. SAR can effectively identify camouflage and penetrate cover, so it has been widely used in military and civilian fields such as remote sensing mapping, military reconnaissance, earthquake relief, etc., and has become an ind...

AI Technical Summary

Problems solved by technology

The disadvantage of this method is that the feature extraction of this method is too simple, and the rich texture information in the SAR image is not considered, and the classification results obviously appear in the confusion of regional classification and the phenomenon of irregular edges.

The disadvantage of this method is that the process of image feature extraction relies too much on grayscale features, and the extraction of texture features is not complete. At the same time, the unsupervised classification method will cause a large loss in classification accuracy, which will cause the final SAR Image classification results with low accuracy

Images