SAR image classification based on multi-feature and composite kernel

Abstract

A SAR image classification method based on multiple features and composite kernel comprises the following steps: 1, inputting an image; 2, extracting spatial features and structural feature by using gray level co-occurrence matrix and multi-level local pattern histogram; 3, fusing the extracted spatial features and the structural features to form a feature fusion matrix; 4, construct a training sample set and a test sample set; 5, generating the superpixel by using the generalized likelihood ratio and the like; 6. Weighing The traditional feature kernel provided by radial basis function and the context information kernel composed of super pixels into a composite kernel to form a new support vector machine. 7, performing classification; 8. Calculating Classification Precision. By using theinvention for classification, the influence of speckle noise can be effectively reduced, the accurate classification of the SAR image is realized, the classification precision is effectively improved,and the method can be used for target recognition and tracking of the SAR image.

Description

technical field [0001] The invention belongs to the technical field of image processing, and in particular relates to a SAR image classification method based on multi-features and composite kernels, which is applied to the classification of SAR images to realize target recognition and tracking. Background technique [0002] Synthetic Aperture Radar (SAR) can acquire high-quality images of different land covers at any time under any weather conditions. Therefore, SAR has been successfully applied in many fields, such as environmental monitoring, land resource mapping and military systems. In recent years, SAR image classification has received increasing attention as an important part of image understanding and interpretation. However, the inherent multiplicative speckle noise and high intra-class variation in SAR images make it difficult for classification methods to obtain satisfactory classification results. How to classify with high precision is still a challenging proble...

AI Technical Summary

Problems solved by technology

[0005] The purpose of the present invention is to address the shortcomings of the above-mentioned prior art, that is, in the SAR image, a single feature cannot fully extract rich information and a single kernel cannot accurately handle the classification problem, and proposes a SAR based on multi-features and composite kernels Image classification method, which proposes a multi-feature extraction combining GLCM and MLPH and a method of applying composite kernel in support vector machine, spatial and structural information is captured by multi-feature extraction, and composite kernel is used to combine texture information and context information weighting integrated into the support vector machine to form a new type of support vector machine

Images