Check patentability & draft patents in minutes with Patsnap Eureka AI!

SAR Target Recognition Method Fused with Convolutional Features and Integrated Extreme Learning Machine

An ultra-limited learning machine and target recognition technology, applied in the field of automatic target recognition of SAR images, can solve the problem of long model training time, and achieve the effect of improving real-time performance and generalization ability, less adjustable parameters, and high recognition accuracy

Active Publication Date: 2020-04-28
HANGZHOU DIANZI UNIV
View PDF6 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

In addition, recognition methods based on deep learning usually have more model parameters, how to set the initial value of the model and the hyperparameters of model training are issues worth studying; (3) the model training time is long

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • SAR Target Recognition Method Fused with Convolutional Features and Integrated Extreme Learning Machine
  • SAR Target Recognition Method Fused with Convolutional Features and Integrated Extreme Learning Machine
  • SAR Target Recognition Method Fused with Convolutional Features and Integrated Extreme Learning Machine

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0013] The present invention will be further described below in conjunction with accompanying drawing.

[0014] Depend on figure 1 Shown, the specific implementation steps of the present invention are as follows:

[0015] Step (1). Preprocessing the original SAR image, reducing the influence of coherent speckle noise through a filtering algorithm, and extracting the SAR target slice image. details as follows:

[0016] Use the mean filtering algorithm to filter the input SAR image, take the two-dimensional center point of the image plane as the coordinate origin, extract the image of the target area with a size of w×h, and scale it so that the image resolution will be 32×32, Divided by 255.0, so that the gray level of the image is in the interval [0 1]. Denote the obtained SAR target slice image as P.

[0017] Step (2). Randomly generate a certain number of two-dimensional convolution kernels with different kernel widths, use these convolution kernels to filter the SAR imag...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention relates to a SAR target recognition method of fusing convolution features and integrating extreme learning machines. In the prior art, when deep convolutional neural networks are applied to SAR image target recognition, in order to improve the recognition accuracy, problems such as sample expansion, model optimization design, and long-term training need to be solved. In order to solve the problem that the accuracy of the classification results and the recognition speed cannot be improved at the same time due to the lack of the number of target samples of each category in the SAR target recognition, the present invention performs filtering and pooling operations on the scaled image through a random convolution kernel , use the method of random extraction to reduce the dimensionality of the obtained features, and finally use the integrated extreme learning machine to obtain the classification result. The method proposed by the invention not only greatly improves the recognition speed, but also improves the recognition accuracy, and has better application prospects.

Description

technical field [0001] The invention belongs to the field of SAR (Synthetic Aperture Radar) image automatic target recognition, and relates to a SAR target recognition method combining random convolution features and an integrated extreme learning machine. Background technique [0002] SAR image automatic target recognition is a key research direction of SAR image interpretation. SAR image target recognition uses data processing methods to classify and identify targets. Its working process is to first find the region of interest in the SAR image, and then classify each region of interest to determine its category. SAR image target recognition has a wide range of applications in the national economy and national defense construction, such as marine monitoring systems, ship target recognition, mine detection, etc. [0003] Feature extraction and classifier design are two key factors affecting target recognition in SAR images. Feature extraction can be divided into broad and ...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Patents(China)
IPC IPC(8): G06K9/62G06N3/08
CPCG06N3/08G06F18/285G06F18/24G06F18/214
Inventor 谷雨徐英冯秋晨郭宝峰
Owner HANGZHOU DIANZI UNIV
Features
  • R&D
  • Intellectual Property
  • Life Sciences
  • Materials
  • Tech Scout
Why Patsnap Eureka
  • Unparalleled Data Quality
  • Higher Quality Content
  • 60% Fewer Hallucinations
Social media
Patsnap Eureka Blog
Learn More

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2025 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com