Unlock instant, AI-driven research and patent intelligence for your innovation.

An Unknown Target Discrimination Method Based on Threshold Optimal Dictionary Learning

A technology of dictionary learning and discrimination methods, applied in character and pattern recognition, instruments, calculations, etc., can solve the problem of increased misjudgment rate and achieve the effect of reducing the impact and improving the discrimination rate

Active Publication Date: 2022-04-22
UNIV OF ELECTRONICS SCI & TECH OF CHINA
View PDF5 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, the conventional subspace method cannot extract robust features at a low SNR, resulting in a significant increase in the misjudgment rate

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
  • An Unknown Target Discrimination Method Based on Threshold Optimal Dictionary Learning
  • An Unknown Target Discrimination Method Based on Threshold Optimal Dictionary Learning
  • An Unknown Target Discrimination Method Based on Threshold Optimal Dictionary Learning

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0035] The following analysis of the practicality of the present invention in conjunction with simulation experiments.

[0036]Design four point goals: "|" Glyphs, "V" glyphs, "stem" fonts, and "small" fonts. The first three goals ("|" Glyphs, "V" fonts, "stem" glyph targets) participate in training as known targets, establishing a threshold optimization dictionary for known targets. The latter target (the "small" font target) does not participate in the training (i.e. as an unknown target). In the one-dimensional distance image of the target attitude angle of (0 °∽60 °) every 1°, the one-dimensional distance image of all the training target attitude angles of 0°, 2°, 4° 、...、60° is taken for training, and the one-dimensional distance image of the remaining attitude angle is used as the test data. Using conventional eigens subspace discriminant method and the method of the present invention to conduct discriminant experiments on the above simulation data.

[0037] In the simulati...

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 belongs to the technical field of radar target recognition, in particular to an unknown target discrimination method based on threshold optimization dictionary learning. The method of the present invention first uses the training data set to establish a super-complete dictionary for the known target through the dictionary learning method based on threshold optimization, and then performs sparse analysis on the one-dimensional range image samples of the target to be discriminated based on the dictionary, and calculates the reconstruction error to realize Discrimination of unknown targets. Due to the introduction of threshold optimization in dictionary learning, the influence of noise on dictionary learning is reduced, and a robust over-complete dictionary can still be obtained when the signal-to-noise ratio is low, thereby improving the discrimination rate of unknown targets. Simulation experiments on five types of targets verify the effectiveness of the method.

Description

Technical field [0001] The present invention belongs to the field of radar target recognition technology, specifically relates to an unknown target discrimination method based on threshold optimization dictionary learning. Background [0002] Broadband radar can obtain a one-dimensional distance image of the target, because the one-dimensional distance image contains information about the structure and shape of the target, relative to the low-resolution radar, this information is more conducive to the classification of the target, at the same time, relative to the two-dimensional image, the one-dimensional distance image is easy to obtain, and can achieve real-time recognition, therefore, based on the one-dimensional distance image of the target identification has become the current radar target recognition hotspot. [0003] When the target to be identified belongs to the training target, many conventional recognition methods based on one-dimensional distance image can obtain go...

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): G06K9/62
CPCG06F18/40
Inventor 周代英张瑛梁菁沈晓峰冯健
Owner UNIV OF ELECTRONICS SCI & TECH OF CHINA