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

A Discriminant Optimal Eigensubspace Feature Extraction Method for One-dimensional Image of Object Library Attributes

An extraction method and technology of target library, which are applied in the field of feature extraction of the optimal eigensubspace for one-dimensional image discrimination of target library attributes, can solve the problem of increasing the misjudgment rate, increase the overlapping area of ​​library-related target and non-library-related target features, It cannot better describe the essential information of the library belonging to the target, so as to achieve the effect of improving the discriminant performance.

Active Publication Date: 2022-02-08
UNIV OF ELECTRONICS SCI & TECH OF CHINA
View PDF10 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, the features extracted by the conventional eigensubspace method are not optimal, and cannot better describe the essential information of the library genus target, which increases the feature overlap area between the library genus target and the non-library target, 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
  • A Discriminant Optimal Eigensubspace Feature Extraction Method for One-dimensional Image of Object Library Attributes
  • A Discriminant Optimal Eigensubspace Feature Extraction Method for One-dimensional Image of Object Library Attributes
  • A Discriminant Optimal Eigensubspace Feature Extraction Method for One-dimensional Image of Object Library Attributes

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0033] The effectiveness of the solution of the present invention is illustrated below in conjunction with a simulation example.

[0034] Design four point objects: "|" font, "V" font, "dry" font and "small" font. The first three objects ("|" font, "V" font, and "dry" font object) participate in training as library objects, and establish a template library for library objects. The latter type of target ("small" font target) does not participate in training (ie as a non-library target). In the one-dimensional range image of every 1° within the range of target attitude angle (0°∽60°), take all the one-dimensional The range images are used for training, and the one-dimensional range images of other attitude angles are used as test data. The discriminant experiment is carried out on the above simulation data by using the conventional eigensubspace discriminant method and the method of the present invention. In the experiment, the one-dimensional distance image data x of the inp...

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

This paper discloses a method for extracting optimal eigensubspace features for object library attribute one-dimensional image discrimination. The method first uses the training one-dimensional range image sample data of all library objects to establish an optimal eigensubspace, extracts the optimal features of the objects, constructs the feature description library of library objects, and discriminates the library attributes of the input objects. The method establishes the optimal eigensubspace through layer-by-layer optimization, which can better describe the difference information between library-generic objects and non- library-generic objects, thereby improving the discriminative performance of object library attributes. Simulation experiments on four types of targets verify the effectiveness of the method.

Description

technical field [0001] The invention belongs to the technical field of radar target recognition, and relates to a method for extracting an optimal eigensubspace feature for discriminating a one-dimensional image of a target database attribute. Background technique [0002] The one-dimensional range image of the target contains information about the structure and shape of the target. Compared with the low-resolution radar, this information is more conducive to the classification of the target. Therefore, the target recognition based on the one-dimensional range image has become the current radar target recognition. hotspot. [0003] The conventional pattern recognition method is to use the training sample data set of the target for training in advance, establish the feature template library of the target, and then identify the input data of the target that participated in the training. The recognition methods of one-dimensional range image have obtained good recognition resu...

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): G06V10/46
CPCG06V10/462G06V2201/07
Inventor 周代英沈晓峰廖阔梁菁张瑛冯健
Owner UNIV OF ELECTRONICS SCI & TECH OF CHINA