Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Target recognition model construction and recognition method and device based on computational ghost imaging

A target recognition and construction method technology, applied in the recognition method and device, in the field of target recognition model construction based on computational ghost imaging, can solve the problem of low recognition efficiency, and achieve the effect of reducing hardware requirements and saving imaging calculation time

Active Publication Date: 2020-09-11
NORTHWEST UNIV(CN)
View PDF4 Cites 6 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] The purpose of the present invention is to provide a target recognition model construction, recognition method and device based on computational ghost imaging, to solve the problems in the prior art based on computational ghost imaging The target recognition method has the problem of low recognition efficiency

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
  • Target recognition model construction and recognition method and device based on computational ghost imaging
  • Target recognition model construction and recognition method and device based on computational ghost imaging
  • Target recognition model construction and recognition method and device based on computational ghost imaging

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0065] In this embodiment, a method for constructing a target recognition model based on computational ghost imaging is disclosed, and the method is executed according to the following steps:

[0066] Step 1. Collect a plurality of first images to obtain a first image set; obtain the label of each first image in the first image set to obtain a label set; the first image includes the target image to be recognized or the same type of the target image to be recognized image;

[0067] In the present invention, similar images of the target image to be recognized refer to images with the same semantics as the target image to be recognized. In the present invention, semantic similarity refers to images belonging to the same category, including plants, animals, daily necessities, etc., such as biological Scientists classify many objects in nature according to family, genus and species. For example, plants are a large category, and its subcategories include orchids, chrysanthemums, pop...

Embodiment 2

[0104] In this embodiment, a target recognition method based on computational ghost imaging is disclosed, and the method is executed according to the following steps:

[0105] Step A, using computational ghost imaging equipment to perform R on the target to be identified 2 times measurement, get R 2 A measured light intensity value is obtained to obtain a sequence of measured light intensity values, wherein R is a positive integer;

[0106] Step B. Input the measured light intensity value sequence into the target recognition model obtained by the target recognition model construction method based on computational ghost imaging in Embodiment 1, and obtain the recognition result.

[0107] In this embodiment, the computational ghost imaging device includes a light source 1, a lens I2, a lens II3, a spatial light modulator 4, a target to be measured 5, a lens III6, a barrel detector 7 and a PC8. The light emitted by the light source 1 is irradiated on the spatial light modulator...

Embodiment 3

[0112] In this embodiment, a device for building a target recognition model based on computational ghost imaging is provided, wherein the device includes an image acquisition module, a simulated light intensity processing module, and a model building module;

[0113] The image acquisition module is used to collect a plurality of first images to obtain a first image set; obtain the label of each first image in the first image set to obtain a label set; the first image includes a target image to be recognized or a target image to be recognized similar images of

[0114] The simulated light intensity processing module is used to perform simulated light intensity processing on each first image in the first image set, obtain a sequence of simulated light intensity values ​​of each first image, and obtain a set of simulated light intensity values;

[0115] The model building module is used to use the simulated light intensity value set as input and the label set as reference output ...

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 discloses a target recognition model construction and recognition method and device based on computational ghost imaging. A target recognition model special for a light intensity value collected by ghost imaging equipment is designed, the characteristics of ghost imaging data calculation are fully considered by a double-path neural network, the light intensity value and a measurementmatrix are used as input information at the same time, and the accuracy of fusion of the output of the two paths of networks is higher than that of independent input of the light intensity information; the method achieves the recognition of an object in a calculation ghost imaging system without an imaging step, saves the imaging calculation time, and completes the quick target recognition. According to the method, imaging calculation is skipped, the imaging calculation time is saved compared with the process of imaging first and then recognition by using a classical ghost imaging calculationmethod, rapid target recognition by using light intensity information is realized, and the hardware requirement of the system is reduced.

Description

technical field [0001] The invention relates to a target recognition method, in particular to a target recognition model construction and recognition method and device based on computational ghost imaging. Background technique [0002] Target recognition has been widely used in military, transportation, medical and other fields, including face recognition, flying object recognition, license plate recognition, etc. A more general method is to directly acquire the target image through an area array detector with spatial resolution, and then realize target recognition through computer vision. However, the area array detector has a limited spectral range and needs to work at a full sampling rate, which has high computational complexity and high hardware requirements. [0003] As an indirect imaging method, ghost imaging uses the second-order and even higher-order correlation properties of the light field to reconstruct the image. By means of area array detection and acquisitio...

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 Applications(China)
IPC IPC(8): G06K9/00G06K9/62G06N3/04G06N3/08
CPCG06N3/08G06V20/00G06N3/045G06F18/241G06F18/214
Inventor 乐明楠李建波张薇李斌赵国英彭进业章勇勤赵万青王珺
Owner NORTHWEST UNIV(CN)
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

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

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com