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

Strong-noise single-photon three-dimensional reconstruction method based on multi-stage degeneration neural network

A neural network and 3D reconstruction technology, applied in neural learning methods, biological neural network models, constraint-based CAD, etc., can solve problems such as imaging performance degradation, and achieve high feasibility, strong universality, and obvious effects

Active Publication Date: 2022-07-01
NANJING UNIV OF POSTS & TELECOMM
View PDF7 Cites 6 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Both methods have improved the reconstruction accuracy, but as the luminous flux gradually decreases, the imaging performance of the two is seriously degraded.

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
  • Strong-noise single-photon three-dimensional reconstruction method based on multi-stage degeneration neural network
  • Strong-noise single-photon three-dimensional reconstruction method based on multi-stage degeneration neural network
  • Strong-noise single-photon three-dimensional reconstruction method based on multi-stage degeneration neural network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0048] In order to deepen the understanding of the present invention, the present invention will be described in further detail below with reference to the embodiments, which are only used to explain the present invention and do not constitute a limitation on the protection scope of the present invention.

[0049] Strong noise single-photon 3D reconstruction method based on multi-stage degenerate neural network, such as figure 1 shown, including the following steps:

[0050] Step 1, get the dataset;

[0051] The data set includes a training data set for network model training and a test data set for network model testing; the training data set is a noise single photon obtained by simulating the NYUV2 data set using a single-photon lidar simulation system model. Photon dataset, the test dataset is a noisy single-photon dataset obtained by simulating the Middlebury 2005 dataset using the same single-photon lidar simulation system model.

[0052] The specific modeling of the si...

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 laser radars, and particularly relates to a high-noise single-photon three-dimensional reconstruction method based on a multi-stage degeneration neural network. The method solves the problems that a classical single-photon imaging method is poor in imaging effect under the low signal photon to background noise ratio, some hyper-parameters need to be adjusted to maintain precision and calculation efficiency, and practicability is limited. The method mainly comprises the following steps that 1, a data set is obtained; step 2, constructing a multi-stage degeneration neural network for single photon reconstruction; 3, training the network by using the training set, verifying the network, and observing whether the trained network achieves an expected effect or not; and step 4, inputting test set data into the network to realize strong-noise single-photon three-dimensional reconstruction, and recovering a depth map.

Description

technical field [0001] The invention belongs to the technical field of laser radar, and in particular relates to a three-dimensional reconstruction method of strong noise single photon based on a multi-stage degenerate neural network. Background technique [0002] Traditional 3D imaging lidar is a common active optical 3D imaging system, which has been widely used in industry, scientific research and even national defense. However, with the maturity of high-sensitivity photon detection technology and high-precision electronic timing technology, lidar technology represented by time-correlated single-photon counting three-dimensional imaging technology has gradually become a development trend in the field of long-distance imaging and non-horizontal imaging. [0003] Although the single-photon counting radar can respond to the echo photon signal at the single-photon level, after adopting the single-photon counting system, while improving the detection sensitivity, the scatterin...

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): G06F30/27G06T17/00G06N3/04G06N3/08G06F111/04G06F119/10
CPCG06F30/27G06T17/00G06N3/08G06N3/084G06F2111/04G06F2119/10G06N3/045
Inventor 陈颖豪王琴李剑陈彦昆
Owner NANJING UNIV OF POSTS & TELECOMM
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