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

Space-time image texture direction detection method and system based on DCNN and transfer learning

An image texture and direction detection technology, applied in neural learning methods, image data processing, image analysis and other directions, can solve problems such as low signal-to-noise ratio, inability to eliminate image noise, false detection, etc., to eliminate image noise and improve detection accuracy. rate effect

Active Publication Date: 2021-08-06
HOHAI UNIV
View PDF2 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

The key core of spatio-temporal image velocity measurement is how to realize the complex nonlinear data prediction process of accurately detecting the main direction of texture from spatio-temporal images. Traditional spatio-temporal image texture direction detection methods cannot eliminate image noise and have strong turbulence conditions for flow velocity fluctuations. If the signal-to-noise ratio is too low, it may cause false detection and other problems.

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
  • Space-time image texture direction detection method and system based on DCNN and transfer learning
  • Space-time image texture direction detection method and system based on DCNN and transfer learning
  • Space-time image texture direction detection method and system based on DCNN and transfer learning

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0027] like Figure 1 ~ 6 As shown, a time-space image texture direction detection method based on DCNN and migration learning, including: a time and space image; a predictive time-space image texture direction prediction model based on deep convolutional neural network and migration learning, Get the prediction result in the temporal image texture direction.

[0028] like Figure 5 As shown, a two-layer convolutional layer base layer, four convolution layer groups (first to fourth volume group), all-connection layer, and all-in-plane texture, and four convolutional layer groups (first to fourth volume layers), full connecting layer and Return layer. Specifically, including 2 layer convolution layer base layer, 9 layers of convolution layer constitutes a first rolling group, and the 12-layer convolution layer constitutes a second roll group, and the 69-layer convolution layer constitutes a third volume group. The 9-layer convolution layer constitutes a fourth volume group; each of t...

Embodiment 2

[0056] According to an embodiment, a time-space image texture direction detecting method based on DCNN and migration, this embodiment provides a time and space image texture direction detection system based on DCNN and migration learning, including: data acquisition module for collecting time and space images Time and Space Image Texture Direction Prediction Module, used to introduce the acquisition time and time-space image input well prediction model based on a deep convolutional neural network and migration learning, and acquire prediction results in the temporal image texture direction.

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 time-space image texture direction detection method and system based on a DCNN (deep convolutional neural network) and transfer learning in the technical field of time-space image texture direction recognition. The method comprises the following steps: acquiring a time-space image; inputting the collected space-time image into a trained space-time image texture direction prediction model based on a deep convolutional neural network and transfer learning, and obtaining a prediction result of the space-time image texture direction; the image noise is eliminated, and the detection accuracy under the turbulent flow condition with strong flow velocity pulsation is improved.

Description

Technical field [0001] The present invention belongs to the field of time and space image texture, and more particularly to a time-time image texture direction detection method and system based on DCNN (deep convolutional neural network) and migration learning. Background technique [0002] The river water surface imaging speed method is an emerging technology applied to a large-scale water surface flow rate field non-contact quantitative measurement. It is of great significance for rendering the flow of rivers, revealing the rule of prototyping, and has important significance of online monitoring of flood flow. It is river dynamics. The foundation of scientific research such as river water and literature. [0003] As a typical application of the river water surface imaging speed method, the Time and Space Image Speedometer (STIV) is a measurement line as the analysis area, by detecting the texture of the synthetic time and space estimation of one - day flow rate measurement meth...

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): G06T7/00G06T7/40G06N3/04G06N3/08
CPCG06T7/0002G06T7/40G06N3/084G06N3/08G06N3/045
Inventor 张振李华宝陈林刘志杰莫岱辉蒋芸
Owner HOHAI UNIV
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