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

Concrete crack identification method and device based on video semantic segmentation technology

A technology of semantic segmentation and crack recognition, applied in neural learning methods, character and pattern recognition, image analysis, etc., can solve the problems of low detection efficiency, high computing cost, poor timeliness, etc., and achieve strong robustness and generalization. , the effect of fast network convergence and reduced dependence

Active Publication Date: 2020-07-31
ZHEJIANG UNIV
View PDF3 Cites 6 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0010] The purpose of the present invention is to provide a concrete crack recognition method and device based on video semantic segmentation technology to solve the current commonly used deep learning CNN method similar to exhaustive, which requires a large amount of data, high computing cost, low detection efficiency, and The problem of poor timeliness

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
  • Concrete crack identification method and device based on video semantic segmentation technology
  • Concrete crack identification method and device based on video semantic segmentation technology
  • Concrete crack identification method and device based on video semantic segmentation technology

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0048] In order to make the object, technical solution and advantages of the present invention clearer, the present invention will be further described in detail below in conjunction with examples of concrete cracks. It should be understood that the specific embodiments described here are only used to explain the present invention, not to limit the present invention.

[0049] In view of the currently commonly used deep learning CNN methods similar to exhaustive, a large amount of data is required, the calculation cost is high, the detection efficiency is low, and the timeliness is poor. The invention is a method that can greatly reduce the data volume demand of the deep learning model of concrete cracks, and can greatly improve the detection efficiency and timeliness, and has strong robustness, good generalization ability and high detection The efficient and accurate crack detection technology is original and more suitable for engineering applications.

[0050] The applicatio...

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 concrete crack identification method and device based on a video semantic segmentation technology, and belongs to the technical field of concrete structure damage detection,and the method comprises the steps: obtaining a crack video, and manually labeling a label in a video picture frame; predicting a future frame and a future label for the labeled frame by using a spatial displacement convolution block, and spreading the future frame and the future label at the same time to obtain a synthetic sample and pre-process the synthetic sample to form a crack database; modifying input and output ports and parameters of data of Deeplabv <3+> to enable the Deeplabv <3+> to accept video input, and establishing a CVN model through video output; taking a convolutional layerin the trained Deeplabv <3+> network as an initial weight of a CVN model for migration; inputting the crack database into the migrated CVN model, and training a concrete crack detection semantic segmentation model CVN for crack data. Compared with a convolutional neural classification network, the method reduces the requirement for the data size, a target can be rapidly and accurately recognized through video input and video output, and the method has practical engineering significance.

Description

technical field [0001] The invention belongs to the technical field of concrete structure damage detection, and in particular relates to a concrete crack recognition method and device based on video semantic segmentation technology. Background technique [0002] Concrete is currently the most widely used building material and is widely used in the construction of infrastructure such as roads, bridges, tunnels, and industrial and civil construction. The damage on the surface of the concrete structure, including cracks, weathering, holes and spalling, etc., visually reflects the durability and safety of the concrete structure. Among them, cracks are the type of damage that causes the most damage to structures and attracts the most attention. Regular crack detection plays a very important role in the maintenance and operation of infrastructure. According to the characteristics such as the shape and position of cracks, the degree of damage inside the structure and the cause of ...

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): G06T7/00G06N3/04G06N3/08G06T7/11
CPCG06T7/0008G06T7/11G06N3/08G06T2207/10016G06N3/045G06T2207/30132G06T2207/20084G06T7/0004G06V10/82G06V20/49G06T7/10G06T2207/20081G06V20/41
Inventor 申永刚俞臻威
Owner ZHEJIANG 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