Supercharge Your Innovation With Domain-Expert AI Agents!

Dam crack detection method based on multi-transfer learning model fusion

A technology of model fusion and transfer learning, applied in the field of image recognition, can solve the problems of non-adaptiveness, ineffective denoising effect, and low edge detection accuracy

Active Publication Date: 2021-02-09
HOHAI UNIV
View PDF5 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] Although the traditional detection algorithm is simple, the denoising effect is not obvious, the edge detection accuracy is not high, and it is not adaptive, so it is not completely suitable for the detection of underwater dam cracks

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
  • Dam crack detection method based on multi-transfer learning model fusion
  • Dam crack detection method based on multi-transfer learning model fusion
  • Dam crack detection method based on multi-transfer learning model fusion

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0067] Embodiments of the invention are described in detail below, examples of which are illustrated in the accompanying drawings. The embodiments described below by referring to the figures are exemplary only for explaining the present invention and should not be construed as limiting the present invention.

[0068] Due to complex environmental factors, the collected images of dam cracks generally have problems such as unclear models, dark brightness, and low contrast. At the same time, too few data sets can easily lead to overfitting of the model. To solve these problems, data augmentation is first performed on dam crack images, and at the same time transfer learning is considered using models trained on road, wall, and bridge crack datasets. After obtaining multiple transfer learning models, try to fuse the multi-model data to obtain more accurate frame regression results. Based on this idea, the present invention proposes a dam crack detection method based on the fusion o...

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 relates to a dam crack detection method based on multi-transfer learning model fusion and relates to the field of image recognition, in particular to a multi-model fusion dam crack detection method. The method first collects road, wall, bridge and dam crack data sets, and performs data enhancement processing on the dam crack picture data set; builds a MobileNet-SSD target detection model, and replaces the original VGG network in the SSD algorithm with a MobileNet network structure; Carry out model training; after the training is completed, extract the MobileNet structure parameters that have been trained in the road crack detection model and wall crack detection model, import the untrained MobileNet-SSD, and freeze the MobileNet structure; use the dam cracks after data enhancement The data set is transferred and learned, and after obtaining multiple models, the multiple models are fused and calculated to improve the accuracy of dam crack detection.

Description

technical field [0001] The invention relates to a dam crack detection method based on multi-transfer learning model fusion and relates to the field of image recognition, in particular to a multi-model fusion dam crack detection method. Background technique [0002] In the past few decades, my country's water conservancy construction has achieved tremendous development. As the most important part of water conservancy construction, dams have exerted huge engineering benefits in our country. However, the dam has been in a complex environment for a long time, and cracks will inevitably occur. Dam cracks are a great hidden danger to dam safety, and the detection of dam cracks is extremely important. However, the underwater environment is complex, and the collected fracture images have shortcomings such as unclear models, dark brightness, and low contrast, making fracture detection extremely difficult. [0003] In response to these problems, many scholars have carried out researc...

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): G06T7/00G06K9/62
CPCG06T7/0004G06T2207/20081G06T2207/20084G06T2207/30132G06F18/25
Inventor 陈峙宇刘凡郑豪杨赛
Owner HOHAI UNIV
Features
  • R&D
  • Intellectual Property
  • Life Sciences
  • Materials
  • Tech Scout
Why Patsnap Eureka
  • Unparalleled Data Quality
  • Higher Quality Content
  • 60% Fewer Hallucinations
Social media
Patsnap Eureka Blog
Learn More

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

© 2025 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com