Building seismic damage grade classification method based on deep learning

A technology of deep learning and hierarchical classification, applied in the field of hierarchical classification, to achieve the effects of easy access, reduced interference, and reduced manpower and material resources

Active Publication Date: 2021-02-19
HARBIN INST OF TECH
View PDF10 Cites 7 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0006] The purpose of the present invention is to solve the problem of accurately classifying the damage level of buildings in the earthquake damage area in the earthquake damage assess

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
  • Building seismic damage grade classification method based on deep learning
  • Building seismic damage grade classification method based on deep learning
  • Building seismic damage grade classification method based on deep learning

Examples

Experimental program
Comparison scheme
Effect test

Example Embodiment

[0067]Example

[0068]CombineFigure 5 According to the post-earthquake building image data and the post-earthquake building image data collected on the Internet, the post-earthquake building damage level classification method is based on the deep learning of the present invention:

[0069]The first step is specifically: the post-earthquake building image data and the high-resolution post-earthquake building image data of different structures collected from the Internet form image set A. According to the format of the Cityscapes data set, use the Labelme tool to analyze each of the images. Class objects are labeled and named. Produce diversified building image data sets after the earthquake.

[0070]The second step is specifically: applying DeepLabV3+ as the basic image segmentation model, where the resolution of the input image of the input layer is not limited, and pre-training the image segmentation model using the Cityscapes data set to obtain the model MSP . Select the classification you...

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 provides a building seismic damage grade classification method based on deep learning, and relates to the field of seismic engineering, and the method comprises the steps: obtaining a seismic damage and damage image based on the photographing of a single building in a region, and carrying out the quick and accurate segmentation of a building in the photographed image through an imagesegmentation neural network based on deep learning, extracting a seismic building from a complex image background; and then carrying out seismic damage classification on the segmented building through an image classification neural network based on deep learning. According to the method, accurate damage classification can be quickly and efficiently carried out on buildings in the seismic area, the image classification method based on deep learning can replace on-site manual evaluation by utilizing the calculation capability and the quick decision capability of the deep neural network, the time and energy consumption is greatly reduced, under the condition that the accuracy is guaranteed, whether buildings in a large-range seismic area collapse or not is efficiently judged.

Description

technical field [0001] The invention belongs to the technical field of earthquake engineering, in particular to a method for classifying earthquake damage levels of buildings based on deep learning. Background technique [0002] In the earthquake damage assessment of the building area, it is necessary to quickly assess the damage of the building in the earthquake-damaged area after the earthquake, so as to obtain the damage level and quantity of the building after the earthquake in time, which is convenient for auxiliary decision-making and emergency rescue after the earthquake. . How to accurately classify the damage levels of buildings in the earthquake-damaged area and at the same time meet the timeliness requirements, that is, to investigate the distribution and quantity of disasters in a timely manner after the earthquake and implement a rapid assessment of earthquake damage is an effective way to reduce the loss of earthquake disasters . Based on this situation, it i...

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
IPC IPC(8): G06K9/00G06K9/34G06K9/62G06N3/04G06N3/08
CPCG06N3/08G06V20/13G06V10/267G06N3/045G06F18/241G06F18/214
Inventor 黄永于建琦林旭川李惠
Owner HARBIN INST OF TECH
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products