CN116630814B - Quick positioning and evaluating method for building disasters based on machine learning - Google Patents
Quick positioning and evaluating method for building disasters based on machine learning Download PDFInfo
- Publication number
- CN116630814B CN116630814B CN202310904906.5A CN202310904906A CN116630814B CN 116630814 B CN116630814 B CN 116630814B CN 202310904906 A CN202310904906 A CN 202310904906A CN 116630814 B CN116630814 B CN 116630814B
- Authority
- CN
- China
- Prior art keywords
- building
- image
- disaster
- neural network
- images
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 41
- 238000010801 machine learning Methods 0.000 title claims abstract description 13
- 238000013527 convolutional neural network Methods 0.000 claims abstract description 44
- 238000011156 evaluation Methods 0.000 claims abstract description 31
- 238000007781 pre-processing Methods 0.000 claims abstract description 7
- 238000012549 training Methods 0.000 claims description 21
- 238000001914 filtration Methods 0.000 claims description 12
- 230000006870 function Effects 0.000 claims description 6
- 238000012545 processing Methods 0.000 claims description 5
- 238000012876 topography Methods 0.000 claims description 5
- ORILYTVJVMAKLC-UHFFFAOYSA-N Adamantane Natural products C1C(C2)CC3CC1CC2C3 ORILYTVJVMAKLC-UHFFFAOYSA-N 0.000 claims description 3
- 238000013528 artificial neural network Methods 0.000 claims description 3
- 238000001514 detection method Methods 0.000 claims description 3
- 238000011176 pooling Methods 0.000 claims description 3
- 238000000547 structure data Methods 0.000 claims description 3
- 230000002401 inhibitory effect Effects 0.000 claims description 2
- 230000004807 localization Effects 0.000 claims 2
- 230000002708 enhancing effect Effects 0.000 claims 1
- 238000011160 research Methods 0.000 description 3
- 238000000605 extraction Methods 0.000 description 2
- 230000001537 neural effect Effects 0.000 description 2
- 230000006978 adaptation Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000009977 dual effect Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/10—Terrestrial scenes
- G06V20/176—Urban or other man-made structures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/0464—Convolutional networks [CNN, ConvNet]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/084—Backpropagation, e.g. using gradient descent
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
- G06Q50/08—Construction
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/74—Image or video pattern matching; Proximity measures in feature spaces
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/764—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using classification, e.g. of video objects
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/82—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using neural networks
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A10/00—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE at coastal zones; at river basins
- Y02A10/40—Controlling or monitoring, e.g. of flood or hurricane; Forecasting, e.g. risk assessment or mapping
Abstract
The invention discloses a quick positioning and evaluating method for building disasters based on machine learning, which comprises the steps of obtaining building disaster images, preprocessing the building disaster images, constructing a convolutional neural network of building components, optimizing the convolutional neural network, inputting the preprocessed building disaster images into the optimized convolutional neural network, extracting features of the preprocessed building disaster images, classifying the preprocessed building disaster images according to the features to obtain classified images, cutting the classified images to obtain refined images, creating BIM (building information model) according to the building components, inputting the refined images into the BIM to correlate and identify the building components, obtaining correlation content, integrating the correlation content by using a time axis to obtain the variation trend of the building components, and evaluating disaster grades according to the variation trend. The method not only can improve the quick positioning evaluation precision, but also has better interpretation.
Description
Technical Field
The invention relates to the technical field of positioning evaluation, in particular to a quick positioning evaluation method for building disasters based on machine learning.
Background
The positioning evaluation technology is widely applied in the field of building disasters, and can help building managers to timely and efficiently acquire disaster positions of buildings and reasons for disasters, so that rapid positioning and evaluation of building disasters are realized. At present, the building disasters have the characteristics of huge image quantity, various building components, high information density and the like, and the rapid positioning evaluation of the building disasters has more uncertain factors, so that the rapid positioning evaluation of the building disasters has larger uncertainty. Although some building disaster rapid positioning evaluation methods have been invented, the problem of uncertainty of building disaster rapid positioning evaluation cannot be effectively solved.
Disclosure of Invention
The invention aims to provide a quick positioning and evaluating method for building disasters based on machine learning.
In order to achieve the above purpose, the invention is implemented according to the following technical scheme:
the invention comprises the following steps:
a, acquiring a building disaster image, and preprocessing the building disaster image, wherein the building disaster image comprises positioning information;
b, constructing a convolutional neural network of a building member, optimizing the convolutional neural network, inputting the preprocessed building disaster image into the optimized convolutional neural network, extracting the features of the preprocessed building disaster image, and classifying the preprocessed building disaster image according to the features to obtain a classified image;
c, performing image cutting on the classified images to obtain refined images, creating a BIM (building information modeling) according to building components, inputting the refined images into the BIM for association and identification of the building components, and obtaining association content;
and D, integrating the associated content by using a time axis to obtain the change trend of the building component, and evaluating the disaster grade according to the change trend.
Further, the preprocessing method in the step a includes converting the building disaster image into a gray image by using a weighted average method, increasing brightness of the gray image by using more pixels, reducing image brightness by using smaller pixels, obtaining an enhanced image, and removing noise of the enhanced image by using gaussian filtering to obtain a processed image.
Further, the method for constructing the building element convolutional neural network comprises the following steps:
constructing different building component images of a research building into a data set, taking AlexNet as a convolutional neural network architecture, adding a convolutional layer, a pooling layer and a full-connection layer, and taking cross entropy as a loss function of the convolutional neural network:
wherein the loss function is C, and the vector corresponding to the kth image is expressed asThe output of the kth Zhang Juanji neural network is +.>;
Adam is used as an optimizer of the convolutional neural network, a data set is input into the convolutional neural model for training, the weight of the convolutional neural network is updated by adopting a back propagation algorithm, and the characteristics and classification rules of building components are learned.
Further, the method for optimizing the convolutional neural network comprises the following steps:
a. initializing a convolutional neural network model parameter, dividing a data set into a plurality of small-batch training image sets, inputting the small-batch training image sets into a convolutional neural network for training, repeating the operation until the last small-batch training image set, and calculating a cross entropy gradient to adjust the convolutional neural network parameter;
wherein the network parameters areThe iteration number is r, the learning rate is +.>Cross entropy gradient +.>;
b. Inputting the small training image set into the adjusted convolutional neural network, randomly initializing a population, and determining the population scale; and continuously iterating the small-batch training image set until the weight gradient is reduced to 0.0001, and obtaining the optimal optimizing path according to the iteration times and the generated random number to obtain the size of the convolution kernel.
Further, the method for obtaining the classification image comprises the steps of identifying the preprocessed building disaster image according to the features, classifying the preprocessed building disaster image according to the building components corresponding to the features if the preprocessed building disaster image is identified, updating the features until the features are matched if the preprocessed building disaster image is not identified, and repeating the identification until all the preprocessed building disaster images are completely identified, wherein the preprocessed building disaster images after the classification of the building components are classified into classification images.
Further, the method for obtaining the thinned image comprises the steps of processing the thinned image by using Gaussian filtering, calculating gradient strength and direction of each pixel point in the thinned image after Gaussian filtering, determining real and potential Gaussian filtering to process edges of the thinned image to obtain an edge image by applying double-threshold detection, and inhibiting isolated weak edge cutting of the edge image to obtain the thinned image.
Further, the method for obtaining the associated content comprises the steps of using the BIM model to note component parameters of the refined image, comparing the spatial distribution, basic features and shapes of the features with the refined image, matching the features with the refined image if the compared results are the same, otherwise traversing other features until the features are matched, outputting a matching result, and marking the refined image on the component in the corresponding BIM model according to the matching result to carry out association and identification to obtain the associated content.
Further, the method for evaluating disaster grades according to the change trend comprises the following steps:
collecting building disaster related data, analyzing the disaster related data and the change trend, determining an index for evaluating the disaster grade, and determining an evaluation standard according to the index and expert experience; and comparing the disaster data with the evaluation standard to determine a disaster grade, wherein the data related to the building disaster comprise meteorological data, geological topography data, building structure data, building inclination data and building vibration data.
The beneficial effects of the invention are as follows:
the invention relates to a quick positioning and evaluating method for building disasters based on machine learning, which has the following technical effects compared with the prior art:
1. according to the invention, through the steps of acquiring the building disaster image, positioning the building image, identifying the building components, cutting the image, correlating the image and analyzing the disaster cause, the accuracy of the rapid positioning evaluation of the building disaster can be improved, so that the accuracy of the rapid positioning evaluation is improved, the labor and time cost can be greatly saved, the working efficiency is improved, the rapid positioning and evaluation of the building disaster can be realized, the potential information of the building disaster image can be timely found and processed, the method has important significance for the rapid positioning evaluation of the building disaster, and the method can adapt to the rapid positioning evaluation requirements of the building disasters of different buildings and different positions, and has a certain universality.
2. The method can comprehensively analyze the image of the building disaster, converts the image positioning problem into the feature extraction problem by utilizing the image feature extraction, and realizes the accurate control of the rapid positioning evaluation of the building disaster by analyzing the relationship between the change trend of the building disaster and the building information. The method not only can improve the accuracy of quick positioning and evaluation of the building disasters, but also has better interpretability, and can be directly applied to a quick positioning and evaluation system of the building disasters.
Drawings
Fig. 1 is a flow chart of steps of a quick positioning and evaluating method for building disasters based on machine learning.
Detailed Description
The invention is further described by the following specific examples, which are presented to illustrate, but not to limit, the invention.
The invention discloses a quick positioning and evaluating method for building disasters based on machine learning, which comprises the following steps:
as shown in fig. 1, in this embodiment, the steps include:
a, acquiring a building disaster image, and preprocessing the building disaster image, wherein the building disaster image comprises positioning information;
b, constructing a convolutional neural network of a building member, optimizing the convolutional neural network, inputting the preprocessed building disaster image into the optimized convolutional neural network, extracting the features of the preprocessed building disaster image, and classifying the preprocessed building disaster image according to the features to obtain a classified image;
c, performing image cutting on the classified images to obtain refined images, creating a BIM (building information modeling) according to building components, inputting the refined images into the BIM for association and identification of the building components, and obtaining association content;
and D, integrating the associated content by using a time axis to obtain the change trend of the building component, and evaluating the disaster grade according to the change trend.
In this embodiment, the preprocessing method in step a includes converting the building disaster image into a gray image by using a weighted average method, increasing the brightness of the gray image by using more pixels, reducing the image brightness by using smaller pixels, obtaining an enhanced image, and removing noise of the enhanced image by using gaussian filtering to obtain a processed image;
in actual evaluation, a problem of subsidence of a building body and wall cracks occurs due to earthquake flood and the like of a place where a abandoned cell is located, 300 building disaster images of the cell are collected as initial data at intervals of one month by taking the cell as a research object, and the collection is continued for one year.
In this embodiment, the method of constructing a convolutional neural network of a building component includes:
constructing different building component images of a research building into a data set, taking AlexNet as a convolutional neural network architecture, adding a convolutional layer, a pooling layer and a full-connection layer, and taking cross entropy as a loss function of the convolutional neural network:
wherein the loss function is C, and the vector corresponding to the kth image is expressed asThe output of the kth Zhang Juanji neural network is +.>;
Adam is used as an optimizer of the convolutional neural network, a data set is input into the convolutional neural model for training, the weight of the convolutional neural network is updated by adopting a back propagation algorithm, and the characteristics and classification rules of building components are learned.
In this embodiment, the method for optimizing the convolutional neural network includes:
a. initializing a convolutional neural network model parameter, dividing a data set into a plurality of small-batch training image sets, inputting the small-batch training image sets into a convolutional neural network for training, repeating the operation until the last small-batch training image set, and calculating a cross entropy gradient to adjust the convolutional neural network parameter;
wherein the network parameters areThe iteration number is r, the learning rate is +.>Cross entropy gradient +.>;
b. Inputting the small training image set into the adjusted convolutional neural network, randomly initializing a population, and determining the population scale; and continuously iterating the small-batch training image set until the weight gradient is reduced to 0.0001, and obtaining the optimal optimizing path according to the iteration times and the generated random number to obtain the size of the convolution kernel.
In the actual evaluation, the size and learning rate of the convolution kernel after optimization are 192x128 and 0.0001 respectively.
In this embodiment, the method for obtaining the classification image includes identifying the preprocessed building disaster image according to the features, classifying the preprocessed building disaster image according to the building components corresponding to the features if the preprocessed building disaster image is identified, updating the features until the features are matched if the preprocessed building disaster image is not identified, and repeating the identifying until all the preprocessed building disaster images are completely identified, wherein the preprocessed building disaster image classified according to the building components is the classification image;
in actual evaluation, the building elements had walls, columns, beams, doors and windows, and the number of images of feature recognition was divided into 130 walls, 17 columns, 26 beams, 69 doors and 58 windows.
In this embodiment, the method for obtaining a thinned image by performing image cutting on the classified image according to the texture color and shape includes processing the thinned image by using gaussian filtering, calculating gradient strength and direction of each pixel point in the thinned image after gaussian filtering, determining real and potential gaussian filtering by applying dual threshold detection, processing edges of the thinned image to obtain an edge image, and suppressing isolated weak edge cutting of the edge image to obtain a thinned image.
In this embodiment, the method for obtaining the associated content includes using the BIM model to note component parameters of the refined image, comparing spatial distribution, basic features and shapes of the features with the refined image, and matching if the comparison result is the same, otherwise traversing other features until matching, outputting a matching result, and marking the refined image on the component in the corresponding BIM model according to the matching result for association and identification to obtain the associated content.
In this embodiment, the method for evaluating disaster grades according to the variation trend includes:
collecting building disaster related data, analyzing the disaster related data and the change trend, determining an index for evaluating the disaster grade, and determining an evaluation standard according to the index and expert experience; comparing the disaster data with the evaluation standard to determine a disaster grade, wherein the data related to the building disaster comprise meteorological data, geological topography data, building structure data, building inclination data and building vibration data;
in practical evaluation, the building vibration data of the building element walls, columns, beams, doors and windows were 0.96, 0.824, 0.762, 0.436, 0.399, respectively, the geological topography data of the building element walls, columns, beams, doors and windows were 0.851, 0.651, 0.585, 0.467, 0.394, the building inclination data of the building element walls, columns, beams, doors and windows were 0.615, 0.587, 0.482, 0.467, 0.351, the weather data of the building element walls, columns, beams, doors and windows were 0.733, 0.641, 0.598, 0.473, 0.372, the building inclination data of the building element walls, columns, beams, doors and windows were 0.591, 0.551, 0.471, 0.437, 0.384, the building disaster related data are weighted according to the importance degree to be building vibration data 0.312, geological topography data 0.276, building inclination data 0.205, meteorological data 0.113 and building inclination data 0.094, the indexes of building disaster evaluation grades are damage degree, structural stability, inclination, earthquake resistance, structural material performance and environment adaptation degree, the disaster evaluation of building component walls, columns, beams, doors and windows is 0.8, 0.69, 0.61, 0.46 and 0.38, the standard disaster evaluation grade is 0.8-1, the disaster grade is 0.4-0.79, the disaster grade is 0-0.39, and the building disaster grades of walls, columns, beams, doors and windows are three-grade, two-grade and one-grade respectively.
The foregoing description of the preferred embodiments of the invention is not intended to limit the invention to the precise form disclosed, and any such modifications, equivalents, and alternatives falling within the spirit and scope of the invention are intended to be included within the scope of the invention.
Claims (6)
1. The quick positioning and evaluating method for the building disasters based on machine learning is characterized by comprising the following steps of:
A. obtaining a building disaster image, and preprocessing the building disaster image, wherein the building disaster image comprises positioning information;
B. the method for constructing the convolutional neural network of the building component comprises the steps of constructing the convolutional neural network of the building component, optimizing the convolutional neural network, inputting the preprocessed building disaster image into the optimized convolutional neural network, extracting the preprocessed building disaster image characteristics, classifying the preprocessed building disaster image according to the characteristics to obtain a classified image, and the method for constructing the convolutional neural network of the building component comprises the following steps:
constructing different building component images of a building into a data set, taking AlexNet as a convolutional neural network architecture, adding a convolutional layer, a pooling layer and a full-connection layer, and taking cross entropy as a loss function of the convolutional neural network:
wherein the loss function is C, and the vector corresponding to the kth image is expressed asThe output of the kth Zhang Juanji neural network is +.>;
Taking Adam as an optimizer of the convolutional neural network, inputting a data set into a convolutional neural network model for training, updating the weight of the convolutional neural network by adopting a back propagation algorithm, and learning the characteristics and classification rules of building components;
a method of optimizing the convolutional neural network, comprising:
a. initializing a convolutional neural network model parameter, randomly dividing a training image into a plurality of small-batch training image sets, inputting the small-batch training image sets into a convolutional neural network for training, repeating the operation until the last small-batch training image set, and calculating a cross entropy gradient to adjust the convolutional neural network parameter;
wherein the network parameters areThe iteration number is r, the learning rate is +.>Cross entropy gradient +.>;
b. Inputting the small training image set into the adjusted convolutional neural network, randomly initializing a population, and determining the population scale; obtaining an optimal optimizing path according to the iteration times and the generated random number, and obtaining the size of a convolution kernel;
C. performing image cutting on the classified images to obtain refined images, creating a BIM (building information modeling) according to building components, inputting the refined images into the BIM for association and identification of the building components, and obtaining association content;
D. and integrating the associated content by using a time axis to obtain the change trend of the building component, and evaluating the disaster grade according to the change trend.
2. The method according to claim 1, wherein the preprocessing in step a includes removing noise from the building disaster image by gaussian filtering, converting the building disaster image after denoising into a gray scale image, and enhancing pixel gray values of the gray scale image in a spatial domain to obtain a processed image.
3. The quick positioning and evaluating method for building disaster based on machine learning according to claim 1, wherein the method for obtaining refined images comprises the steps of identifying the preprocessed building disaster images according to the features, classifying the preprocessed building disaster images according to building components corresponding to the features if the preprocessed building disaster images are identified, updating the features until the features are matched if the preprocessed building disaster images are not identified, and repeating the identification until all the preprocessed building disaster images are completely identified, and taking the building disaster images classified by the building components as classified images.
4. The quick localization evaluation method of building disasters based on machine learning according to claim 1, wherein the method for obtaining a refined image by performing image cutting on the classified image comprises the steps of processing the refined image by using Gaussian filtering, calculating gradient strength and direction of each pixel point in the refined image after Gaussian filtering, determining true and potential Gaussian filtering by applying double-threshold detection, processing edges of the refined image to obtain an edge image, and inhibiting isolated weak edge cutting of the edge image to obtain the refined image.
5. The quick positioning and evaluating method for building disasters based on machine learning according to claim 1, wherein the method for obtaining relevant contents comprises the steps of using the BIM model to note component parameters of the refined image, comparing the spatial distribution, basic features and shapes of the features with the refined image, matching the compared results, otherwise traversing the other features until the matched results are output, marking the refined image on the component in the corresponding BIM model according to the matched results, and obtaining relevant contents through association and identification.
6. The machine learning-based rapid localization assessment method for building disasters of claim 1, wherein the method for assessing disaster grades according to the change trend comprises the following steps: collecting data related to building disasters, analyzing the disaster data and the change trend, determining indexes for evaluating disaster grades, and determining evaluation standards according to the indexes and expert experience; and comparing the disaster data with the evaluation standard to determine a disaster grade, wherein the data related to the building disaster comprise meteorological data, geological topography data, building structure data, building inclination data and building vibration data.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202310904906.5A CN116630814B (en) | 2023-07-24 | 2023-07-24 | Quick positioning and evaluating method for building disasters based on machine learning |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202310904906.5A CN116630814B (en) | 2023-07-24 | 2023-07-24 | Quick positioning and evaluating method for building disasters based on machine learning |
Publications (2)
Publication Number | Publication Date |
---|---|
CN116630814A CN116630814A (en) | 2023-08-22 |
CN116630814B true CN116630814B (en) | 2023-09-29 |
Family
ID=87590605
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202310904906.5A Active CN116630814B (en) | 2023-07-24 | 2023-07-24 | Quick positioning and evaluating method for building disasters based on machine learning |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN116630814B (en) |
Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2019095886A (en) * | 2017-11-20 | 2019-06-20 | 株式会社パスコ | Building damage estimation device |
KR20200079708A (en) * | 2018-12-26 | 2020-07-06 | 한국건설기술연구원 | Database construction system of ultra-high speed structural performance for optimal retrofit strategy of structure vulnerable to multi-hazard disaster |
CN114091897A (en) * | 2021-11-19 | 2022-02-25 | 云南电网有限责任公司电力科学研究院 | Neural network-based building damage assessment method after natural disaster |
CN114677522A (en) * | 2022-03-17 | 2022-06-28 | 北京科技大学 | Building structure type prediction method and system based on building pattern spots and deep learning |
CN114792020A (en) * | 2022-04-12 | 2022-07-26 | 云昇昇安全科技(大连)有限责任公司 | Method and system for quickly evaluating building earthquake resistance toughness based on machine learning |
CN115147408A (en) * | 2022-08-30 | 2022-10-04 | 广东省有色工业建筑质量检测站有限公司 | Intelligent detection and identification method for dangerous house |
CN115512247A (en) * | 2022-09-27 | 2022-12-23 | 同济大学 | Regional building damage grade assessment method based on image multi-parameter extraction |
CN116204947A (en) * | 2022-10-11 | 2023-06-02 | 中国人民解放军63921部队 | Frame shear structure collapse area rapid assessment method based on neural network |
-
2023
- 2023-07-24 CN CN202310904906.5A patent/CN116630814B/en active Active
Patent Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2019095886A (en) * | 2017-11-20 | 2019-06-20 | 株式会社パスコ | Building damage estimation device |
KR20200079708A (en) * | 2018-12-26 | 2020-07-06 | 한국건설기술연구원 | Database construction system of ultra-high speed structural performance for optimal retrofit strategy of structure vulnerable to multi-hazard disaster |
CN114091897A (en) * | 2021-11-19 | 2022-02-25 | 云南电网有限责任公司电力科学研究院 | Neural network-based building damage assessment method after natural disaster |
CN114677522A (en) * | 2022-03-17 | 2022-06-28 | 北京科技大学 | Building structure type prediction method and system based on building pattern spots and deep learning |
CN114792020A (en) * | 2022-04-12 | 2022-07-26 | 云昇昇安全科技(大连)有限责任公司 | Method and system for quickly evaluating building earthquake resistance toughness based on machine learning |
CN115147408A (en) * | 2022-08-30 | 2022-10-04 | 广东省有色工业建筑质量检测站有限公司 | Intelligent detection and identification method for dangerous house |
CN115512247A (en) * | 2022-09-27 | 2022-12-23 | 同济大学 | Regional building damage grade assessment method based on image multi-parameter extraction |
CN116204947A (en) * | 2022-10-11 | 2023-06-02 | 中国人民解放军63921部队 | Frame shear structure collapse area rapid assessment method based on neural network |
Non-Patent Citations (2)
Title |
---|
刘照球 ; 万福磊 ; 李云贵 ; .结构灾害评估关键构件信息库框架研究.建筑科学.2012,(09),第79-83页. * |
张令心 ; 钟江荣 ; 林旭川 ; 孙利民 ; 公茂盛 ; 纪晓东 ; 鲍跃全 ; 何浩祥 ; .区域与城市地震风险评估与监测技术研究项目及研究进展.地震科学进展.2020,(03),第1-19页. * |
Also Published As
Publication number | Publication date |
---|---|
CN116630814A (en) | 2023-08-22 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109118479B (en) | Capsule network-based insulator defect identification and positioning device and method | |
CN104091321B (en) | It is applicable to the extracting method of the multi-level point set feature of ground laser radar point cloud classifications | |
CN115063409B (en) | Method and system for detecting surface material of mechanical cutter | |
CN103020649A (en) | Forest type identification method based on texture information | |
CN108171119B (en) | SAR image change detection method based on residual error network | |
CN111914902B (en) | Traditional Chinese medicine identification and surface defect detection method based on deep neural network | |
CN113919226B (en) | Mining vegetation ecological cumulative effect disturbance range identification method based on weight | |
CN113989257A (en) | Electric power comprehensive pipe gallery settlement crack identification method based on artificial intelligence technology | |
CN116206185A (en) | Lightweight small target detection method based on improved YOLOv7 | |
CN113920255B (en) | High-efficient mapping system based on point cloud data | |
CN109859199B (en) | Method for detecting quality of freshwater seedless pearls through SD-OCT image | |
Vatsavai et al. | Probabilistic change detection framework for analyzing settlement dynamics using very high-resolution satellite imagery | |
CN1252588C (en) | High spectrum remote sensing image combined weighting random sorting method | |
CN112348750A (en) | SAR image change detection method based on threshold fusion and neighborhood voting | |
CN110349119B (en) | Pavement disease detection method and device based on edge detection neural network | |
CN116630814B (en) | Quick positioning and evaluating method for building disasters based on machine learning | |
Paget et al. | Determining the floe-size distribution of East Antarctic sea ice from digital aerial photographs | |
Wijaya et al. | Building crack due to lombok earthquake classification based on glcm features and svm classifier | |
CN111507423B (en) | Engineering quantity measuring method for cleaning transmission line channel | |
CN114419443A (en) | Automatic remote-sensing image cultivated land block extraction method and system | |
CN116129280B (en) | Method for detecting snow in remote sensing image | |
CN117409329B (en) | Method and system for reducing false alarm rate of underground cavity detection by three-dimensional ground penetrating radar | |
CN109726724B (en) | Water gauge image feature weighted learning identification method under shielding condition | |
CN113392924B (en) | Identification method of acoustic-electric imaging log and related equipment | |
CN117765395A (en) | Road extraction method and system for generalizing semantic segmentation algorithm |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |