WO2020199538A1 - Bridge key component disease early-warning system and method based on image monitoring data - Google Patents

Bridge key component disease early-warning system and method based on image monitoring data Download PDF

Info

Publication number
WO2020199538A1
WO2020199538A1 PCT/CN2019/108089 CN2019108089W WO2020199538A1 WO 2020199538 A1 WO2020199538 A1 WO 2020199538A1 CN 2019108089 W CN2019108089 W CN 2019108089W WO 2020199538 A1 WO2020199538 A1 WO 2020199538A1
Authority
WO
WIPO (PCT)
Prior art keywords
disease
image
bridge
early warning
data
Prior art date
Application number
PCT/CN2019/108089
Other languages
French (fr)
Chinese (zh)
Inventor
明图章
陆永泉
闫志刚
胡靖�
丁军华
蒋龙泉
张贵忠
成礼平
李西芝
李波
Original Assignee
中设设计集团股份有限公司
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by 中设设计集团股份有限公司 filed Critical 中设设计集团股份有限公司
Publication of WO2020199538A1 publication Critical patent/WO2020199538A1/en

Links

Images

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
    • G01N21/84—Systems specially adapted for particular applications
    • G01N21/88—Investigating the presence of flaws or contamination
    • G01N21/8851—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
    • G01N21/84—Systems specially adapted for particular applications
    • G01N21/88—Investigating the presence of flaws or contamination
    • G01N21/95—Investigating the presence of flaws or contamination characterised by the material or shape of the object to be examined
    • 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
    • G06N3/045—Combinations of networks
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T5/00—Image enhancement or restoration
    • G06T5/70—Denoising; Smoothing
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T5/00—Image enhancement or restoration
    • G06T5/77—Retouching; Inpainting; Scratch removal
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/0002—Inspection of images, e.g. flaw detection
    • G06T7/0004—Industrial image inspection
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/60—Analysis of geometric attributes
    • G06T7/62—Analysis of geometric attributes of area, perimeter, diameter or volume
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N7/00—Television systems
    • H04N7/18—Closed-circuit television [CCTV] systems, i.e. systems in which the video signal is not broadcast
    • H04N7/183—Closed-circuit television [CCTV] systems, i.e. systems in which the video signal is not broadcast for receiving images from a single remote source
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
    • G01N21/84—Systems specially adapted for particular applications
    • G01N21/88—Investigating the presence of flaws or contamination
    • G01N21/8851—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges
    • G01N2021/8854—Grading and classifying of flaws
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
    • G01N21/84—Systems specially adapted for particular applications
    • G01N21/88—Investigating the presence of flaws or contamination
    • G01N21/8851—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges
    • G01N2021/8854—Grading and classifying of flaws
    • G01N2021/8874—Taking dimensions of defect into account
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
    • G01N21/84—Systems specially adapted for particular applications
    • G01N21/88—Investigating the presence of flaws or contamination
    • G01N21/8851—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges
    • G01N2021/8887—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges based on image processing techniques
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/20—Special algorithmic details
    • G06T2207/20081—Training; Learning
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/30—Subject of image; Context of image processing
    • G06T2207/30108—Industrial image inspection
    • G06T2207/30132—Masonry; Concrete

Definitions

  • the invention belongs to the technical field of bridge structure disease monitoring, and specifically relates to an early warning system and method for bridge critical component disease based on image monitoring data.
  • the operating quality of the bridge structure is very important to ensure the normal operation of the road network and driving safety. Therefore, it is necessary to carry out the quality monitoring of the bridge structure, especially the key components of the force.
  • Modern bridges have different bridge types and complex forces, and there are many key parts, such as supports, connecting sections, cable connections, etc., and mastering their quality conditions is the basis for effective maintenance plans.
  • the traditional manual inspection method has the problem of strong subjective judgment and low efficiency, and there is also a certain risk to the personal safety of the inspectors.
  • the technology of using sensors to monitor key components of bridges has been widely used, but its collection speed is slow, the error is large, and the layout is complicated. Once it fails, it is difficult to repair.
  • Convolutional neural network is an effective technology for computer machine vision recognition. At this stage, it is mainly used in the detection of tunnel diseases and has achieved good detection results. However, in the monitoring of key components of bridges, machine vision technology considering deep learning has not been effectively applied, so it is necessary to establish a monitoring system based on deep learning machine vision technology in bridge structures.
  • the present invention proposes a bridge critical component disease early warning system and method based on image monitoring data.
  • image analysis and evaluation early warning system combined with depth Learn the machine vision technology to realize the effective identification and classification of different diseases, and carry out the assessment and early warning of the disease development rate and severity according to the image processing technology.
  • an early warning system for bridge critical component diseases based on image monitoring data which includes an image acquisition and data transmission system, an image processing system and an evaluation early warning system;
  • the image acquisition and data transmission system consists of high-definition cameras, solar components, auxiliary light sources, wireless transmission networks and data receiving terminals.
  • High-definition cameras are set up on key components of the bridge to collect and transmit image data to the data receiving terminal at a certain frequency; image processing
  • the system is used for automatic reading, optimization, disease identification and classification of high-definition images; the evaluation and early warning system takes the binary images of the disease as the research object, and if the disease development rate and severity exceed the limit threshold, it will give an early warning.
  • An early warning method for bridge critical component diseases based on image monitoring data including the following steps:
  • the present invention establishes an early warning system and method for bridge critical component diseases based on image monitoring data, and provides effective monitoring technology for the quality status of the bridge structure during operation.
  • Real-time monitoring images are acquired through high-definition cameras installed at key components.
  • the high-definition images acquired at a certain acquisition frequency can monitor the quality of the structure and obtain a comprehensive and effective overall bridge structure in time.
  • Fig. 1 is a schematic block diagram of an early warning system for critical components of a bridge based on image monitoring data.
  • Figure 2 is the original image of the disease in the embodiment.
  • Figure 3 is a disease binarization diagram in the embodiment.
  • an early warning system for bridge critical component diseases based on image monitoring data which includes an image acquisition and data transmission system, an image processing system, and an evaluation and early warning system;
  • the image acquisition and data transmission system is composed of a high-definition camera, solar components, auxiliary light sources, a wireless transmission network, and a data receiving end.
  • the high-definition cameras are installed at each key component of the bridge, and image data is collected and transmitted to the data receiving end at a certain frequency.
  • the image processing system includes the automatic reading, optimization, disease identification and classification of high-definition images to realize the machine vision analysis of the operational quality of the monitoring components; the evaluation and early warning system takes the binary image of the disease as the research object to carry out the development of possible diseases Tracking and status evaluation, if the disease development rate and severity exceed the limit threshold, an early warning or warning will be issued.
  • the high-definition camera is fixed on the key component of the bridge and takes pictures.
  • the solar module is used to power the high-definition camera and auxiliary light source.
  • the auxiliary light source is used to increase the brightness when the optical fiber is dark, and adjust the shooting angle and distance to cover the range of the component.
  • the captured image is a 256-color RGB image, the minimum resolution is 2432 ⁇ 2048, and the acquisition frequency can be adjusted between 1 min and 1 d according to the actual situation;
  • the image acquisition and data transmission system adopts a 4G wireless transmission module to transmit the original captured image and store it in the data receiving terminal according to the shooting number and shooting time.
  • the key components of the bridge include bridge joints, cable ends, supports, box girder structures and piers.
  • the image processing system is used for automatic image reading, image optimization, component disease identification and classification.
  • Use MATLAB software to read the 256-color RGB image of the key structural parts of the bridge according to the number and time sequence, and use the built-in function imadjust and discrete cosine transform to enhance the brightness, improve the contrast and remove the noise of the image data, and then input the image data into the volume
  • Product neural network machine vision model if the image recognition is disease-free, the output flag 0, the crack-type disease output flag 1, the pothole disease output flag 2, and the binary image of the disease image are output.
  • the convolutional neural network machine vision model can be improved based on the VGGNet model as the kernel basis.
  • the "Tensorflow+Python" system framework is used to build a full convolutional neural network model.
  • the model is composed of 5 layers of convolutional layers and softmax output layers.
  • the 3-layer fully connected layer in the original VGGNet is replaced with a convolutional layer, and the layers are separated by max-pooling and mean-pooling, and the activation function is the Maxout function
  • the convolutional neural network machine vision model uses the disease images of the existing bridge components for training.
  • the typical disease of each type of component selects 200 to 500 original disease images, and uses the PHOTOSHOP software to manually mark and classify the diseases in the images.
  • the original image is used as the input data of the model, and the corresponding disease-labeled image is used as the output data of the model.
  • the correct recognition rate of the model needs to reach more than 95%.
  • the evaluation and early warning system analyzes the binary image of the disease, uses MATLAB's own functions Bwmorph and Minboundrect to determine the length and distribution area of the crack disease; uses direct statistics of the number of 1-value pixels of the binary disease image to determine the pothole Disease area
  • the present invention also provides an early warning method for bridge critical component diseases based on the above system, which includes the following steps:
  • the development of the disease that may occur is tracked and the condition is evaluated. If the component disease degree exceeds the limit threshold, an early warning or warning will be given.
  • an image acquisition and data transmission system As shown in Figure 1, according to the functional modules of the system of the present invention, an image acquisition and data transmission system, an image processing system, and an evaluation and early warning system are respectively set up.
  • First set up an image acquisition and data transmission system Set up fixed brackets to install high-definition cameras at stress-sensitive places such as bridge joints, cable ends, supports, box girder structures and piers, and adjust the lens angle and focal length to make the target area clear and appear.
  • the resolution of the selected high-definition camera needs to be above 2432 ⁇ 2048, which can identify 0.5mm cracks;
  • the detection equipment of the force-bearing complex components collects images at a frequency of 30 minutes, and transmits them to the image storage device through the signal transmitting device and the gigabit wireless network.
  • the monitoring images of the key structural parts of the bridge are stored according to the number. Extract the first image of each high-definition camera as the initial image that characterizes the condition of the component, and record the corresponding time and position number, as shown in Figure 2.
  • the TensorFlow program framework to build a convolutional neural network machine vision model, select the VGGNet model as the kernel template for training, and build a recognition and classification system for diseases of different key structural parts of the bridge.
  • the serial number the 256-color RGB monitoring image is used as the input data of the model, and the corresponding disease label map is output after image recognition.
  • Bwmorph extracts the skeleton polyline composed of a single pixel of the crack, counts the number of pixels contained in the polyline and multiplies it by the pixel size to obtain the crack length;
  • Minboundrect function Establish the smallest rectangle surrounding the crack, and measure the influence range of the crack by the area of the rectangle.
  • pothole disease count the number of pixels with a value of 1 in the binarized image, and multiply it by the area of the individual pixel to get the pothole area.

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • General Health & Medical Sciences (AREA)
  • Health & Medical Sciences (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Immunology (AREA)
  • Signal Processing (AREA)
  • Pathology (AREA)
  • Biochemistry (AREA)
  • Analytical Chemistry (AREA)
  • Chemical & Material Sciences (AREA)
  • Molecular Biology (AREA)
  • Data Mining & Analysis (AREA)
  • Software Systems (AREA)
  • Biophysics (AREA)
  • General Engineering & Computer Science (AREA)
  • Computing Systems (AREA)
  • Mathematical Physics (AREA)
  • Biomedical Technology (AREA)
  • Computational Linguistics (AREA)
  • Multimedia (AREA)
  • Evolutionary Computation (AREA)
  • Quality & Reliability (AREA)
  • Artificial Intelligence (AREA)
  • Geometry (AREA)
  • Image Analysis (AREA)

Abstract

A bridge key component disease early-warning system and method based on image monitoring data. The system comprises an image collection and data transmission system, an image processing system and an assessment early-warning system. The image collection and data transmission system consists of high-definition cameras, a solar module, an auxiliary light source, a wireless transmission network and a data receiving end; the high-definition cameras are provided on key structure parts of a bridge, and image data is collected according to a certain frequency and transmitted to the data receiving end. The image processing system has the functions of automatic reading and optimization of a high-definition image and disease recognition and classification, and achieves machine vision analysis on monitoring component operation quality. The assessment early-warning system performs tracking and condition evaluation on the development of a possible disease by using a disease binary image as a research object, and if the disease development rate and severity exceed a threshold limit value, an early-warning is given. Automatic monitoring of bridge key components is achieved, and the early-warning system and method are of great significance in ensuring the safety of the structure.

Description

基于图像监控数据的桥梁关键构件病害预警系统及方法Early warning system and method for bridge key component diseases based on image monitoring data 技术领域Technical field
本发明属于桥梁结构病害监测技术领域，具体涉及一种基于图像监控数据的桥梁关键构件病害预警系统及方法。The invention belongs to the technical field of bridge structure disease monitoring, and specifically relates to an early warning system and method for bridge critical component disease based on image monitoring data.
背景技术Background technique
桥梁结构的运营质量对保障路网正常运行与行车安全至关重要，因此，开展桥梁结构，特别是受力关键构件的质量监测十分必要。现代桥梁桥型各异、受力复杂，存在众多关键部位，如支座、连接段、缆索连接处等，掌握其质量状况是有效制定维养方案的基础。对于大型桥梁结构，传统的人工巡检方式存在主观判断性强且效率较低的问题，并且对巡检人员的人身安全也存在一定的危险性。利用传感器监测桥梁关键构件的技术已经得到广泛应用，但其采集速度慢、误差较大且布置复杂，一旦失效难以维修。此外，传统图像监控技术以人工观测、识别为主，近年来计算机图像处理与病害识别技术虽逐渐得到应用，但由于图像质量低、背景干扰物复杂、识别算法有效性较差等因素，导致监控图像的自动化处理技术未能在桥梁结构的监测中得以广泛应用。The operating quality of the bridge structure is very important to ensure the normal operation of the road network and driving safety. Therefore, it is necessary to carry out the quality monitoring of the bridge structure, especially the key components of the force. Modern bridges have different bridge types and complex forces, and there are many key parts, such as supports, connecting sections, cable connections, etc., and mastering their quality conditions is the basis for effective maintenance plans. For large bridge structures, the traditional manual inspection method has the problem of strong subjective judgment and low efficiency, and there is also a certain risk to the personal safety of the inspectors. The technology of using sensors to monitor key components of bridges has been widely used, but its collection speed is slow, the error is large, and the layout is complicated. Once it fails, it is difficult to repair. In addition, traditional image monitoring technology is mainly based on manual observation and recognition. In recent years, computer image processing and disease recognition technology have gradually been applied. However, due to factors such as low image quality, complex background interference, and poor recognition algorithm effectiveness, the monitoring Automatic image processing technology has not been widely used in the monitoring of bridge structures.
随着深度学习与人工智能在工程领域的逐步普及，工程结构病害的图像识别精度得到了大幅度提高。卷积神经网络是计算机机器视觉识别的有效技术，现阶段主要应用于隧道病害的检测，取得了较好的检测效果。然而，在桥梁关键构件的监测工作中，考虑深度学习的机器视觉技术尚未得到有效应用，因此有必要建立基于深度学习机器视觉技术在桥梁结构中的监测系统。With the gradual popularity of deep learning and artificial intelligence in the engineering field, the accuracy of image recognition of engineering structural diseases has been greatly improved. Convolutional neural network is an effective technology for computer machine vision recognition. At this stage, it is mainly used in the detection of tunnel diseases and has achieved good detection results. However, in the monitoring of key components of bridges, machine vision technology considering deep learning has not been effectively applied, so it is necessary to establish a monitoring system based on deep learning machine vision technology in bridge structures.
发明内容Summary of the invention
发明目的：针对以上现状与存在的问题，本发明提出一种基于图像监控数据的桥梁关键构件病害预警系统及方法，通过建立桥梁关键构件的图像采集与传输、图像分析与评估预警系统，结合深度学习机器视觉技术实现对不同病害的有效识别与分类，并根据图像处理技术开展病害发展速率与严重程度的评估和预警。Purpose of the invention: In view of the above current situation and existing problems, the present invention proposes a bridge critical component disease early warning system and method based on image monitoring data. By establishing a bridge critical component image collection and transmission, image analysis and evaluation early warning system, combined with depth Learn the machine vision technology to realize the effective identification and classification of different diseases, and carry out the assessment and early warning of the disease development rate and severity according to the image processing technology.
实现本发明目的的技术解决方案为：一种基于图像监控数据的桥梁关键构件病害预警系统，该系统包括图像采集与数据传输系统、图像处理系统及评估预警系统；The technical solution to achieve the objective of the present invention is: an early warning system for bridge critical component diseases based on image monitoring data, which includes an image acquisition and data transmission system, an image processing system and an evaluation early warning system;
图像采集与数据传输系统由高清相机、太阳能组件、辅助光源、无线传输网络及数据接收端组成，在桥梁各关键构件设置高清相机，按照一定频率进行图像数据采集并传输至数据接收端；图像处理系统用于高清图像的自动化读取、优化、病害识别与分类； 评估预警系统以病害二值化图像为研究对象，若病害发展速率与严重程度超过极限阈值即进行预警。The image acquisition and data transmission system consists of high-definition cameras, solar components, auxiliary light sources, wireless transmission networks and data receiving terminals. High-definition cameras are set up on key components of the bridge to collect and transmit image data to the data receiving terminal at a certain frequency; image processing The system is used for automatic reading, optimization, disease identification and classification of high-definition images; the evaluation and early warning system takes the binary images of the disease as the research object, and if the disease development rate and severity exceed the limit threshold, it will give an early warning.
一种基于图像监控数据的桥梁关键构件病害预警方法，包括以下步骤：An early warning method for bridge critical component diseases based on image monitoring data, including the following steps:
在桥梁各关键构件处设置高清相机，进行图像数据采集并传输至数据接收端；Set up high-definition cameras at each key component of the bridge to collect and transmit image data to the data receiving end;
对采集的图像进行自动化读取、优化、病害识别与分类；Automatically read, optimize, identify and classify the collected images;
以构件的病害二值化图像为研究对象，若构件病害程度超过极限阈值即进行预警。Taking the component disease binarization image as the research object, if the component disease degree exceeds the limit threshold, an early warning is given.
与现有技术相比，本发明的技术方案具有以下有益技术效果：Compared with the prior art, the technical solution of the present invention has the following beneficial technical effects:
(1)本发明建立了一种基于图像监控数据的桥梁关键构件病害预警系统及方法，针对桥梁结构在运营过程中的质量状况提供了有效监测技术。(1) The present invention establishes an early warning system and method for bridge critical component diseases based on image monitoring data, and provides effective monitoring technology for the quality status of the bridge structure during operation.
(2)通过设置于关键构件处的高清相机获取实时监测图像，按照一定采集频率所获取的高清图像可对结构质量进行监控，及时获取全面、有效的桥梁整体结构工作状况。(2) Real-time monitoring images are acquired through high-definition cameras installed at key components. The high-definition images acquired at a certain acquisition frequency can monitor the quality of the structure and obtain a comprehensive and effective overall bridge structure in time.
(3)采用深度学习机器视觉进行病害识别与分类，可较好地对复杂背景与干扰物下的病害图像进行识别，并按照类型实现分类处理；根据不同病害特征，采用相应的评估模式分析病害的发展趋势与严重程度，根据预定阈值开展预警。(3) The use of deep learning machine vision for disease recognition and classification can better recognize disease images under complex backgrounds and interferences, and realize classification processing according to types; according to different disease characteristics, use corresponding evaluation modes to analyze disease Early warning is carried out according to the predetermined threshold value of the development trend and severity.
附图说明Description of the drawings
图1为本发明基于图像监控数据的桥梁关键构件病害预警系统原理框图。Fig. 1 is a schematic block diagram of an early warning system for critical components of a bridge based on image monitoring data.
图2为实施例中的病害原始图像。Figure 2 is the original image of the disease in the embodiment.
图3为实施例中的病害二值化图。Figure 3 is a disease binarization diagram in the embodiment.
具体实施方式detailed description
结合图1，一种基于图像监控数据的桥梁关键构件病害预警系统，该系统包含图像采集与数据传输系统、图像处理系统及评估预警系统；Combined with Figure 1, an early warning system for bridge critical component diseases based on image monitoring data, which includes an image acquisition and data transmission system, an image processing system, and an evaluation and early warning system;
所述图像采集与数据传输系统由高清相机、太阳能组件、辅助光源、无线传输网络及数据接收端组成，在桥梁各关键构件处设置高清相机，按照一定频率进行图像数据采集并传输至数据接收端；图像处理系统包括高清图像的自动化读取、优化、病害识别与分类，实现对监测构件运营质量的机器视觉分析；评估预警系统以病害二值化图像为研究对象，对可能发生病害的发展进行跟踪与状况评价，若病害发展速率与严重程度超过极限阈值即进行预警或发出警告。The image acquisition and data transmission system is composed of a high-definition camera, solar components, auxiliary light sources, a wireless transmission network, and a data receiving end. The high-definition cameras are installed at each key component of the bridge, and image data is collected and transmitted to the data receiving end at a certain frequency. The image processing system includes the automatic reading, optimization, disease identification and classification of high-definition images to realize the machine vision analysis of the operational quality of the monitoring components; the evaluation and early warning system takes the binary image of the disease as the research object to carry out the development of possible diseases Tracking and status evaluation, if the disease development rate and severity exceed the limit threshold, an early warning or warning will be issued.
进一步的，所述高清相机固定于桥梁关键构件部位并进行拍摄，太阳能组件用于为高清相机和辅助光源供电，辅助光源用于在光纤较暗时增强亮度，调节拍摄角度与距离 以覆盖构件范围，并通过调整焦距使其识别精度达到0.5mm以上，拍摄图像为256色RGB图像，最小分辨率为2432×2048，采集频率可根据实际情况在1min～1d之间调节；Further, the high-definition camera is fixed on the key component of the bridge and takes pictures. The solar module is used to power the high-definition camera and auxiliary light source. The auxiliary light source is used to increase the brightness when the optical fiber is dark, and adjust the shooting angle and distance to cover the range of the component. , And by adjusting the focal length to make the recognition accuracy reach 0.5mm or more, the captured image is a 256-color RGB image, the minimum resolution is 2432×2048, and the acquisition frequency can be adjusted between 1 min and 1 d according to the actual situation;
所述图像采集与数据传输系统采用4G无线传输模块，将原始拍摄图像传输并按照拍摄编号、拍摄时间存储于数据接收终端。The image acquisition and data transmission system adopts a 4G wireless transmission module to transmit the original captured image and store it in the data receiving terminal according to the shooting number and shooting time.
桥梁关键构件包括桥梁接缝、缆索端部、支座、箱梁结构与桥墩。The key components of the bridge include bridge joints, cable ends, supports, box girder structures and piers.
进一步的，所述图像处理系统用于自动化图像读取、图像优化、构件病害识别与分类。利用MATLAB软件按照编号与时间顺序读取桥梁关键结构部位的256色RGB图像，采用自带函数imadjust与离散余弦变换对图像数据进行增强亮度、提高对比度与去除噪点处理，其后将图像数据输入卷积神经网络机器视觉模型，若图像识别无病害则输出标识0、裂缝类病害输出标识1、坑洞病害输出标识2，并输出病害图像的二值化图像。Further, the image processing system is used for automatic image reading, image optimization, component disease identification and classification. Use MATLAB software to read the 256-color RGB image of the key structural parts of the bridge according to the number and time sequence, and use the built-in function imadjust and discrete cosine transform to enhance the brightness, improve the contrast and remove the noise of the image data, and then input the image data into the volume Product neural network machine vision model, if the image recognition is disease-free, the output flag 0, the crack-type disease output flag 1, the pothole disease output flag 2, and the binary image of the disease image are output.
所述卷积神经网络机器视觉模型可基于VGGNet模型为内核基础进行改进，采用“Tensorflow+Python”系统框架构建全卷积神经网络模型，模型采用5层卷积层、softmax输出层构成，并将原VGGNet中的3层全连接层替换为卷积层，层与层之间采用max-pooling与mean-pooling分开，激活函数为Maxout函数The convolutional neural network machine vision model can be improved based on the VGGNet model as the kernel basis. The "Tensorflow+Python" system framework is used to build a full convolutional neural network model. The model is composed of 5 layers of convolutional layers and softmax output layers. The 3-layer fully connected layer in the original VGGNet is replaced with a convolutional layer, and the layers are separated by max-pooling and mean-pooling, and the activation function is the Maxout function
卷积神经网络机器视觉模型采用已有桥梁构件的病害图像进行训练，每类构件的典型病害选取200～500张病害原始图像，并采用PHOTOSHOP软件将图像中的病害进行人工标记与分类，以病害原始图像作为模型输入数据，对应的病害标记图像作为模型输出数据，模型正确识别率需达到95％以上。The convolutional neural network machine vision model uses the disease images of the existing bridge components for training. The typical disease of each type of component selects 200 to 500 original disease images, and uses the PHOTOSHOP software to manually mark and classify the diseases in the images. The original image is used as the input data of the model, and the corresponding disease-labeled image is used as the output data of the model. The correct recognition rate of the model needs to reach more than 95%.
进一步的，所述评估预警系统对病害二值化图像进行分析，采用MATLAB自带函数Bwmorph与Minboundrect确定裂缝病害的长度与分布面积；采用直接统计二值化病害图像的1值像素数量确定坑洞病害的面积；Further, the evaluation and early warning system analyzes the binary image of the disease, uses MATLAB's own functions Bwmorph and Minboundrect to determine the length and distribution area of the crack disease; uses direct statistics of the number of 1-value pixels of the binary disease image to determine the pothole Disease area
对比同一构件不同时间的病害图像数据，针对裂缝病害，逐步对比裂缝长度与分布面积的变化趋势；对于坑洞病害，逐步对比面积的变化趋势。若病害的变化趋势增长速率超过规定阈值则进行预警；若裂缝病害的长度、或裂缝与坑洞病害的面积超过规定最大值则发出警报。Compare the disease image data of the same component at different times, and gradually compare the change trend of the crack length and the distribution area for the crack disease; for the pothole disease, gradually compare the change trend of the area. If the growth rate of the change trend of the disease exceeds the prescribed threshold, an early warning is given; if the length of the crack disease or the area of the crack and the pothole disease exceeds the prescribed maximum value, an alarm is issued.
本发明还提供一种基于上述系统的桥梁关键构件病害预警方法，包括以下步骤：The present invention also provides an early warning method for bridge critical component diseases based on the above system, which includes the following steps:
在桥梁各关键构件位置设置高清相机，进行图像数据采集并及时传输至数据接收端；Set up high-definition cameras at the locations of key components of the bridge to collect image data and transmit them to the data receiving end in time;
对采集的图像进行自动化读取、优化、病害识别与分类；Automatically read, optimize, identify and classify the collected images;
以构件的病害二值化图像为研究对象，对可能发生病害的发展进行跟踪与状况评价， 若构件病害程度超过极限阈值即进行预警或发出警告。Taking the component disease binarization image as the research object, the development of the disease that may occur is tracked and the condition is evaluated. If the component disease degree exceeds the limit threshold, an early warning or warning will be given.
下面结合具体实施例对本发明的目的作进一步详细地描述，该实例仅用于更加清楚地说明本发明的技术方案，而不能以此来限制本发明的保护范围。The purpose of the present invention will be described in further detail below in conjunction with specific embodiments. This example is only used to illustrate the technical solution of the present invention more clearly, and cannot be used to limit the protection scope of the present invention.
实施例Example
如图1所示，根据本发明系统的功能模块，分别设置图像采集与数据传输系统、图像处理系统及评估预警系统。As shown in Figure 1, according to the functional modules of the system of the present invention, an image acquisition and data transmission system, an image processing system, and an evaluation and early warning system are respectively set up.
首先设置图像采集与数据传输系统，在桥梁接缝、缆索端部、支座、箱梁结构与桥墩等受力敏感处设置固定支架安装高清相机，调整镜头角度与焦距使得目标区域清晰并均出现在拍摄范围内，选定高清相机的分辨率需达到2432×2048以上，可识别0.5mm的裂缝；First set up an image acquisition and data transmission system. Set up fixed brackets to install high-definition cameras at stress-sensitive places such as bridge joints, cable ends, supports, box girder structures and piers, and adjust the lens angle and focal length to make the target area clear and appear. Within the shooting range, the resolution of the selected high-definition camera needs to be above 2432×2048, which can identify 0.5mm cracks;
在各高清相机范围内进行尺寸标定，采用实际构件上两标定点的距离除以图像中对于两点的距离，计算得到图像的像素尺寸。Perform size calibration within the range of each high-definition camera, divide the distance between two calibration points on the actual component by the distance between the two points in the image, and calculate the pixel size of the image.
安装太阳能组件与辅助光源设备，并与高清相机连接以提供电源，降低安装与后续维养难度。Install solar modules and auxiliary light source equipment, and connect with high-definition cameras to provide power, reducing the difficulty of installation and subsequent maintenance.
受力复杂构件的检测设备按照30min的频率进行图像采集，并通过信号发射装置与千兆无线网传输至图像存储设备中，各桥梁关键结构部位的监测图像按照编号存储。提取各高清相机的第一幅图像作为表征构件状况的初始图像，并记录相应的时间与位置编号，如图2所示。The detection equipment of the force-bearing complex components collects images at a frequency of 30 minutes, and transmits them to the image storage device through the signal transmitting device and the gigabit wireless network. The monitoring images of the key structural parts of the bridge are stored according to the number. Extract the first image of each high-definition camera as the initial image that characterizes the condition of the component, and record the corresponding time and position number, as shown in Figure 2.
利用TensorFlow程序框架搭建卷积神经网络机器视觉模型，选用VGGNet模型作为内核模板训练并搭建不同桥梁关键结构部位病害的识别与分类系统。按照编号顺序依次将256色RGB监测图像作为模型的输入端数据，进行图像识别后输出相应的病害的标签图。Use the TensorFlow program framework to build a convolutional neural network machine vision model, select the VGGNet model as the kernel template for training, and build a recognition and classification system for diseases of different key structural parts of the bridge. According to the serial number, the 256-color RGB monitoring image is used as the input data of the model, and the corresponding disease label map is output after image recognition.
采用MATLAB程序im2bw函数对病害标签图进行二值化处理，如图3所示，根据标签数据自动选用相关进行分析。针对裂缝病害，采用Bwmorph与Minboundrect函数计算裂缝病害的长度与分布面积：Bwmorph提取裂缝单像素组成的骨架折线，统计该折线所包含的像素数并乘以像素尺寸后即可得到裂缝长度；Minboundrect函数建立包围裂缝的最小矩形，以矩形面积衡量裂缝的影响范围。针对坑洞病害，统计二值化图像中数值为1的像素数量，乘以单独像素面积即可得到坑洞面积。Use the MATLAB program im2bw function to binarize the disease label map, as shown in Figure 3, automatically select correlations for analysis based on the label data. For crack diseases, use the Bwmorph and Minboundrect functions to calculate the length and distribution area of the crack diseases: Bwmorph extracts the skeleton polyline composed of a single pixel of the crack, counts the number of pixels contained in the polyline and multiplies it by the pixel size to obtain the crack length; Minboundrect function Establish the smallest rectangle surrounding the crack, and measure the influence range of the crack by the area of the rectangle. For the pothole disease, count the number of pixels with a value of 1 in the binarized image, and multiply it by the area of the individual pixel to get the pothole area.
对比构件病害在不同时段的二值化图像数据，对于裂缝病害，若裂缝长度增加率超 过5mm/d、分布面积超过10cm 2/d即进行发展趋势预警；对于坑洞病害，若面积发展速率超过1cm 2/d即进行发展趋势预警。 Comparing the binarized image data of component diseases in different periods, for crack diseases, if the crack length increase rate exceeds 5mm/d and the distribution area exceeds 10cm 2 /d, the development trend warning will be given; for pothole diseases, if the area development rate exceeds 1cm 2 /d is the early warning of development trend.
计算构件病害的二值化图像数据，对于裂缝病害，若裂缝长度达到0.2m则进行裂缝严重程度警告；对于坑洞病害，若面积达到0.01m 2则进行坑洞严重程度警告。 Calculate the binary image data of component diseases. For crack diseases, if the crack length reaches 0.2m, the crack severity warning will be issued; for pothole diseases, if the area reaches 0.01m 2 , the pothole severity warning will be issued.

Claims (10)

  1. 一种基于图像监控数据的桥梁关键构件病害预警系统，其特征在于：该系统包括图像采集与数据传输系统、图像处理系统及评估预警系统；An early warning system for bridge critical component diseases based on image monitoring data is characterized in that: the system includes an image acquisition and data transmission system, an image processing system, and an evaluation and early warning system;
    图像采集与数据传输系统由高清相机、太阳能组件、辅助光源、无线传输网络及数据接收端组成，在桥梁各关键构件设置高清相机，按照一定频率进行图像数据采集并传输至数据接收端；图像处理系统用于高清图像的自动化读取、优化、病害识别与分类；评估预警系统以病害二值化图像为研究对象，若病害发展速率与严重程度超过极限阈值即进行预警。The image acquisition and data transmission system consists of high-definition cameras, solar components, auxiliary light sources, wireless transmission networks and data receiving terminals. High-definition cameras are set up on key components of the bridge to collect and transmit image data to the data receiving terminal at a certain frequency; image processing The system is used for automatic reading, optimization, disease identification and classification of high-definition images; the evaluation and early warning system uses disease binary images as the research object, and if the disease development rate and severity exceed the limit threshold, it will give early warning.
  2. 根据权利要求1所述的基于图像监控数据的桥梁关键构件病害预警系统，其特征在于：高清相机固定于桥梁关键构件并进行拍摄，调节拍摄角度与距离以覆盖构件范围，并通过调整焦距使其识别精度达到0.5mm以上，拍摄图像为256色RGB图像，最小分辨率为2432×2048，采集频率在1min～1d之间调节；The bridge critical component disease early warning system based on image monitoring data according to claim 1, characterized in that: the high-definition camera is fixed to the bridge critical component and shoots, adjusts the shooting angle and distance to cover the range of the component, and adjusts the focal length to make it The recognition accuracy is above 0.5mm, the captured image is a 256-color RGB image, the minimum resolution is 2432×2048, and the acquisition frequency is adjusted between 1min and 1d;
    所述图像采集与数据传输系统采用4G无线传输模块，将原始拍摄图像传输并按照拍摄编号、拍摄时间存储于数据接收终端。The image acquisition and data transmission system adopts a 4G wireless transmission module to transmit the original captured image and store it in the data receiving terminal according to the shooting number and shooting time.
  3. 根据权利要求1或2所述的基于图像监控数据的桥梁关键构件病害预警系统，其特征在于：桥梁关键构件包括桥梁接缝、缆索端部、支座、箱梁结构与桥墩。The bridge critical component disease early warning system based on image monitoring data according to claim 1 or 2, characterized in that: the bridge critical components include bridge joints, cable ends, supports, box girder structures and bridge piers.
  4. 根据权利要求1所述的基于图像监控数据的桥梁关键构件病害预警系统，其特征在于，所述图像处理系统用于自动化图像读取、图像优化、构件病害识别与分类，具体方法为：利用MATLAB软件按照编号与时间顺序读取桥梁关键构件的256色RGB图像，采用自带函数imadjust与离散余弦变换对图像数据进行增强亮度、提高对比度与去除噪点处理，其后将图像数据输入卷积神经网络机器视觉模型，若图像识别无病害则输出标识0，裂缝类病害输出标识1，坑洞病害输出标识2，并输出病害图像的二值化图像。The bridge critical component disease early warning system based on image monitoring data according to claim 1, wherein the image processing system is used for automatic image reading, image optimization, component disease identification and classification, and the specific method is: using MATLAB The software reads the 256-color RGB image of the key components of the bridge according to the serial number and time sequence, and uses the built-in function imadjust and discrete cosine transform to enhance the brightness, improve the contrast and remove the noise of the image data, and then input the image data into the convolutional neural network In the machine vision model, if the image recognition is disease-free, it will output flag 0, crack type disease output flag 1, pothole disease output flag 2, and output the binary image of the disease image.
  5. 根据权利要求4所述的基于图像监控数据的桥梁关键构件病害预警系统，其特征在于：所述卷积神经网络机器视觉模型可基于VGGNet模型为内核基础进行改进，采用“Tensorflow+Python”系统框架构建全卷积神经网络模型，模型由5层卷积层、softmax输出层构成，并将原VGGNet中的3层全连接层替换为卷积层，层与层之间采用max-pooling与mean-pooling分开，激活函数为Maxout函数；The bridge critical component disease early warning system based on image monitoring data according to claim 4, characterized in that: the convolutional neural network machine vision model can be improved based on the VGGNet model based on the kernel, using the "Tensorflow+Python" system framework Construct a fully convolutional neural network model. The model consists of 5 layers of convolutional layers and softmax output layers. The 3 layers of fully connected layers in the original VGGNet are replaced with convolutional layers, and max-pooling and mean- are used between layers. Pooling is separated, and the activation function is Maxout function;
    卷积神经网络机器视觉模型采用已有桥梁构件的病害图像进行训练，每类构件的典型病害选取200～500张病害原始图像，并采用PHOTOSHOP软件将图像中的病害进行人工标记与分类，以病害原始图像作为模型输入数据，对应的病害标记图像作为模型输 出数据，模型正确识别率需达到95％以上。The convolutional neural network machine vision model uses the disease images of the existing bridge components for training. The typical disease of each type of component selects 200 to 500 original disease images, and uses the PHOTOSHOP software to manually mark and classify the diseases in the images. The original image is used as the input data of the model, and the corresponding disease-labeled image is used as the output data of the model. The correct recognition rate of the model needs to reach more than 95%.
  6. 根据权利要求1所述的基于图像监控数据的桥梁关键构件病害预警系统，其特征在于：所述评估预警系统对病害二值化图像进行分析，采用MATLAB自带函数Bwmorph与Minboundrect确定裂缝病害的长度与分布面积；采用直接统计二值化病害图像的1值像素数量确定坑洞病害的面积；The bridge critical component disease early warning system based on image monitoring data according to claim 1, characterized in that: the evaluation early warning system analyzes the binary image of the disease, and uses MATLAB's own functions Bwmorph and Minboundrect to determine the length of the crack disease And the distribution area; directly count the number of 1-value pixels in the binary disease image to determine the area of the pothole disease;
    对比同一构件不同时间的病害图像数据，针对裂缝病害，逐步对比裂缝长度与分布面积的变化趋势；对于坑洞病害，逐步对比面积的变化趋势；若病害的变化趋势增长速率超过规定阈值则进行预警；若裂缝病害的长度、或裂缝与坑洞病害的面积超过规定最大值则发出警报。Compare the disease image data of the same component at different times, and gradually compare the change trend of the crack length and the distribution area for the crack disease; for the pothole disease, gradually compare the change trend of the area; if the growth rate of the disease change trend exceeds the specified threshold, an early warning is given ; If the length of the crack disease or the area of the crack and the pothole disease exceeds the specified maximum value, an alarm will be issued.
  7. 一种基于图像监控数据的桥梁关键构件病害预警方法，其特征在于，包括以下步骤：An early warning method for bridge critical component diseases based on image monitoring data is characterized in that it includes the following steps:
    在桥梁各关键构件处设置高清相机，进行图像数据采集并传输至数据接收端；Set up high-definition cameras at each key component of the bridge to collect and transmit image data to the data receiving end;
    对采集的图像进行自动化读取、优化、病害识别与分类；Automatically read, optimize, identify and classify the collected images;
    以构件的病害二值化图像为研究对象，若构件病害程度超过极限阈值即进行预警。Taking the component disease binarization image as the research object, if the component disease degree exceeds the limit threshold, an early warning is given.
  8. 根据权利要求7所述的图像监控数据的桥梁关键构件病害预警方法，其特征在于，桥梁关键构件包括桥梁接缝、缆索端部、支座、箱梁结构与桥墩；The method for early warning of bridge key component diseases based on image monitoring data according to claim 7, wherein the bridge key components include bridge joints, cable ends, supports, box girder structures and bridge piers;
    高清相机固定于桥梁关键构件并进行拍摄，调节拍摄角度与距离以覆盖构件范围，并通过调整焦距使其识别精度达到0.5mm以上，拍摄图像为256色RGB图像，最小分辨率为2432×2048，采集频率可根据实际情况在1min～1d之间调节；The high-definition camera is fixed to the key components of the bridge and shoots, adjust the shooting angle and distance to cover the range of the components, and adjust the focal length to make the recognition accuracy reach 0.5mm or more. The captured image is a 256-color RGB image with a minimum resolution of 2432×2048. The acquisition frequency can be adjusted between 1 min and 1 d according to the actual situation;
    所述图像采集与数据传输系统采用4G无线传输模块，将原始拍摄图像传输并按照拍摄编号、拍摄时间存储于数据接收终端。The image acquisition and data transmission system adopts a 4G wireless transmission module to transmit the original captured image and store it in the data receiving terminal according to the shooting number and shooting time.
  9. 根据权利要求7所述的图像监控数据的桥梁关键构件病害预警方法，其特征在于，对采集的高清图像进行自动化读取、优化、病害识别与分类，具体方法为：The method for early warning of bridge key component diseases based on image monitoring data according to claim 7, characterized in that the automatic reading, optimization, disease identification and classification of the collected high-definition images are carried out by:
    利用MATLAB软件按照编号与时间顺序读取桥梁关键结构部位的256色RGB图像，采用自带函数imadjust与离散余弦变换对图像数据进行增强亮度、提高对比度与去除噪点处理，其后将图像数据输入卷积神经网络机器视觉模型，若图像识别无病害则输出标识0、裂缝类病害输出标识1、坑洞病害输出标识2，并输出病害图像的二值化图像。Use MATLAB software to read the 256-color RGB image of the key structural parts of the bridge according to the number and time sequence, and use the built-in function imadjust and discrete cosine transform to enhance the brightness, improve the contrast and remove the noise of the image data, and then input the image data into the volume Product neural network machine vision model, if the image recognition is disease-free, the output flag 0, the crack-type disease output flag 1, the pothole disease output flag 2, and the binary image of the disease image are output.
    所述卷积神经网络机器视觉模型基于VGGNet模型为内核基础进行改进，采用“Tensorflow+Python”系统框架构建全卷积神经网络模型，模型采用5层卷积层、softmax 输出层构成，并将原VGGNet中的3层全连接层替换为卷积层，层与层之间采用max-pooling与mean-pooling分开，激活函数为Maxout函数；The convolutional neural network machine vision model is improved based on the VGGNet model as the kernel basis, and the full convolutional neural network model is constructed using the "Tensorflow+Python" system framework. The model is composed of 5 layers of convolutional layers and softmax output layers, and the original The 3-layer fully connected layer in VGGNet is replaced with a convolutional layer. The layers are separated by max-pooling and mean-pooling, and the activation function is the Maxout function;
    卷积神经网络机器视觉模型采用已有桥梁构件的病害图像进行训练，每类构件的典型病害选取200～500张病害原始图像，并采用PHOTOSHOP软件将图像中的病害进行人工标记与分类，以病害原始图像作为模型输入数据，对应的病害标记图像作为模型输出数据，模型正确识别率需达到95％以上。The convolutional neural network machine vision model uses the disease images of the existing bridge components for training. The typical disease of each type of component selects 200 to 500 original disease images, and uses the PHOTOSHOP software to manually mark and classify the diseases in the images. The original image is used as the input data of the model, and the corresponding disease-labeled image is used as the output data of the model. The correct recognition rate of the model needs to reach more than 95%.
  10. 根据权利要求7所述的图像监控数据的桥梁关键构件病害预警方法，其特征在于，以病害二值化图像为研究对象，对可能发生病害的发展进行跟踪与状况评价，若病害发展速率与严重程度超过极限阈值即进行预警或发出警告，具体方法为：The method for early warning of bridge critical component diseases based on image monitoring data according to claim 7, characterized in that the binary image of the disease is used as the research object to track and evaluate the development of the disease that may occur. If the degree exceeds the limit threshold, an early warning or a warning is issued. The specific method is:
    对病害二值化图像进行分析，采用MATLAB自带函数Bwmorph与Minboundrect确定裂缝病害的长度与分布面积；采用直接统计二值化病害图像的1值像素数量确定坑洞病害的面积；Analyze the binary image of the disease, use MATLAB's own functions Bwmorph and Minboundrect to determine the length and distribution area of the crack disease; use the direct statistics of the number of 1-value pixels of the binary disease image to determine the area of the pothole disease;
    对比同一构件不同时间的病害图像数据，针对裂缝病害，逐步对比裂缝长度与分布面积的变化趋势；对于坑洞病害，逐步对比面积的变化趋势；若病害的变化趋势增长速率超过规定阈值则进行预警；若裂缝病害的长度、或裂缝与坑洞病害的面积超过规定最大值则发出警报。Compare the disease image data of the same component at different times, and gradually compare the change trend of the crack length and the distribution area for the crack disease; for the pothole disease, gradually compare the change trend of the area; if the growth rate of the disease change trend exceeds the specified threshold, an early warning is given ; If the length of the crack disease or the area of the crack and the pothole disease exceeds the specified maximum value, an alarm will be issued.
PCT/CN2019/108089 2019-04-04 2019-09-26 Bridge key component disease early-warning system and method based on image monitoring data WO2020199538A1 (en)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201910272711.7A CN110031477B (en) 2019-04-04 2019-04-04 Bridge key component disease early warning system and method based on image monitoring data
CN201910272711.7 2019-04-04

Publications (1)

Publication Number Publication Date
WO2020199538A1 true WO2020199538A1 (en) 2020-10-08

Family

ID=67237406

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2019/108089 WO2020199538A1 (en) 2019-04-04 2019-09-26 Bridge key component disease early-warning system and method based on image monitoring data

Country Status (2)

Country Link
CN (1) CN110031477B (en)
WO (1) WO2020199538A1 (en)

Cited By (28)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112419288A (en) * 2020-11-28 2021-02-26 西安石油大学 Unmanned vegetable greenhouse planting method based on computer vision
CN112487925A (en) * 2020-11-25 2021-03-12 上海海事大学 Bridge load damage identification method and system
CN112699736A (en) * 2020-12-08 2021-04-23 江西省交通科学研究院 Bridge bearing fault identification method based on space attention
CN112749733A (en) * 2020-11-27 2021-05-04 江西省交通科学研究院 Bridge disease identification method based on mixed model and image pyramid
CN113063795A (en) * 2021-03-31 2021-07-02 辽宁省交通规划设计院有限责任公司 Method and system for determining defect position of expressway tunnel
CN113128371A (en) * 2021-04-01 2021-07-16 中铁大桥局集团有限公司 Operation-period bridge monitoring system and method based on automatic visual scanning
CN113570580A (en) * 2021-07-30 2021-10-29 重庆予胜远升网络科技有限公司 Power equipment part loosening identification system and method based on machine vision
CN113569908A (en) * 2021-06-24 2021-10-29 河南大学 Bridge technical condition and disease prediction back-stepping calculation method based on deep learning
CN113657542A (en) * 2021-08-27 2021-11-16 中大检测(湖南)股份有限公司 Multi-dimensional correlation early warning system
CN113720438A (en) * 2021-07-23 2021-11-30 鹏城实验室 Component crack information monitoring method and device, terminal equipment and storage medium
CN114841995A (en) * 2022-06-02 2022-08-02 西南交通大学 Deep learning-based railway roadbed fender equipment service state evaluation method
CN114913473A (en) * 2022-03-21 2022-08-16 中国科学院光电技术研究所 Lightweight single-body imaging contact network safety patrol instrument
CN115184378A (en) * 2022-09-15 2022-10-14 北京思莫特科技有限公司 Concrete structure disease detection system and method based on mobile device
CN115710941A (en) * 2022-11-23 2023-02-24 江苏海洋大学 Underwater pile foundation crack loss detection device and detection method based on VR technology
CN115790400A (en) * 2023-01-17 2023-03-14 中大智能科技股份有限公司 Machine vision target calibration method applied to bridge and tunnel structure safety
CN115937595A (en) * 2022-12-20 2023-04-07 中交公路长大桥建设国家工程研究中心有限公司 Bridge apparent anomaly identification method and system based on intelligent data processing
CN116206266A (en) * 2023-05-06 2023-06-02 济宁盛世照明电器有限公司 Vision counterpoint system based on capper
CN116415507A (en) * 2023-05-31 2023-07-11 四川省公路规划勘察设计研究院有限公司 Big transportation bridge intelligent evaluation system based on big data and cloud computing
CN116596508A (en) * 2023-05-12 2023-08-15 广东交科检测有限公司 Slope crack disease development identification recording method
CN116629618A (en) * 2023-07-19 2023-08-22 铁正检测科技有限公司 Bridge disease diagnosis system based on Internet
CN116630899A (en) * 2023-07-21 2023-08-22 四川公路工程咨询监理有限公司 Highway side slope disease monitoring and early warning system
CN116697922A (en) * 2023-08-04 2023-09-05 沂水华辰房地产测绘有限公司 Big data-based monitoring system for engineering measurement and application method thereof
CN116698854A (en) * 2023-08-04 2023-09-05 广州市市政工程试验检测有限公司 Engineering structure health detection method and system
CN116797436A (en) * 2023-08-29 2023-09-22 北京道仪数慧科技有限公司 Processing system for carrying out road disease inspection by utilizing bus
CN116958138A (en) * 2023-09-19 2023-10-27 河南省建筑质量监督检验中心有限公司 Sampling control method, system, equipment and medium for bridge detection
CN117152492A (en) * 2023-08-08 2023-12-01 广东省公路建设有限公司湾区特大桥养护技术中心 Method, system, computer equipment and medium for identifying fatigue crack in steel box girder
CN117152163A (en) * 2023-11-01 2023-12-01 安徽乾劲企业管理有限公司 Bridge construction quality visual detection method
CN113570580B (en) * 2021-07-30 2024-06-07 重庆予胜远升网络科技有限公司 Power equipment component loosening identification system and method based on machine vision

Families Citing this family (14)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110031477B (en) * 2019-04-04 2021-01-01 华设设计集团股份有限公司 Bridge key component disease early warning system and method based on image monitoring data
CN110852576A (en) * 2019-10-22 2020-02-28 中冶建筑研究总院有限公司 Bridge technical condition rapid evaluation method based on intelligent monitoring
CN111272764B (en) * 2020-01-22 2023-04-28 哈尔滨工业大学 Non-contact image identification mobile management and control system and method for large intelligent temporary platform
CN111340772A (en) * 2020-02-24 2020-06-26 湖北工业大学 Reinforced concrete bridge damage detection system and method based on mobile terminal
CN111795978B (en) * 2020-09-08 2020-12-04 湖南大学 Steel bridge structure health state assessment method, device, equipment and storage medium
CN112229586B (en) * 2020-09-30 2022-10-14 石家庄铁道大学 Bridge support health state detection method and terminal equipment
CN112411371A (en) * 2020-12-14 2021-02-26 湖南大学 Highway bridge comprehensive detection method and system based on mobile sensing and vision
CN112903710A (en) * 2021-01-22 2021-06-04 山东高速工程检测有限公司 Method, system and device for monitoring apparent bridge diseases
CN113123317B (en) * 2021-04-20 2022-05-10 交通运输部公路科学研究所 Device for measuring base layer after dismantling concrete pavement slab and state evaluation method
CN113406088A (en) * 2021-05-10 2021-09-17 同济大学 Fixed point type steel box girder crack development observation device
CN113267503A (en) * 2021-05-12 2021-08-17 同济大学 Adsorption type steel structure linear scanning device
CN113505706A (en) * 2021-07-14 2021-10-15 陈小义 Automatic disease identification method and system based on AI intelligence
CN115389523A (en) * 2022-08-26 2022-11-25 成都木格子装饰工程有限公司 Building engineering supervision analysis system based on man-machine cooperation and visual inspection
CN117592820B (en) * 2024-01-18 2024-03-29 贵州顺康检测股份有限公司 Bridge damage disease intelligent recognition system based on computer data analysis

Citations (12)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20080219517A1 (en) * 2007-03-05 2008-09-11 Fotonation Vision Limited Illumination Detection Using Classifier Chains
JP2012026881A (en) * 2010-07-23 2012-02-09 Nippon Telegr & Teleph Corp <Ntt> Collapse detection system and collapse detection method
CN102663344A (en) * 2012-03-04 2012-09-12 南京理工大学常熟研究院有限公司 Damaged road detection device based on image segmentation
CN104751603A (en) * 2015-04-10 2015-07-01 招商局重庆交通科研设计院有限公司 Rockfall monitoring and early warning system and rockfall monitoring and early warning method
CN106841216A (en) * 2017-02-28 2017-06-13 浙江工业大学 Tunnel defect automatic identification equipment based on panoramic picture CNN
CN107862677A (en) * 2017-10-16 2018-03-30 中铁第四勘察设计院集团有限公司 The Tunnel Lining Cracks recognition methods of thresholding algorithm and system between a kind of class based on gradient
CN108288269A (en) * 2018-01-24 2018-07-17 东南大学 Bridge pad disease automatic identifying method based on unmanned plane and convolutional neural networks
CN108332927A (en) * 2018-01-10 2018-07-27 深圳市晟达机械设计有限公司 A kind of bridge surface crack detection device
CN108765404A (en) * 2018-05-31 2018-11-06 南京行者易智能交通科技有限公司 A kind of road damage testing method and device based on deep learning image classification
CN108918539A (en) * 2018-07-27 2018-11-30 同济大学 A kind of apparent Defect inspection device and method of tunnel structure
CN109064449A (en) * 2018-07-04 2018-12-21 中铁大桥科学研究院有限公司 A kind of detection method of bridge surface disease
CN110031477A (en) * 2019-04-04 2019-07-19 中设设计集团股份有限公司 Bridge key component disease early warning system and method based on image monitoring data

Family Cites Families (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101169380B (en) * 2007-10-31 2010-09-15 重庆大学 Bridge cable surface damage dynamic detection method and device
CN101726497B (en) * 2009-12-11 2011-12-28 昆明理工大学 Method for evaluating safety of service bridge structure
CN108734692A (en) * 2018-03-26 2018-11-02 湖南城市学院 A kind of bridge section model system
CN109300126B (en) * 2018-09-21 2022-01-07 重庆建工集团股份有限公司 High-precision intelligent detection method for bridge diseases based on spatial positions
CN109544492A (en) * 2018-10-25 2019-03-29 东南大学 A kind of multi-focus image fusion data set production method based on convolutional neural networks
CN109521019A (en) * 2018-11-09 2019-03-26 华南理工大学 A kind of bridge bottom crack detection method based on unmanned plane vision

Patent Citations (12)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20080219517A1 (en) * 2007-03-05 2008-09-11 Fotonation Vision Limited Illumination Detection Using Classifier Chains
JP2012026881A (en) * 2010-07-23 2012-02-09 Nippon Telegr & Teleph Corp <Ntt> Collapse detection system and collapse detection method
CN102663344A (en) * 2012-03-04 2012-09-12 南京理工大学常熟研究院有限公司 Damaged road detection device based on image segmentation
CN104751603A (en) * 2015-04-10 2015-07-01 招商局重庆交通科研设计院有限公司 Rockfall monitoring and early warning system and rockfall monitoring and early warning method
CN106841216A (en) * 2017-02-28 2017-06-13 浙江工业大学 Tunnel defect automatic identification equipment based on panoramic picture CNN
CN107862677A (en) * 2017-10-16 2018-03-30 中铁第四勘察设计院集团有限公司 The Tunnel Lining Cracks recognition methods of thresholding algorithm and system between a kind of class based on gradient
CN108332927A (en) * 2018-01-10 2018-07-27 深圳市晟达机械设计有限公司 A kind of bridge surface crack detection device
CN108288269A (en) * 2018-01-24 2018-07-17 东南大学 Bridge pad disease automatic identifying method based on unmanned plane and convolutional neural networks
CN108765404A (en) * 2018-05-31 2018-11-06 南京行者易智能交通科技有限公司 A kind of road damage testing method and device based on deep learning image classification
CN109064449A (en) * 2018-07-04 2018-12-21 中铁大桥科学研究院有限公司 A kind of detection method of bridge surface disease
CN108918539A (en) * 2018-07-27 2018-11-30 同济大学 A kind of apparent Defect inspection device and method of tunnel structure
CN110031477A (en) * 2019-04-04 2019-07-19 中设设计集团股份有限公司 Bridge key component disease early warning system and method based on image monitoring data

Cited By (48)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112487925A (en) * 2020-11-25 2021-03-12 上海海事大学 Bridge load damage identification method and system
CN112487925B (en) * 2020-11-25 2024-05-14 上海海事大学 Bridge load damage identification method and system
CN112749733A (en) * 2020-11-27 2021-05-04 江西省交通科学研究院 Bridge disease identification method based on mixed model and image pyramid
CN112419288B (en) * 2020-11-28 2024-05-03 西安石油大学 Unmanned vegetable greenhouse planting method based on computer vision
CN112419288A (en) * 2020-11-28 2021-02-26 西安石油大学 Unmanned vegetable greenhouse planting method based on computer vision
CN112699736A (en) * 2020-12-08 2021-04-23 江西省交通科学研究院 Bridge bearing fault identification method based on space attention
CN112699736B (en) * 2020-12-08 2024-06-07 江西省交通科学研究院有限公司 Bridge bearing disease identification method based on spatial attention
CN113063795A (en) * 2021-03-31 2021-07-02 辽宁省交通规划设计院有限责任公司 Method and system for determining defect position of expressway tunnel
CN113063795B (en) * 2021-03-31 2024-02-02 辽宁省交通规划设计院有限责任公司 Expressway tunnel defect position determining method and system
CN113128371A (en) * 2021-04-01 2021-07-16 中铁大桥局集团有限公司 Operation-period bridge monitoring system and method based on automatic visual scanning
CN113128371B (en) * 2021-04-01 2023-06-23 中铁大桥局集团有限公司 Automatic visual scanning-based operation period bridge monitoring system and method
CN113569908B (en) * 2021-06-24 2024-02-27 河南大学 Bridge technical condition and disease prediction reverse calculation method based on deep learning
CN113569908A (en) * 2021-06-24 2021-10-29 河南大学 Bridge technical condition and disease prediction back-stepping calculation method based on deep learning
CN113720438A (en) * 2021-07-23 2021-11-30 鹏城实验室 Component crack information monitoring method and device, terminal equipment and storage medium
CN113720438B (en) * 2021-07-23 2023-06-16 鹏城实验室 Component crack information monitoring method and device, terminal equipment and storage medium
CN113570580A (en) * 2021-07-30 2021-10-29 重庆予胜远升网络科技有限公司 Power equipment part loosening identification system and method based on machine vision
CN113570580B (en) * 2021-07-30 2024-06-07 重庆予胜远升网络科技有限公司 Power equipment component loosening identification system and method based on machine vision
CN113657542A (en) * 2021-08-27 2021-11-16 中大检测(湖南)股份有限公司 Multi-dimensional correlation early warning system
CN114913473B (en) * 2022-03-21 2023-08-15 中国科学院光电技术研究所 Lightweight monomer type imaging contact net safety inspection instrument
CN114913473A (en) * 2022-03-21 2022-08-16 中国科学院光电技术研究所 Lightweight single-body imaging contact network safety patrol instrument
CN114841995A (en) * 2022-06-02 2022-08-02 西南交通大学 Deep learning-based railway roadbed fender equipment service state evaluation method
CN115184378A (en) * 2022-09-15 2022-10-14 北京思莫特科技有限公司 Concrete structure disease detection system and method based on mobile device
CN115184378B (en) * 2022-09-15 2024-03-29 北京思莫特科技有限公司 Concrete structure disease detection system and method based on mobile equipment
CN115710941A (en) * 2022-11-23 2023-02-24 江苏海洋大学 Underwater pile foundation crack loss detection device and detection method based on VR technology
CN115937595A (en) * 2022-12-20 2023-04-07 中交公路长大桥建设国家工程研究中心有限公司 Bridge apparent anomaly identification method and system based on intelligent data processing
CN115790400A (en) * 2023-01-17 2023-03-14 中大智能科技股份有限公司 Machine vision target calibration method applied to bridge and tunnel structure safety
CN116206266B (en) * 2023-05-06 2023-08-15 济宁盛世照明电器有限公司 Vision counterpoint system based on capper
CN116206266A (en) * 2023-05-06 2023-06-02 济宁盛世照明电器有限公司 Vision counterpoint system based on capper
CN116596508A (en) * 2023-05-12 2023-08-15 广东交科检测有限公司 Slope crack disease development identification recording method
CN116596508B (en) * 2023-05-12 2024-03-08 广东交科检测有限公司 Slope crack disease development identification recording method
CN116415507B (en) * 2023-05-31 2023-08-08 四川省公路规划勘察设计研究院有限公司 Big transportation bridge intelligent evaluation system based on big data and cloud computing
CN116415507A (en) * 2023-05-31 2023-07-11 四川省公路规划勘察设计研究院有限公司 Big transportation bridge intelligent evaluation system based on big data and cloud computing
CN116629618A (en) * 2023-07-19 2023-08-22 铁正检测科技有限公司 Bridge disease diagnosis system based on Internet
CN116629618B (en) * 2023-07-19 2023-10-03 铁正检测科技有限公司 Bridge disease diagnosis system based on Internet
CN116630899B (en) * 2023-07-21 2023-10-20 四川公路工程咨询监理有限公司 Highway side slope disease monitoring and early warning system
CN116630899A (en) * 2023-07-21 2023-08-22 四川公路工程咨询监理有限公司 Highway side slope disease monitoring and early warning system
CN116697922B (en) * 2023-08-04 2023-12-12 沂水华辰房地产测绘有限公司 Big data-based monitoring system for engineering measurement and application method thereof
CN116698854B (en) * 2023-08-04 2023-11-21 广州市市政工程试验检测有限公司 Engineering structure health detection method and system
CN116698854A (en) * 2023-08-04 2023-09-05 广州市市政工程试验检测有限公司 Engineering structure health detection method and system
CN116697922A (en) * 2023-08-04 2023-09-05 沂水华辰房地产测绘有限公司 Big data-based monitoring system for engineering measurement and application method thereof
CN117152492B (en) * 2023-08-08 2024-04-19 广东省公路建设有限公司湾区特大桥养护技术中心 Method, system, computer equipment and medium for identifying fatigue crack in steel box girder
CN117152492A (en) * 2023-08-08 2023-12-01 广东省公路建设有限公司湾区特大桥养护技术中心 Method, system, computer equipment and medium for identifying fatigue crack in steel box girder
CN116797436B (en) * 2023-08-29 2023-10-31 北京道仪数慧科技有限公司 Processing system for carrying out road disease inspection by utilizing bus
CN116797436A (en) * 2023-08-29 2023-09-22 北京道仪数慧科技有限公司 Processing system for carrying out road disease inspection by utilizing bus
CN116958138B (en) * 2023-09-19 2023-12-19 河南省建筑质量监督检验中心有限公司 Sampling control method, system, equipment and medium for bridge detection
CN116958138A (en) * 2023-09-19 2023-10-27 河南省建筑质量监督检验中心有限公司 Sampling control method, system, equipment and medium for bridge detection
CN117152163B (en) * 2023-11-01 2024-02-27 安徽乾劲企业管理有限公司 Bridge construction quality visual detection method
CN117152163A (en) * 2023-11-01 2023-12-01 安徽乾劲企业管理有限公司 Bridge construction quality visual detection method

Also Published As

Publication number Publication date
CN110031477B (en) 2021-01-01
CN110031477A (en) 2019-07-19

Similar Documents

Publication Publication Date Title
WO2020199538A1 (en) Bridge key component disease early-warning system and method based on image monitoring data
CN108918539B (en) Apparent disease detection device and method for tunnel structure
CN108109385B (en) System and method for identifying and judging dangerous behaviors of power transmission line anti-external damage vehicle
CN103487729B (en) Based on the power equipments defect detection method that ultraviolet video and infrared video merge
KR20200013148A (en) Method, system and computer program for providing defect analysis service of concrete structure
CN107588733A (en) A kind of abrasion of pantograph pan on-line measuring device and method based on image
CN103413150A (en) Power line defect diagnosis method based on visible light image
CN103808723A (en) Exhaust gas blackness automatic detection device for diesel vehicles
CN110046584B (en) Road crack detection device and detection method based on unmanned aerial vehicle inspection
CN112528979B (en) Transformer substation inspection robot obstacle distinguishing method and system
CN105550710A (en) Nonlinear fitting based intelligent detection method for running exception state of contact network
CN113947555A (en) Infrared and visible light fused visual system and method based on deep neural network
CN112508911A (en) Rail joint touch net suspension support component crack detection system based on inspection robot and detection method thereof
CN116297199B (en) Detection system for automatically detecting lithium battery parameters based on machine vision
CN111667655A (en) Infrared image-based high-speed railway safety area intrusion alarm device and method
CN113205039A (en) Power equipment fault image identification and disaster investigation system and method based on multiple DCNNs
CN116310274A (en) State evaluation method for power transmission and transformation equipment
CN113673614B (en) Metro tunnel foreign matter intrusion detection device and method based on machine vision
CN107147877A (en) FX night fog day condition all-weather colorful video imaging system and its construction method
CN105987665B (en) Early warning monitoring device and method for sag point change range of ultra-high voltage transmission line
CN116086547B (en) Contact net icing detection method based on infrared imaging and meteorological monitoring
CN108198164A (en) A kind of detection device and method of cable tunnel cable integrity
CN117152617A (en) Urban flood identification method and system
CN107194923B (en) Ultraviolet image diagnosis method for defect inspection of contact network power equipment
CN113657621A (en) Hidden danger monitoring method and system

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 19922566

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 19922566

Country of ref document: EP

Kind code of ref document: A1