CN111551562A - Bridge pavement structure damage identification method and system - Google Patents

Abstract

The invention discloses a bridge pavement structure damage identification method and a system, which are used for collecting bridge pavement video data and vehicle position data and realizing the matching of time and space dimensions of the data; receiving data transmitted by a data acquisition system, and realizing the processing of structured data and unstructured data through a GPS data module and a video data processing module; and finally, the structural damage identification subsystem identifies the damage of the road surface through the abnormal GPS identification module, the video correction module and the structural damage identification module and issues a damage prompt. The invention utilizes the video acquisition equipment and the GPS acquisition equipment to acquire the multi-source data set, can accurately identify the structural damage of the bridge pavement and provide reliable information for the bridge operation and maintenance management department, thereby formulating a reasonable and efficient operation and maintenance scheme and management measures and lightening the intensity of workers.

Description

A method and system for damage identification of bridge pavement structure

technical field

The invention relates to the field of intelligent transportation, and more particularly, to a method and system for identifying damage to bridge pavement structures.

Background technique

Bridges are key nodes in the transportation network, and their health status directly affects the safe operation of vehicles. Overload operation of vehicles is common in my country, and bridge structures, especially pavement structures, are often damaged during their service life, which will lead to accelerated performance degradation, followed by a sharp decline in structural bearing capacity, which greatly increases the risk of driving. Bridge structural health monitoring can effectively identify various damages of bridge structures and provide accurate information for the assessment of bridge structural conditions.

At present, bridge pavement damage is mainly realized through manual inspection, which is highly subjective, has a large workload and a high cost. Moreover, there are many bridges in my country and the workload is large. The detection will also affect the road traffic operation and cause traffic congestion. . In a word, it is very important to improve the automation level of bridge pavement damage detection.

SUMMARY OF THE INVENTION

The invention provides a bridge pavement structure damage identification method and system, which fully utilizes multi-source data fusion to improve the accuracy of pavement damage, improve detection efficiency and accuracy, and save detection costs.

For solving the above-mentioned technical problems, the technical scheme of the present invention is as follows:

A method for identifying damage to a bridge pavement structure, comprising the following steps:

S1: Use roadside video acquisition equipment to collect image data of the bridge pavement, the bridge pavement is divided into grids, wherein each grid is matched with GPS data points corresponding to the bridge geographic information using a data matching module, and the GPS data and image data are stored in the data storage module;

S2: Use the video data processing module to slice the image data obtained in S1 to obtain the image of each grid, and use the manual calibration method to determine the damage status of each grid pavement. Images without pavement damage are classified as normal samples, normal samples are marked as 0, and damaged samples are marked as 1;

S3: The GPS data processing module uses the damage samples and normal samples to construct a training data set, uses the training data set to train the model, and obtains a trained model. The model is used to identify the damage of the pavement structure, and the input of the model is all the data in the grid. GPS data, the output is the state of the grid, that is, 0 or 1, 0 means the road surface is normal, 1 means the road surface is damaged;

S4: The vehicle transmits the GPS data collected in real time to the abnormal GPS identification module through the wireless communication network, the video proofreading module slices the image collected by the roadside video collection equipment, and the structural damage identification module outputs the damage status of the road surface, including two types: damage and Normal; if normal, the output is normal; if damaged, map matching is performed to determine the location of the damage, and at the same time, the management personnel are notified. It is a normal sample, and the output is normal. At the same time, the management personnel are notified. The management personnel watch the video to determine whether there is damage. If there is damage, report the damage type and location. If it is normal, it will be marked as a normal sample and the output is normal.

Preferably, the GPS data points in step S1 are collected by a vehicle-mounted GPS device.

Preferably, the grid size of the bridge pavement after grid division is 0.1m×0.1m.

Preferably, the image obtained by slicing in step S2 is consistent with the size of the grid.

Preferably, the model training method is any one of neural network, support vector machine and classification tree or other methods.

Preferably, the model identifies both the presence of structural damage and the location of the damage.

Preferably, the structural damage includes cracks, subsidence and rutting.

Preferably, the GPS data includes three variables: time stamp, longitude, and latitude.

A bridge pavement structure damage identification system based on the above-mentioned identification method, comprising:

The data storage module is used to store the GPS data collected by the vehicle-mounted GPS equipment and the image data collected by the roadside equipment;

The data matching module is used to obtain data from the data storage module, and match GPS data and video slice data to the divided grid;

The video data processing module is used for slicing the video data, manually calibrating the category of the video slicing, calibrating the grid division state according to the position of the video slicing, and transmitting the calibration result to the GPS data processing module;

The GPS data processing module is used to receive the results of the data matching module, and aggregate the GPS data in each grid as an input variable, train the structural damage identification model, and transmit the trained identification model to the abnormal GPS identification module;

The abnormal GPS identification module takes the GPS data collected in real time as input, and determines the state of the grid in real time based on the identification model;

The video proofreading module is used to extract video slices of a specific grid, further confirm the state of the grid, and transmit the results to the structural damage identification module;

Structural damage identification module, used to transmit the final identification result.

Compared with the prior art, the beneficial effects of the technical solution of the present invention are:

1. The present invention integrates roadside collection equipment data and vehicle-mounted collection equipment data to improve the accuracy of detection;

2. The present invention improves the automation level of detection based on the machine learning method;

3. The present invention can accurately identify the structural damage of the bridge pavement, and provide a reliable management basis for the traffic planning and operation management departments.

Description of drawings

FIG. 1 is a schematic flow chart of the method of the present invention.

FIG. 2 is a bridge meshing method in the present invention.

FIG. 3 is a schematic diagram of the system structure of the present invention.

Detailed ways

The accompanying drawings are for illustrative purposes only, and should not be construed as limitations on this patent;

In order to better illustrate this embodiment, some parts of the drawings are omitted, enlarged or reduced, which do not represent the size of the actual product;

It will be understood by those skilled in the art that some well-known structures and their descriptions may be omitted from the drawings.

The technical solutions of the present invention will be further described below with reference to the accompanying drawings and embodiments.

Example 1

This embodiment provides a method for identifying damage to a bridge pavement structure, as shown in Figure 1, including the following steps:

S1: Use roadside video acquisition equipment to collect the image data of the bridge pavement. The bridge pavement is divided into grids, as shown in Figure 2, wherein each grid is matched with GPS data corresponding to the bridge geographic information using a data matching module Point, GPS data and image data are stored in the data storage module;

S2: Use the video data processing module to slice the image data obtained in S1 to obtain the image of each grid, and use the manual calibration method to determine the damage status of each grid pavement. Images without pavement damage are classified as normal samples, normal samples are marked as 0, and damaged samples are marked as 1;

S3: The GPS data processing module uses the damage samples and normal samples to construct a training data set, uses the training data set to train the model, and obtains a trained model. The model is used to identify the damage of the pavement structure, and the input of the model is all the data in the grid. GPS data, the output is the state of the grid, i.e. 0 or 1;

S4: The vehicle transmits the GPS data collected in real time to the abnormal GPS identification module through the wireless communication network, the video proofreading module slices the image collected by the roadside video collection equipment, and the structural damage identification module outputs the damage status of the road surface, including two types: damage and Normal; if normal, the output is normal; if damaged, map matching is performed to determine the location of the damage, and at the same time, the management personnel are notified. For normal samples, the output is normal.

The GPS data points in step S1 are collected by a vehicle-mounted GPS device.

The grid size of the bridge pavement after grid division is 0.1m×0.1m.

The image obtained by slicing in step S2 is consistent with the size of the grid.

The above model training method is any one of neural network, support vector machine and classification tree.

The model identifies both the presence of structural damage and the location of the damage.

The structural damage includes cracks, subsidence, and rutting.

GPS data includes three variables: time stamp, longitude, and latitude.

Example 2

This embodiment provides a bridge pavement structural damage identification system based on the identification method described in Embodiment 1, as shown in FIG. 3 , including a data acquisition subsystem, a data processing subsystem and a structural damage identification subsystem, wherein:

The data acquisition subsystem includes:

The data storage module is used to store the GPS data collected by the vehicle-mounted GPS equipment and the image data collected by the roadside equipment;

The data matching module is used to obtain data from the data storage module, and match GPS data and video slice data to the divided grid;

The data processing subsystem includes:

The video data processing module is used for slicing the video data, manually calibrating the category of the video slicing, calibrating the grid division state according to the position of the video slicing, and transmitting the calibration result to the GPS data processing module;

The GPS data processing module is used to receive the results of the data matching module, and aggregate the GPS data in each grid as an input variable, train the structural damage identification model, and transmit the trained identification model to the abnormal GPS identification module;

The structural damage identification subsystem includes:

The abnormal GPS identification module takes the GPS data collected in real time as input, and determines the state of the grid in real time based on the identification model;

The video proofreading module is used to extract video slices of a specific grid, further confirm the state of the grid, and transmit the results to the structural damage identification module;

Structural damage identification module, used to transmit the final identification result.

The same or similar reference numbers correspond to the same or similar parts;

The terms describing the positional relationship in the accompanying drawings are only used for exemplary illustration, and should not be construed as a limitation on this patent;

Obviously, the above-mentioned embodiments of the present invention are only examples for clearly illustrating the present invention, rather than limiting the embodiments of the present invention. For those of ordinary skill in the art, changes or modifications in other different forms can also be made on the basis of the above description. There is no need and cannot be exhaustive of all implementations here. Any modifications, equivalent replacements and improvements made within the spirit and principle of the present invention shall be included within the protection scope of the claims of the present invention.

Claims (9)

1. a bridge pavement structure damage identification method, is characterized in that, comprises the following steps: S1: Use roadside video acquisition equipment to collect image data of the bridge pavement, the bridge pavement is divided into grids, wherein each grid is matched with GPS data points corresponding to the bridge geographic information using a data matching module, and the GPS data and image data are stored in the data storage module; S2: Use the video data processing module to slice the image data obtained in S1 to obtain the image of each grid, and use the manual calibration method to determine the damage status of each grid pavement. Images without pavement damage are classified as normal samples, normal samples are marked as 0, and damaged samples are marked as 1; S3: The GPS data processing module uses the damage samples and normal samples to construct a training data set, uses the training data set to train the model, and obtains a trained model. The model is used to identify the damage of the pavement structure, and the input of the model is all the data in the grid. GPS data, the output is the state of the grid, i.e. 0 or 1; S4: The vehicle transmits the GPS data collected in real time to the abnormal GPS identification module through the wireless communication network, the video proofreading module slices the image collected by the roadside video collection equipment, and the structural damage identification module outputs the damage status of the road surface, including two types: damage and Normal; if normal, the output is normal; if damaged, map matching is performed to determine the location of the damage. 2 . The method for identifying damage to a bridge pavement structure according to claim 1 , wherein the GPS data points in step S1 are collected by a vehicle-mounted GPS device. 3 . 3 . The method for identifying damage to a bridge pavement structure according to claim 2 , wherein the grid size of the bridge pavement after grid division is 0.1m×0.1m. 4 . 4 . The method for identifying damage to bridge pavement structures according to claim 3 , wherein the image obtained by slicing in step S2 is consistent with the size of the grid. 5 . 5 . The method for identifying damage to bridge pavement structure according to claim 4 , wherein the model training method is any one of neural network, support vector machine and classification tree. 6 . 6 . The method for identifying damage to a bridge pavement structure according to claim 5 , wherein the model simultaneously identifies whether the structure is damaged and the location of the damage. 7 . 7 . The method for identifying damage to a bridge pavement structure according to claim 6 , wherein the structural damage includes cracks, subsidence and rutting. 8 . 8 . The method for identifying damage to a bridge pavement structure according to claim 7 , wherein the GPS data includes three variables: time stamp, longitude and latitude. 9 . 9. A bridge pavement structure damage identification system based on the identification method according to claims 1 to 8, characterized in that, comprising: The data storage module is used to store the GPS data collected by the vehicle-mounted GPS equipment and the image data collected by the roadside equipment; The data matching module is used to obtain data from the data storage module, and match GPS data and video slice data to the divided grid; The video data processing module is used for slicing the video data, manually calibrating the category of the video slicing, calibrating the grid division state according to the position of the video slicing, and transmitting the calibration result to the GPS data processing module; The GPS data processing module is used to receive the results of the data matching module, and aggregate the GPS data in each grid as an input variable, train the structural damage identification model, and transmit the trained identification model to the abnormal GPS identification module; The abnormal GPS identification module takes the GPS data collected in real time as input, and determines the state of the grid in real time based on the identification model; The video proofreading module is used to extract video slices of a specific grid, further confirm the state of the grid, and transmit the results to the structural damage identification module; Structural damage identification module, used to transmit the final identification result.

Images