CN115620041B - Pavement disease sensing and service state evaluation method - Google Patents
Pavement disease sensing and service state evaluation method Download PDFInfo
- Publication number
- CN115620041B CN115620041B CN202211304508.1A CN202211304508A CN115620041B CN 115620041 B CN115620041 B CN 115620041B CN 202211304508 A CN202211304508 A CN 202211304508A CN 115620041 B CN115620041 B CN 115620041B
- Authority
- CN
- China
- Prior art keywords
- disease
- distribution
- pavement
- data
- weight
- 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
Classifications
-
- 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
- G06V10/762—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using clustering, e.g. of similar faces in social networks
- G06V10/763—Non-hierarchical techniques, e.g. based on statistics of modelling distributions
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/26—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion
-
- 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
- G06V10/77—Processing image or video features in feature spaces; using data integration or data reduction, e.g. principal component analysis [PCA] or independent component analysis [ICA] or self-organising maps [SOM]; Blind source separation
- G06V10/778—Active pattern-learning, e.g. online learning of image or video features
-
- 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
- Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
- Y02A90/10—Information and communication technologies [ICT] supporting adaptation to climate change, e.g. for weather forecasting or climate simulation
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Multimedia (AREA)
- General Physics & Mathematics (AREA)
- Databases & Information Systems (AREA)
- Evolutionary Computation (AREA)
- Software Systems (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Computing Systems (AREA)
- Artificial Intelligence (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Health & Medical Sciences (AREA)
- Probability & Statistics with Applications (AREA)
- Image Processing (AREA)
- Investigating Or Analysing Materials By Optical Means (AREA)
Abstract
The invention discloses a pavement disease sensing and service state evaluation method, which comprises the steps of obtaining image data and radar data, and obtaining pavement disease data by performing disease identification on the image data and the radar data; and counting the pavement damage data distribution data, and carrying out different dimension evaluation on the pavement service state based on pavement damage distribution to obtain a pavement service state evaluation result. Through the technical scheme, the road surface disease detection method and the road surface disease detection system can be used for effectively detecting road surface diseases and evaluating the service state.
Description
Technical Field
The invention relates to the technical field of pavement damage sensing and evaluation, in particular to a pavement damage sensing and service state evaluation method.
Background
Asphalt pavement is a typical road tunnel pavement form, fatigue cracking and reflection cracking are tunnel asphalt pavement defects which are the most typical and widely occur, and the occurrence of the cracks greatly reduces the integrity of the pavement, has a larger influence on the bearing capacity of the pavement and seriously threatens the driving safety. At present, the traditional detection or perception technology for asphalt pavement cracks mainly comprises the following steps: manual inspection, rapid detection vehicle and various embedded sensors. Most areas still rely on manual inspection, but manual inspection is inefficient and has large errors. Along with the rapid development of image processing technology, laser scanning technology and other technologies, the rapid detection system for the pavement is rapid in development, the application of the rapid detection system obviously improves the single detection speed, but the identification precision of the prior art is low, and the requirements on the precise management and maintenance of the pavement cannot be met.
The detection and treatment of the operation road surface diseases become key problems of road surface management and maintenance, the service road is regularly and comprehensively and healthily detected, the whole health condition of the road surface structure is analyzed, the road surface state, the diseases such as cracks, water leakage and the like are perceived, and timely repair and maintenance work is carried out according to the severity degree of the diseases, so that the road surface driving safety coefficient is effectively improved, and the economic loss and the potential safety hazard are reduced. The research on the road surface state and disease perception in the operation period is developed, great help can be provided for the management maintenance and the treatment and the safety of the road, the urgent need of guaranteeing the health and long-term service of the road surface structure is met, and great theoretical significance and engineering value are achieved. In the prior art, the problems that effective and accurate detection and evaluation of the pavement are difficult exist, and aiming at the problems, a pavement disease sensing and service state evaluation method is needed.
Disclosure of Invention
In order to solve the problems that the pavement is difficult to effectively and accurately detect and evaluate in the prior art, the invention provides a pavement disease sensing and service state evaluation method which can effectively sense and detect pavement diseases and evaluate service states.
In order to achieve the technical purpose, the invention provides the following technical scheme: a pavement disease sensing and service state evaluation method comprises the following steps:
acquiring image data and radar data, and identifying diseases by the image data and the radar data to obtain pavement disease data; and (3) counting pavement damage data distribution data, and carrying out different dimensional evaluation on the pavement service state based on pavement damage distribution to obtain a pavement service state evaluation result.
Optionally, the process of acquiring the image data and the radar data includes:
collecting image data through high definition image collecting equipment, and collecting radar data through a ground penetrating radar; wherein the image data corresponds to radar data.
Optionally, the process of identifying the disease on the image data includes:
and identifying the image data through a clustering algorithm and a deep learning network to obtain pavement surface disease data, wherein the pavement surface disease data comprises crack disease and water leakage disease, and counting the position and the range of the pavement surface disease data.
Optionally, the process of disease identification on radar data includes:
and carrying out segmentation processing on the radar data, and identifying through a deep learning model according to the segmented radar data to obtain pavement internal disease data, wherein the pavement internal disease data comprise internal cavity diseases and water damage diseases, and counting the positions and the ranges of the pavement internal disease data.
Optionally, the process of counting the pavement disease distribution data includes:
and extracting position and range data in the pavement disease data, and carrying out statistics integration on the position and range data to obtain pavement disease distribution.
Optionally, the process of evaluating the service state of the road surface in different dimensions includes:
dividing intervals of the transverse dimension and the longitudinal dimension of the detected pavement, carrying out weight distribution on different intervals of the transverse dimension and the longitudinal dimension, and analyzing pavement disease distribution according to a weight distribution result to obtain a total disease score, wherein the pavement disease distribution comprises internal disease distribution and surface disease distribution; and performing grade evaluation on the total disease score to obtain a pavement service state evaluation result.
Optionally, the process of analyzing the road surface disease distribution according to the weight distribution result includes:
judging the surface disease distribution according to the sections of the transverse dimension and the longitudinal dimension, judging to obtain the disease range in the section to which the surface disease distribution belongs, and calculating the disease range according to the weight distribution result to obtain the surface disease scores under different dimensions;
converting and predicting the internal disease distribution to obtain predicted surface disease distribution, judging the predicted surface disease distribution according to the sections of the transverse dimension and the longitudinal dimension to obtain a predicted disease range in the section of the predicted surface disease distribution, and calculating the predicted disease range according to the weight distribution result to obtain internal disease scores under different dimensions;
and carrying out weighted average calculation on the surface disease score and the internal disease score in different dimensions to obtain the total disease score.
Optionally, before the disease identification is performed on the image data, the method further comprises:
preprocessing the image data, and identifying diseases of the preprocessed image, wherein the preprocessing comprises image enhancement, gray scale processing and binary segmentation.
The invention has the following technical effects:
according to the technical scheme, the surface diseases and the internal diseases of the pavement can be effectively and accurately identified and detected through the deep learning algorithm, meanwhile, on the basis of identification and detection, the pavement is evaluated in different dimensions, so that the evaluation is more comprehensive, the service state of the pavement can be more effectively represented, reference data can be provided for highway management maintenance, management and safety according to the service state, subsequent maintenance management is facilitated, and the method has strong practicability.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions of the prior art, the drawings that are needed in the embodiments will be briefly described below, it being obvious that the drawings in the following description are only some embodiments of the present invention, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a schematic flow chart of a method according to an embodiment of the present invention;
fig. 2 is a schematic structural diagram of a YOLO series deep learning model for image data according to an embodiment of the present invention.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
As shown in FIG. 1, the invention provides a pavement disease sensing and service state evaluation method, which comprises the following steps:
collecting high-definition images through high-definition image collecting equipment aiming at the surface of a pavement, collecting radar signals through a ground penetrating radar aiming at the interior of the pavement, processing image data and radar data through a deep learning model and a clustering algorithm, generating pavement disease types and position ranges corresponding to the surface and the interior, counting the pavement disease types and the position ranges, generating relevant distribution, evaluating the relevant distribution from two transverse dimensions and longitudinal dimensions, and generating a final evaluation result.
The high-definition image acquisition equipment in the acquisition equipment adopts a Dali technology DLSC-RL series image acquisition system to acquire high-definition images, and carries out relevant pretreatment on the images after acquisition, including image enhancement, gray scale treatment and binary segmentation, and the binary images are identified through a deep learning model.
The high-definition image is identified through a clustering algorithm and a deep learning model, and specifically comprises the following steps: firstly, clustering high-definition images by a clustering algorithm, wherein the clustering algorithm adopts a K-meas clustering algorithm, the clustered images are input into a deep learning model for processing on the basis of clustering, crack diseases and water seepage diseases are identified, and positions and ranges of the diseases are marked on the images; the deep learning model used for the image data employs a YOLO series deep learning algorithm. As shown in fig. 2, the model structure includes a BACKBONE structure, a NECK structure and a HEAD structure connected in sequence, wherein the BACKBONE structure includes a FOCUS layer, a first CBL layer, a first CSP layer, a second CBL layer, a second CSP layer, a third CBL layer, a third CSP layer, a fourth CBL layer and a SPP layer connected in sequence; the NECK structure comprises a fourth CSP layer, a fifth CBL layer, a first upsampling layer, a first concat layer, a fifth CSP layer, a sixth CBL layer, a second upsampling layer, a second concat layer and a sixth CSP layer which are sequentially connected, and the structure outputs through a convolution layer in the HEAD structure; the other end input of the first concat layer is connected with the output of the third CSP layer; the other end input of the second concat layer is connected with the output of the second CSP layer; the NECK structure also comprises a third concat layer and a seventh CSP layer which are sequentially connected, wherein the input of the third concat layer is connected with the output of the sixth CBL layer, the input of the other end of the third concat layer is connected with the output of the sixth CSP layer through the CBL layer, and the seventh CSP layer outputs through a convolution layer in the HEAD structure; the NECK structure also comprises a fourth concat layer and an eighth CSP layer which are sequentially connected, wherein the input of the fourth concat layer is connected with the output of the fifth CBL layer, the input of the other end of the fourth concat layer is connected with the output of the seventh CSP layer through the CBL layer, and the eighth CSP layer outputs through a convolution layer in the HEAD structure; the model is trained by collecting images, the trained deep learning model is used for identifying high-definition images, the surface diseases are identified and generated, and after the surface diseases in different image data are identified, the positions and the ranges of the surface disease data identified by the images are counted.
The radar data corresponds to the image data, and the data under the same road section is identified by the radar data through the deep learning network, and specifically comprises the following steps: the radar data is sampled, the radar data is segmented into signals with the same length, the segmented signals are identified through a deep learning model, a YOLO deep learning algorithm is adopted for the deep learning model used for the radar signals, radar image characteristics are analyzed through forward modeling and indoor inversion experiments of a finite time domain difference method before identification, simulation results are compared with actual measurement results to determine abnormal characteristics, a radar database is built, the deep learning model is trained through the radar database, a trained deep learning model is generated, the radar signals are segmented in the same length in the time domain before training is carried out by using the trained deep learning model, the segmented radar signals are input into the deep learning model for identification, the input length of an input layer in the deep learning model corresponds to the segmented length of the radar signals, identification of the radar signals is achieved, and deep internal diseases are identified. After identifying the split radar signals, the internal disease locations and ranges in the radar signals are counted.
And extracting the positions and the ranges in the counted pavement disease data, wherein in the acquisition process, the positions of different road sections are corresponding to different time points, so that the position and the range data under the same disease category are combined and spliced according to time sequence to form the position and the range data of different types of diseases in the whole pavement, namely pavement disease distribution.
After the road surface disease distribution is obtained, the road surface is evaluated, and in the evaluation process: firstly dividing the intervals of the transverse dimension and the longitudinal dimension of the road surface, wherein the transverse dimension corresponds to the width of the road, under the condition of the road width, different points of the road width correspond to the running positions of vehicles, the influence of road surface diseases on the vehicles on the road width is normally distributed, the influence of the road surface diseases on the running of the vehicles is analyzed from the transverse dimension, the length of the road corresponds to the longitudinal dimension, the density of the large-scale diseases is analyzed from the longitudinal dimension,
dividing the road width transverse dimension into a plurality of sections at equal intervals, and distributing weights conforming to normal distribution for different sections, for example, dividing the road width transverse dimension into 7 sections, wherein the transverse section weights are respectively 0.05, 0.1, 0.15, 0.2, 0.15, 0.1 and 0.05, and the transverse sections and the weights can also be adjusted according to manual experience and actual conditions;
in the longitudinal interval weight distribution process, setting initial longitudinal weight which is set to 0.1, wherein the longitudinal interval and the weight can be adjusted according to manual experience and actual conditions; after the setting is finished, when the road surface interval of which the road surface disease distribution range is larger than a certain threshold value and larger than the threshold value is continuous, the initial longitudinal weight of the corresponding interval is increased by a fixed value, and the threshold value is 40% of the range of the interval, and can be adjusted according to manual experience and actual conditions; according to the number of times that the pavement damage interval is continuously larger than the threshold value, the weight of the corresponding interval is lifted to realize longitudinal weight distribution, for example, the pavement damage distribution range in one interval is larger than the threshold value, the weight of the two intervals is lifted to be a value obtained by adding one fixed value to the initial weight, if the pavement damage distribution range in the next interval is still larger than the threshold value, the weight of the three intervals is lifted to be the initial weight and added to be two fixed values, and the like until the pavement damage distribution range is smaller than or equal to the threshold value, so that the continuous interval weight is the interval continuous number times multiplied by the fixed value plus the initial longitudinal weight, wherein the fixed value can be set to be twenty times of the initial longitudinal weight.
After the weight rules of different sections are set, calculating the road surface disease distribution according to the different sections, firstly, carrying out statistics judgment on the road surface disease distribution according to the transverse dimension sections, carrying out statistics judgment on the range of the road surface disease distribution in the different transverse dimension sections, and carrying out product calculation on the value of the range after statistics and the value of the weight in the section to obtain a transverse surface disease score; and carrying out statistics judgment on the road table disease distribution according to the longitudinal dimension intervals, carrying out statistics judgment on the range of the road table disease distribution under the intervals with different longitudinal dimensions, searching the continuous intervals according to the rules of the continuous intervals, modifying the weights in the intervals, and obtaining the disease scores of different types and different dimensions of the surface according to the weights and the disease ranges under the intervals.
The method comprises the steps of calculating a disease score corresponding to internal disease distribution in parallel, firstly converting and predicting the internal disease distribution, converting the internal disease distribution into surface disease distribution, predicting the internal disease distribution through a convolutional neural network model, acquiring relevant training data through historical cases of converting the internal disease into the surface disease and relevant data in a database, training the convolutional neural network through the training data, predicting the internal disease distribution through a trained network model, and calculating the predicted surface disease distribution according to a weight rule.
The predicted surface disease distribution is the same as the actual surface disease distribution in calculation mode, firstly, the predicted road surface disease distribution is statistically judged according to a transverse dimension interval, the predicted road surface disease distribution range under different transverse dimension intervals is statistically judged, and the product calculation is carried out on the numerical value of the counted range and the numerical value of the weight under the interval to obtain the transverse surface disease score; and carrying out statistics judgment on the road table disease distribution according to the longitudinal dimension intervals, carrying out statistics judgment on the road table disease distribution ranges under the intervals with different longitudinal dimensions, searching the continuous intervals according to the rules of the continuous intervals, modifying the weights in the intervals, obtaining predicted road table disease scores with different types and different dimensions according to the weights and the disease ranges under the intervals, and taking the scores as internal disease scores under different dimensions.
The surface and internal disease scores are subjected to different dimension weighted average calculation, wherein different calculation weights are set for different dimension values of different types, the surface disease total score is calculated firstly, specifically, the crack disease and the water leakage disease in the surface disease score are subjected to weighted average calculation, the crack disease score weight and the water leakage disease are respectively set to 0.6 and 0.4, the weight can also be adjusted according to manual experience and actual conditions to generate final surface disease total scores in different dimensions, the cavity disease and the water damage disease in the internal disease score are respectively set to 0.6 and 0.4, the cavity disease weight and the water damage disease weight can also be adjusted according to manual experience and actual conditions to generate final internal disease total scores in different dimensions, the weight distribution is carried out on different dimensions, the transverse weight and the longitudinal weight are respectively set to 0.5, and the final surface disease total scores and the internal total scores in different dimensions are subjected to weighted average calculation to obtain the disease total scores.
Under the condition that the whole road surface has the diseases, calculating final total surface disease scores, internal disease scores and total disease scores under different dimensions, taking the total whole road surface disease scores obtained by calculation in the process as a grading base, and taking 10%, 20% and 30% of the total whole road surface disease scores as grading points to grade 4 grades, wherein the first grade is below 10% and indicates that the road surface is normal, only detection is needed, and treatment and trimming are not needed; the second grade is 10% -20%, which indicates that the road surface is slightly damaged, the road surface needs to be detected, and small-range treatment and modification are carried out; the third grade is 20% -30%, which indicates that the road surface is high in disease, the road surface needs to be detected, treated and modified, and refurbished if necessary; and the fourth grade is more than 30%, which indicates that the road surface is very ill, indicates that the road surface needs to be detected urgently, carries out large-scale treatment and modification, and judges whether the road surface needs to be renovated.
Then after determining that the total disease score of the whole domain is judged, grading the total surface disease score and the total internal disease score of different dimensions in different grades, wherein the total surface disease score and the total internal disease score of different dimensions are used as grading points according to 10%, 20% and 30%, the disease grades are sequentially graded into slight, normal, higher and very high, the slight, normal, higher and very high in the transverse dimension correspond to slight, normal, higher and very high influences on driving influence respectively, and the detection, small-range treatment, large-range treatment and renovation treatment are respectively corresponding to the influences of the grades; different grades in the longitudinal dimension correspond to slight, normal, higher and very high influence of the density degree respectively, and correspond to detection, small-range treatment, large-range treatment and renovation treatment respectively. Through evaluating the grades, and carrying out relevant treatment modes aiming at different grades, the road is effectively treated, the relevant driving safety and road flatness can be effectively ensured, and the method has strong practicability.
The foregoing has shown and described the basic principles, principal features and advantages of the invention. It will be understood by those skilled in the art that the present invention is not limited to the embodiments described above, and that the above embodiments and descriptions are merely illustrative of the principles of the present invention, and various changes and modifications may be made without departing from the spirit and scope of the invention, which is defined in the appended claims. The scope of the invention is defined by the appended claims and equivalents thereof.
Claims (7)
1. The pavement disease sensing and service state evaluation method is characterized by comprising the following steps of:
acquiring image data and radar data, and identifying diseases by the image data and the radar data to obtain pavement disease data; counting the pavement damage data distribution data, and carrying out different dimensional evaluation on the pavement service state based on pavement damage distribution to obtain a pavement service state evaluation result;
the process for analyzing the road surface disease distribution according to the weight distribution result comprises the following steps:
judging the surface disease distribution according to the sections of the transverse dimension and the longitudinal dimension, judging to obtain the disease range in the section to which the surface disease distribution belongs, and calculating the disease range according to the weight distribution result to obtain the surface disease scores under different dimensions;
converting and predicting the internal disease distribution to obtain predicted surface disease distribution, judging the predicted surface disease distribution according to the sections of the transverse dimension and the longitudinal dimension to obtain a predicted disease range in the section of the predicted surface disease distribution, and calculating the predicted disease range according to the weight distribution result to obtain internal disease scores under different dimensions;
carrying out weighted average calculation on the surface disease score and the internal disease score in different dimensions to obtain a total disease score;
the specific process for obtaining the surface disease score under different dimensions comprises the following steps:
dividing the road width transverse dimension into a plurality of sections at equal intervals, carrying out weight distribution conforming to normal distribution aiming at different sections, dividing the road width transverse dimension into 7 sections, wherein the transverse section weights are respectively 0.05, 0.1, 0.15, 0.2, 0.15, 0.1 and 0.05, and the transverse sections and the weights can be adjusted according to manual experience and actual conditions;
in the longitudinal interval weight distribution process, setting initial longitudinal weight which is set to 0.1, wherein the longitudinal interval and the weight can be adjusted according to manual experience and actual conditions; after the setting is finished, when the road surface interval of which the road surface disease distribution range is larger than a certain threshold value and larger than the threshold value is continuous, the initial longitudinal weight of the corresponding interval is increased by a fixed value, and the threshold value is 40% of the range of the interval, and can be adjusted according to manual experience and actual conditions; according to the number of times that the pavement damage interval is continuously larger than the threshold value, the weight of the corresponding interval is lifted to realize longitudinal weight distribution, the pavement damage distribution range in one interval is larger than the threshold value, the weight of the two intervals is lifted to be a value obtained by adding a fixed value to the initial weight, if the pavement damage distribution range in the next interval is still larger than the threshold value, the weight of the three intervals is lifted to be the initial weight and added to be two fixed values, and the like until the pavement damage distribution range is smaller than or equal to the threshold value, so that the continuous interval weight is the interval continuous number times multiplied by the fixed value plus the initial longitudinal weight, wherein the fixed value can be set to be twenty times of the initial longitudinal weight;
after the weight rules of different sections are set, calculating the road surface disease distribution according to the different sections, firstly, carrying out statistics judgment on the road surface disease distribution according to the transverse dimension sections, carrying out statistics judgment on the range of the road surface disease distribution in the different transverse dimension sections, and carrying out product calculation on the value of the range after statistics and the value of the weight in the section to obtain a transverse surface disease score; and carrying out statistics judgment on the road table disease distribution according to the longitudinal dimension intervals, carrying out statistics judgment on the range of the road table disease distribution under the intervals with different longitudinal dimensions, searching the continuous intervals according to the rules of the continuous intervals, modifying the weights in the intervals, and obtaining the disease scores of different types and different dimensions of the surface according to the weights and the disease ranges under the intervals.
2. The method according to claim 1, characterized in that:
the process of acquiring image data and radar data includes:
collecting image data through high definition image collecting equipment, and collecting radar data through a ground penetrating radar; wherein the image data corresponds to radar data.
3. The method according to claim 1, characterized in that:
the disease identification process for the image data comprises the following steps:
and identifying the image data through a clustering algorithm and a deep learning network to obtain pavement surface disease data, wherein the pavement surface disease data comprises crack disease and water leakage disease, and counting the position and the range of the pavement surface disease data.
4. The method according to claim 1, characterized in that:
the disease identification process for the radar data comprises the following steps:
and carrying out segmentation processing on the radar data, and identifying through a deep learning model according to the segmented radar data to obtain pavement internal disease data, wherein the pavement internal disease data comprise internal cavity diseases and water damage diseases, and counting the positions and the ranges of the pavement internal disease data.
5. The method according to claim 1, characterized in that:
the process for counting the pavement disease data distribution data comprises the following steps:
and extracting position and range data in the pavement disease data, and carrying out statistics integration on the position and range data to obtain pavement disease distribution.
6. The method according to claim 1, characterized in that:
the process for evaluating the service state of the road surface in different dimensions comprises the following steps:
dividing intervals of the transverse dimension and the longitudinal dimension of the detected pavement, carrying out weight distribution on different intervals of the transverse dimension and the longitudinal dimension, and analyzing pavement disease distribution according to a weight distribution result to obtain a total disease score, wherein the pavement disease distribution comprises internal disease distribution and surface disease distribution; and performing grade evaluation on the total disease score to obtain a pavement service state evaluation result.
7. The method according to claim 1, characterized in that:
before the disease identification is carried out on the image data, the method further comprises the following steps:
preprocessing the image data, and identifying diseases of the preprocessed image, wherein the preprocessing comprises image enhancement, gray scale processing and binary segmentation.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211304508.1A CN115620041B (en) | 2022-10-24 | 2022-10-24 | Pavement disease sensing and service state evaluation method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211304508.1A CN115620041B (en) | 2022-10-24 | 2022-10-24 | Pavement disease sensing and service state evaluation method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115620041A CN115620041A (en) | 2023-01-17 |
| CN115620041B true CN115620041B (en) | 2023-08-29 |
Family
ID=84864520
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211304508.1A Active CN115620041B (en) | 2022-10-24 | 2022-10-24 | Pavement disease sensing and service state evaluation method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115620041B (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2012184624A (en) * | 2011-03-08 | 2012-09-27 | Geo Search Co Ltd | Evaluation method for soundness of pavement |
| CN108572248A (en) * | 2018-04-16 | 2018-09-25 | 长沙理工大学 | A method of based on non-destructive testing technology evaluation path technology status |
| CN114266892A (en) * | 2021-12-20 | 2022-04-01 | 江苏燕宁工程科技集团有限公司 | Pavement disease identification method and system for multi-source data deep learning |
| CN115017979A (en) * | 2022-05-23 | 2022-09-06 | 广东华路交通科技有限公司 | Pavement condition evaluation method for semi-rigid base asphalt pavement |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| DE102018203684A1 (en) * | 2018-03-12 | 2019-09-12 | Zf Friedrichshafen Ag | Identification of objects using radar data |
-
2022
- 2022-10-24 CN CN202211304508.1A patent/CN115620041B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2012184624A (en) * | 2011-03-08 | 2012-09-27 | Geo Search Co Ltd | Evaluation method for soundness of pavement |
| CN108572248A (en) * | 2018-04-16 | 2018-09-25 | 长沙理工大学 | A method of based on non-destructive testing technology evaluation path technology status |
| CN114266892A (en) * | 2021-12-20 | 2022-04-01 | 江苏燕宁工程科技集团有限公司 | Pavement disease identification method and system for multi-source data deep learning |
| CN115017979A (en) * | 2022-05-23 | 2022-09-06 | 广东华路交通科技有限公司 | Pavement condition evaluation method for semi-rigid base asphalt pavement |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115620041A (en) | 2023-01-17 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108257114A (en) | A kind of transmission facility defect inspection method based on deep learning | |
| CN110675036B (en) | Plateau river ecological health evaluation method based on random forest optimization microbial index | |
| CN114169548B (en) | BIM-based highway bridge management and maintenance PHM system and method | |
| CN111241994B (en) | Deep learning remote sensing image rural highway sanded road section extraction method | |
| CN114675261B (en) | Urban underground disease body full life cycle management method based on ground penetrating radar | |
| CN114581764B (en) | Underground structure crack disease discriminating method based on deep learning algorithm | |
| CN113918538B (en) | New road maintenance data migration system based on artificial neural network | |
| CN116168356B (en) | Vehicle damage judging method based on computer vision | |
| CN115879343B (en) | Ballastless track self-compacting concrete void two-stage identification method | |
| CN115841466A (en) | Automatic quantitative assessment method for defects of drainage pipe network | |
| CN111353640B (en) | Method for constructing wind speed prediction model by combination method | |
| CN113640380A (en) | Multi-stage classification method and system for rail damage detection | |
| CN115169401A (en) | Cutter abrasion loss prediction method based on multi-scale DenseNet-ResNet-GRU model | |
| CN115271204A (en) | Pavement performance dynamic prediction method based on high-frequency intelligent patrol data | |
| CN115204688A (en) | Comprehensive evaluation method for health of drainage system | |
| CN111724074A (en) | Pavement lesion detection early warning method and system based on deep learning | |
| CN109632942B (en) | Inversion method of pipeline defect size based on ensemble learning | |
| CN115620041B (en) | Pavement disease sensing and service state evaluation method | |
| CN115063337A (en) | Intelligent maintenance decision-making method and device for buried pipeline | |
| CN115438547A (en) | Overall evaluation method and system based on pavement service state | |
| CN116681330A (en) | Road tunnel electromechanical system running state classification and comprehensive evaluation method | |
| CN113569491B (en) | Wheel set size detection data analysis and correction method and device | |
| CN115290057A (en) | Channel software row state monitoring system and method | |
| CN115345846A (en) | Intelligent grading method and system for grain size of medium and low carbon steel | |
| CN114492507A (en) | Method for predicting residual life of bearing under digital-analog cooperative driving |
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 |