CN117250201A - Swimming pool brick layer defect detection method and system - Google Patents
Swimming pool brick layer defect detection method and system Download PDFInfo
- Publication number
- CN117250201A CN117250201A CN202311473166.0A CN202311473166A CN117250201A CN 117250201 A CN117250201 A CN 117250201A CN 202311473166 A CN202311473166 A CN 202311473166A CN 117250201 A CN117250201 A CN 117250201A
- Authority
- CN
- China
- Prior art keywords
- defect
- sample
- detection
- water level
- brick layer
- 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.)
- Granted
Links
- 230000007547 defect Effects 0.000 title claims abstract description 522
- 238000001514 detection method Methods 0.000 title claims abstract description 231
- 239000011449 brick Substances 0.000 title claims abstract description 135
- 230000009182 swimming Effects 0.000 title claims abstract description 105
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 claims abstract description 118
- 238000004458 analytical method Methods 0.000 claims abstract description 107
- 238000000034 method Methods 0.000 claims abstract description 52
- 238000001579 optical reflectometry Methods 0.000 claims abstract description 45
- 230000003287 optical effect Effects 0.000 claims description 46
- 238000012549 training Methods 0.000 claims description 29
- 238000012360 testing method Methods 0.000 claims description 25
- 238000012423 maintenance Methods 0.000 claims description 21
- 238000013507 mapping Methods 0.000 claims description 10
- 238000003066 decision tree Methods 0.000 claims description 9
- 238000013527 convolutional neural network Methods 0.000 claims description 7
- 238000012544 monitoring process Methods 0.000 claims description 7
- 238000005336 cracking Methods 0.000 claims description 5
- 238000010801 machine learning Methods 0.000 claims description 5
- 238000012545 processing Methods 0.000 abstract description 11
- 230000008569 process Effects 0.000 description 6
- 230000000694 effects Effects 0.000 description 5
- 238000005516 engineering process Methods 0.000 description 5
- 238000004422 calculation algorithm Methods 0.000 description 3
- 238000007796 conventional method Methods 0.000 description 3
- 238000007689 inspection Methods 0.000 description 3
- 238000013145 classification model Methods 0.000 description 2
- 230000002950 deficient Effects 0.000 description 2
- 238000011156 evaluation Methods 0.000 description 2
- 230000036541 health Effects 0.000 description 2
- 239000011159 matrix material Substances 0.000 description 2
- 230000008439 repair process Effects 0.000 description 2
- 238000012706 support-vector machine Methods 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000012407 engineering method Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012806 monitoring device Methods 0.000 description 1
- 230000000737 periodic effect Effects 0.000 description 1
- 238000007637 random forest analysis Methods 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 238000011179 visual inspection Methods 0.000 description 1
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
-
- 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
- G01N2021/8883—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 involving the calculation of gauges, generating models
-
- 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
Landscapes
- Chemical & Material Sciences (AREA)
- Biochemistry (AREA)
- Pathology (AREA)
- Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Immunology (AREA)
- Analytical Chemistry (AREA)
- General Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Signal Processing (AREA)
- Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)
Abstract
The invention provides a swimming pool brick layer defect detection method and a swimming pool brick layer defect detection system, which relate to the technical field of data processing and comprise the following steps: extracting to obtain minimum water level information and maximum water level information, carrying out primary prediction on the defects of the brick layer to obtain a primary defect analysis result, wherein the primary defect analysis result comprises defect grades, matching and obtaining defect detection scale when the defect grade is larger than 0, dividing and optically detecting detection areas to obtain a plurality of light reflectivity information, carrying out secondary defect analysis on the brick layer to obtain a plurality of defect areas, analyzing and obtaining image detection scale, dividing and image acquisition on the detection areas to obtain a plurality of local images, carrying out tertiary defect analysis on the brick layer to obtain a plurality of defect information, and taking the plurality of defect information as the brick layer defect detection result of the target swimming pool. The invention solves the technical problems that the traditional method can only detect obvious defects generally, is difficult to find hidden or tiny defects, and is subjective in judging the defect grade of the brick layer of the swimming pool, so that the detection result is low in accuracy and poor in comprehensiveness.
Description
Technical Field
The invention relates to the technical field of data processing, in particular to a swimming pool brick layer defect detection method and system.
Background
In the use process of the swimming pool, due to various reasons, such as long-term use, environmental change and the like, the brick layer of the swimming pool can possibly have the defects of cracks, falling off, damage and the like, and the timely discovery and repair of the defects are very important to ensure the safety of the swimming pool, prevent water leakage and prolong the service life of the swimming pool.
However, there are some problems in the conventional method for detecting defects of brick layers of swimming pools, firstly, the conventional method generally requires professional technicians to perform manual detection, and a great deal of time and manpower resources are consumed, so that the efficiency is low and the cost is high; secondly, the judgment of the defect level by the traditional method often depends on subjective opinion of a detector, and misjudgment or omission is easy to cause; third, some minor defects may be difficult to detect by visual inspection, and conventional methods have difficulty in effectively detecting these minor problems.
Therefore, in order to solve the above problems and improve the efficiency and accuracy of detecting defects of a brick layer of a swimming pool, a new defect detecting method is needed to achieve the improvement of the efficiency and accuracy of detection, and minute defects can be found to ensure the safety and normal use of the swimming pool.
Disclosure of Invention
The application provides a swimming pool brick layer defect detection method and system, and aims to solve the technical problems that the traditional method can only detect obvious defects generally, is difficult to find hidden or tiny defects, is subjective in defect grade judgment of a swimming pool brick layer, and is low in accuracy and poor in comprehensiveness of detection results.
In view of the above, the present application provides a method and a system for detecting defects of a brick layer of a swimming pool.
In a first aspect of the disclosure, a method for detecting defects of a brick layer of a swimming pool is provided, the method is applied to a device for detecting defects of a brick layer of a swimming pool, the device comprises a water level testing module, a light detecting module, an image detecting module and a defect analyzing module, and the method comprises: detecting average water level information of a target swimming pool in a plurality of preset time windows through a water level testing module, and extracting and obtaining minimum water level information and maximum water level information; performing primary prediction on the brick layer defects based on the minimum water level information and the maximum water level information to obtain primary defect analysis results, wherein the primary defect analysis results comprise defect grades; when the defect grade in the primary defect analysis result is greater than 0, matching to obtain a defect detection scale, and dividing and optically detecting a detection area in the target swimming pool according to the defect detection scale by the optical detection module to obtain a plurality of light reflectivity information; performing secondary defect analysis on the brick layer according to the light reflectivity information to obtain a plurality of defect areas, and analyzing and obtaining an image detection scale according to the defect area number and the defect grade of the defect areas; dividing detected local areas and acquiring images in the defect areas according to the image detection scale by an image detection module to obtain a plurality of local images; and performing three-time defect analysis on the brick layers of the plurality of local images through the defect analysis module to obtain a plurality of defect information which is used as a brick layer defect detection result of the target swimming pool.
In another aspect of the disclosure, a swimming pool tile layer defect detection system is provided, the system is applied to a swimming pool tile layer defect detection device, the device includes a water level test module, a light detection module, an image detection module and a defect analysis module, the system is used in the above method, the system includes: the water level information acquisition unit is used for detecting the average water level information of the target swimming pool in a plurality of preset time windows through the water level test module, and extracting and obtaining the minimum water level information and the maximum water level information; the defect primary prediction unit is used for carrying out primary prediction on the brick layer defects based on the minimum water level information and the maximum water level information to obtain primary defect analysis results, wherein the primary defect analysis results comprise defect grades; the regional optical detection unit is used for matching and acquiring a defect detection scale when the defect grade in the primary defect analysis result is greater than 0, and dividing and optically detecting detection regions in the target swimming pool according to the defect detection scale by the optical detection module to acquire a plurality of light reflectivity information; the secondary defect analysis unit is used for performing secondary defect analysis on the brick layer according to the light reflectivity information to obtain a plurality of defect areas, and analyzing and obtaining an image detection scale according to the defect area number and the defect grade of the defect areas; the regional image acquisition unit is used for dividing detected local areas and acquiring images in the plurality of defect areas according to the image detection scale by the image detection module to obtain a plurality of local images; and the third defect analysis unit is used for performing third defect analysis on the brick layers of the plurality of local images through the defect analysis module to obtain a plurality of defect information which is used as a brick layer defect detection result of the target swimming pool.
One or more technical solutions provided in the present application have at least the following technical effects or advantages:
the comprehensive swimming pool brick layer defect detection device is provided by comprehensively utilizing the water level test module, the light detection module, the image detection module and the defect analysis module, so that the detection of the swimming pool brick layer defect can be completed in a short time, and the working efficiency is improved; the minimum and maximum water level information is obtained through the water level testing module, and the optical detection and image processing technology is combined to perform multiple defect analysis and comprehensive consideration, so that the defect grade can be accurately judged, the defect information is obtained, and the accuracy of defect analysis is improved; through multiple defect analysis, including light reflectivity information and local image acquisition and analysis, various defect information of the swimming pool brick layer, including obvious and tiny defects, can be comprehensively detected and obtained, and the comprehensiveness and accuracy of detection are improved. In summary, by introducing the automation equipment and the image processing technology, the swimming pool brick layer defect detection method solves the technical problems of low detection efficiency, inaccurate defect analysis, lack of comprehensiveness and the like in the traditional method, and achieves the technical effects of high efficiency, high speed, accuracy, reliability, comprehensiveness and delicacy.
The foregoing description is only an overview of the technical solutions of the present application, and may be implemented according to the content of the specification in order to make the technical means of the present application more clearly understood, and in order to make the above-mentioned and other objects, features and advantages of the present application more clearly understood, the following detailed description of the present application will be given.
Drawings
Fig. 1 is a schematic flow chart of a method for detecting defects of a brick layer of a swimming pool according to an embodiment of the present application;
fig. 2 is a schematic structural diagram of a swimming pool brick layer defect detecting system according to an embodiment of the present application.
Reference numerals illustrate: the system comprises a water level information acquisition unit 10, a defect primary prediction unit 20, a region optical detection unit 30, a secondary defect analysis unit 40, a region image acquisition unit 50 and a tertiary defect analysis unit 60.
Detailed Description
According to the swimming pool brick layer defect detection method, the technical problems that the traditional method can only detect obvious defects generally, hidden or tiny defects are difficult to find, and the defect grade judgment of the swimming pool brick layer is subjective, so that the accuracy of detection results is low and the comprehensiveness is poor are solved.
Having described the basic principles of the present application, various non-limiting embodiments of the present application will now be described in detail with reference to the accompanying drawings.
Example 1
As shown in fig. 1, an embodiment of the present application provides a method for detecting a defect of a brick layer of a swimming pool, the method is applied to a device for detecting a defect of a brick layer of a swimming pool, the device includes a water level testing module, a light detecting module, an image detecting module and a defect analyzing module, and the method includes:
detecting average water level information of a target swimming pool in a plurality of preset time windows through a water level testing module, and extracting and obtaining minimum water level information and maximum water level information;
the swimming pool brick layer defect detection method is applied to a swimming pool brick layer defect detection device, and the device comprises a water level test module, a light detection module, an image detection module and a defect analysis module.
The water level testing module is correctly installed in a target swimming pool, a testing time range is set according to actual conditions and specific requirements, the time range is divided into a plurality of preset time windows to detect water level changes in different time periods, for example, the time for filling the swimming pool is a complete time range, the time range is divided into ten parts, and each part is a preset time window.
And in each preset time window, measuring and recording water level information in real time through a water level testing module, obtaining water level data of the swimming pool, and carrying out average calculation on the water level data recorded in each preset time window when each preset time window is finished, so as to obtain average water level information in the time window. On account of the fluctuation influence of the water level caused by the defect of the brick layer, the fluctuation condition of the average water level is observed in each time window, and if the fluctuation of the water level obviously exceeds the normal range, the defect of the brick layer can exist.
By comparing the average water level over each time window, minimum water level information is identified, the minimum water level representing the lowest water level of the pool over a given time period, and the maximum water level representing the highest water level of the pool over the given time period.
Performing primary prediction on the brick layer defects based on the minimum water level information and the maximum water level information to obtain primary defect analysis results, wherein the primary defect analysis results comprise defect grades;
different defect grades are defined according to the swimming pool use standard or industry standard, such as slight, medium and serious, a defect primary classifier is adopted for classifying and identifying according to the minimum water level information and the maximum water level information, a primary defect analysis result is obtained, the result comprises the number and the size of brick layer defects in the swimming pool, the proper grade is matched in the defined defect grades according to the number and the size of the brick layer defects, for example, the number and the size exceed the serious threshold, and the defect grade is rated as serious.
Further, a primary defect analysis result is obtained, and the method includes:
acquiring a sample minimum water level information record and a sample maximum water level information record based on the water level monitoring information in the target swimming pool history time;
According to the brick layer maintenance record in the history time of the target swimming pool, a sample primary defect analysis result record is obtained, and the sample primary defect analysis result is determined according to the number and the size of the brick layer defects in the target swimming pool;
the sample minimum water level information record and the sample maximum water level information record are adopted as input, the sample primary defect analysis result record is adopted as output, a defect primary classifier is constructed, and training and updating are carried out;
and classifying and identifying the minimum water level information and the maximum water level information by adopting the updated defect primary classifier to obtain the primary defect analysis result.
Water level monitoring data is obtained over a historical time frame of the target pool, and may be from sensors, water level gauges or other water level monitoring devices, and the water level monitoring data is consolidated into a form suitable for processing, such as by storing it in a data table, with each record containing water level values and corresponding time stamps. Traversing the water level monitoring data, finding out the minimum water level value and the maximum water level value, and recording the found minimum water level value and the found maximum water level value and the corresponding time stamp as a sample minimum water level and a sample maximum water level record.
Brick layer maintenance records are obtained over a historical time frame of the target swimming pool, which records may include evaluations, repairs, or replacement of brick layer defects during periodic inspection or maintenance. The brick layer maintenance records are also arranged into a form suitable for processing, for example, the brick layer maintenance records are stored in a data table, each record is ensured to contain the number and the size of defects and corresponding time stamps, the result of one-time defect analysis of the sample is determined according to the number and the size of the defects of the sample, and specific judgment standards can be defined according to actual conditions, for example, the health degree grade of the brick layer is divided according to the threshold value of the number and the area of the defects.
And aligning the water level monitoring information list with the brick layer maintenance record list according to the time stamp, so that the minimum water level information record of the sample and the maximum water level information record of the sample are in one-to-one correspondence with the one-time defect analysis result record of the sample.
And sorting the sample minimum water level information record and the sample maximum water level information record to serve as input data of the model, sorting the sample primary defect analysis result record to serve as output data, and associating each data sample with a corresponding sample primary defect analysis result to form a construction data set. The constructed data set is divided into a training set and a testing set, wherein the training set is used for training and parameter updating of the model, and the testing set is used for evaluating the performance and tuning of the model.
And selecting a proper machine learning algorithm to construct a classifier, wherein the classifier comprises a decision tree, a random forest, a Support Vector Machine (SVM) and the like, training the classifier by using a training set, and updating parameters according to a training result to ensure that the model can accurately learn the relation between a sample minimum water level information record and a sample maximum water level information record and a primary defect analysis result. The performance of the trained classifier is evaluated by using the test set, tuning is performed according to the evaluation result, such as algorithm parameter adjustment, different characteristic engineering methods try, and the like, so as to obtain the defect primary classifier meeting the requirements, and the classifier can help to automate the defect analysis process and provide quick and accurate judgment and decision support.
Inputting the minimum water level information and the maximum water level information into a updated defect primary classifier, predicting the minimum water level information and the maximum water level information by using the classifier, giving a classification label of a primary defect analysis result according to a mode and association learned before by the classifier, and obtaining a primary defect analysis result according to an output label of the classifier, wherein the result represents the health degree or defect condition of a swimming pool brick layer. Therefore, the automatic defect analysis process can be realized, and a time-accurate judgment basis is provided for the maintenance and management of the swimming pool.
When the defect grade in the primary defect analysis result is greater than 0, matching to obtain a defect detection scale, and dividing and optically detecting a detection area in the target swimming pool according to the defect detection scale by the optical detection module to obtain a plurality of light reflectivity information;
and under the condition that the defect grade is larger than 0, further matching the defect detection scale, and matching the corresponding defect detection scale according to the defect grade in the primary defect analysis result, wherein the defect detection scale is the scale information of the area size of each light reflection test, and the larger the defect grade is, the more defects possibly showing that the brick layer is problematic are larger in scale, and the smaller defect detection scale is required, so that the region division and the optical detection are more finely carried out.
The areas in the target swimming pool are divided according to the selected defect detection scale, and the areas can be divided based on a grid and uniform sampling method, so that the size of each area is ensured to be in accordance with the defect detection scale, and the method is suitable for optical detection. The optical detection module is used for carrying out optical detection on each divided area, the optical detection module can use an optical sensor or a camera to capture light reflectivity information, the light reflectivity information of each area is obtained through optical detection, and the existence of defects can lead to uneven and discontinuous surfaces of the swimming pool tiles so as to generate changes in light reflectivity reflection intensity. According to the obtained light reflectivity information, the possible positions of the defects of the brick layer can be identified by comparing the light reflectivity differences between different areas, and then the defective areas are obtained.
Further, the defect detection scale is obtained, the method comprising:
judging whether the defect grade in the primary defect analysis result is greater than 0;
if not, finishing the detection of the brick layer defect, and if so, carrying out mapping matching in a defect detection scale matcher according to the defect grade to obtain the defect detection scale, wherein the defect detection scale matcher comprises mapping relations of a plurality of sample defect grades and a plurality of sample defect detection scales, each defect detection scale comprises the size of a region for carrying out optical detection, and the size of the defect detection scale is inversely related to the size of the defect grade.
Checking the defect grade of the primary defect analysis result given by the classifier, judging whether the grade is greater than 0, and if the defect grade is greater than 0, indicating that the brick layer defect exists; otherwise, it indicates that no brick layer defect is detected. If the defect grade is less than or equal to 0, the swimming pool brick layer is considered to be in a normal state, and no defects requiring further treatment exist, and the brick layer defect detection process can be finished.
If the defect grade is greater than 0, further detection of brick layer defects is required.
A defect detection scale matcher is trained in advance, and comprises a plurality of mapping relations of sample defect levels and corresponding sample defect detection scales, wherein the mapping relations are used for converting the defect levels into the corresponding defect detection scales. And according to the defect grade in the primary defect analysis result, mapping and matching are carried out by using a defect detection scale matcher, and the corresponding defect detection scale is obtained by searching the mapping relation between the defect grade and the defect detection scale. This scale represents the size of the area where the light detection is performed and is inversely related to the size of the defect level, that is, as the defect level increases, the defect detection scale becomes smaller because the larger the defect level, the more defects that indicate problems with the brick layer are likely to be and the larger the scale, the smaller the defect detection scale is required for finer area division and optical detection.
Performing secondary defect analysis on the brick layer according to the light reflectivity information to obtain a plurality of defect areas, and analyzing and obtaining an image detection scale according to the defect area number and the defect grade of the defect areas;
for each detection area, analyzing the corresponding light reflectivity information, identifying the area with the brick layer defect by comparing the light reflectivity differences among different areas, and determining a plurality of defect areas in the target swimming pool based on the analysis result of the light reflectivity information, wherein the areas correspond to the problem areas on the brick layer and need further inspection and processing.
And further analyzing according to the obtained multiple defect areas and corresponding defect grades, and evaluating the overall defect condition according to the number of the defect areas and the defect grade. By analyzing the number and defect levels of the plurality of defect areas, an image detection scale is deduced, and the same as the previous defect detection scale, more defect areas and higher defect levels correspond to smaller image detection scales, so that area division and image detection can be performed more finely.
Further, a plurality of defect areas are obtained, the method comprising:
Obtaining a plurality of sample light reflectivity information records and a plurality of sample secondary defect detection result records, wherein the plurality of sample light reflectivity information records respectively comprise sample light reflectivity information of the inner bottom surface and different side surfaces of a target swimming pool, and the sample secondary defect detection result comprises whether brick layer defects are included or not;
respectively adopting the plurality of sample light reflectivity information records and a plurality of sample secondary defect detection result records as input and output, and constructing an optical defect detector based on machine learning and training and updating, wherein the optical defect detector comprises a plurality of optical defect classification branches;
adopting an updated optical defect detector to identify and classify the light reflectivity information to obtain a plurality of secondary defect detection results;
and extracting the region with the secondary defect detection result to obtain a plurality of defect regions.
Sample light reflectance information is collected for the interior bottom and various sides of the target pool, which information can be obtained by measuring the interior bottom and sides of the pool using light reflectance measuring devices or sensors, each sample having a corresponding light reflectance information record. And carrying out secondary defect detection on each sample according to the light reflectivity information records of the plurality of samples, and recording whether the detection result of the brick layer defect is included in each sample, namely judging whether the brick layer defect exists in the sample or not, recording the detection result, and obtaining a corresponding secondary defect detection result record of the plurality of samples.
A plurality of sample light reflectivity information records are prepared as inputs of training data, and a corresponding plurality of sample secondary defect detection result records are output of the training data, and an optical defect detector model is constructed based on machine learning, wherein the model adopts a deep learning technology, such as a Convolutional Neural Network (CNN), and a plurality of optical defect classification branches are designed inside the model, and each branch corresponds to a different type of defect.
Training and updating the optical defect detector using the prepared training data, enabling the model to learn the characteristics and rules of the optical defects by recording the light reflectivity information as input, and accurately predicting the result of whether there is a brick layer defect based on these characteristics. The trained optical defect detector is evaluated, and the performance and accuracy of the model are evaluated by using a group of test data which are not used in training, so that the brick layer defect can be effectively detected, and the brick layer defect detector has low false alarm rate and low false alarm rate.
Through the steps, the optical defect detector meeting the requirements is obtained, more accurate and reliable optical defect detection can be realized, and the effect of identifying the defects of the brick layer of the swimming pool is improved.
The plurality of light reflectivity information is supplied as input to the optical defect detector, each of which is detected and classified using the optical defect detector, which has learned the characteristics and rules of the optical defects through training and updating, so that it is possible to judge whether or not there is a brick layer defect based on the input light reflectivity information and classify it. And generating a plurality of secondary defect detection result records according to the detection and classification results, and recording whether the brick layer defects and the specific classification of the defects are contained for each piece of light reflectivity information.
For samples marked as having a brick layer defect, the corresponding defect region is extracted from its position in the light reflectivity information, for example, by calibrating the defect region by pixel coordinates or using a bounding box to indicate the location of the defect. And storing the extracted defect areas to obtain a plurality of defect area sets, wherein each area represents one detected brick layer defect.
Further, the method includes analyzing a detection scale of the acquired image, the method including:
acquiring a sample defect area quantity set and a sample defect grade record according to a maintenance data record of a swimming pool brick layer, and setting an acquired sample image detection scale set, wherein the size of the image detection scale is inversely related to the size of the sample defect area quantity and the size of the sample defect grade;
Based on the decision tree, constructing an image detection decision device by adopting the sample defect area quantity set, the sample defect grade record and the sample image detection scale set;
and adopting an image detection decision maker to obtain the image detection scale by decision classification according to the defect area quantity and the defect grade of the plurality of defect areas.
A maintenance data record of the pool brick layer is obtained, the number of detected defect areas is calculated for each sample, and these numbers are assembled into a sample defect area number set. Defect grade information for each sample is extracted from a maintenance data record of the pool brick layer, the grades representing the severity of the defect, and the grades are recorded and associated with the corresponding samples.
And acquiring a sample image detection scale set, wherein the size of the detection scale is set according to the number of sample defect areas and the size of sample defect levels, and when the number of the defect areas and the defect levels are large, the detection scale of the image is reduced so as to capture the defects more finely. In this way, the proper image detection scale can be adaptively set according to the defect conditions of different samples, so as to obtain more accurate and effective defect detection results.
And arranging the sample defect area quantity set, the sample defect grade record and the sample image detection scale set into a form of a feature matrix and a target vector so as to carry out decision tree training. The feature matrix and the target vector are used for training a decision tree, the decision tree can be used for constructing a classification model to predict whether the image has defects by analyzing the relation between the features, and the classification model comprehensively considers the features of the sample, such as the number of defective areas, the defect level, the image detection scale and the like, so that an accurate defect detection result is provided. And constructing an image detection decision maker according to the decision tree model obtained through training, wherein the decision maker receives input features including the number of sample defect areas, the sample defect level and the sample image detection scale, and predicts based on the branching conditions of the decision tree.
The method comprises the steps of collecting the number and the grade of defect areas of a plurality of defect areas, transmitting the number and the grade of the defect areas to an image detection decision maker, classifying the input characteristics by the decision maker based on rules and branching conditions learned in a training process, and determining the detection scale applicable to the image according to an output classification result of the decision maker.
Dividing detected local areas and acquiring images in the defect areas according to the image detection scale by an image detection module to obtain a plurality of local images;
further, a plurality of local images are obtained, the method comprising:
traversing and dividing the defect area information according to the image detection scale by the image detection module to obtain a plurality of image detection areas;
and acquiring the plurality of images to detect the local images, and obtaining a plurality of local images.
The image detection module is used for detecting the defect areas in the whole image, each defect area is divided according to the determined image detection scale, the corresponding local area in the image is divided, for example, a window with a fixed size is defined around the defect area or the local area is divided in a self-adaptive mode according to the size of the defect area, a plurality of image detection areas are obtained after traversing the division, and each defect area can be divided into a plurality of local areas.
And acquiring images of the corresponding image detection areas by using an image sensor, such as a high-definition camera, for each image detection area, traversing the plurality of image detection areas to acquire a plurality of local images, wherein each image represents one image detection area.
And performing three-time defect analysis on the brick layers of the plurality of local images through the defect analysis module to obtain a plurality of defect information which is used as a brick layer defect detection result of the target swimming pool.
Further, the method comprises the steps of:
based on a defect analysis module, acquiring a brick layer maintenance data record of a swimming pool, acquiring a plurality of sample local image sets of a bottom surface and a plurality of side surfaces, and acquiring a plurality of sample defect information sets, wherein the sample defect information comprises arching, cracking or falling;
based on a convolutional neural network, constructing a brick layer three-time defect analysis model comprising a plurality of brick layer defect identification branches, and respectively performing supervision training update on the plurality of brick layer defect identification branches by adopting the plurality of sample local image sets and the plurality of sample defect information sets;
and based on the updated three-time defect analysis model of the brick layer, inputting corresponding brick layer defect identification branches according to the positions of the plurality of local images in the target swimming pool, and obtaining the plurality of defect information.
Maintenance data for the brick layer of the swimming pool is collected and recorded, including inspection dates, location information, and other relevant maintenance parameters. According to the acquired maintenance data record, the bottom surface and the plurality of side surfaces are selected to be used as the interested areas, and corresponding image acquisition equipment such as a camera is utilized to acquire a plurality of sample local image sets, so that good image quality is ensured, and details of a swimming pool brick layer can be clearly displayed. And carrying out defect identification on the sample local image set based on maintenance parameters, analyzing and extracting defect information such as arching, cracking or falling off for each local image, and recording the information to form a plurality of sample defect information sets.
Based on the architecture of Convolutional Neural Network (CNN), a model suitable for identifying brick layer defects is created, which comprises a plurality of branches, each of which is specially responsible for identifying different types of brick layer defects, such as arching, cracking or falling off, etc. And taking the acquired local image sets of the plurality of samples as input data, and taking the corresponding defect information sets of the plurality of samples as target labels, so as to ensure the corresponding relation between the images and the labels.
And for each brick layer defect identification branch, performing supervision training by using the sample local image set and the sample defect information set, and optimizing model parameters by minimizing the difference between the prediction output and the real label in the training process so as to improve the identification accuracy of the brick layer defects. As training progresses, the model is updated periodically with a new set of sample local images and sample defect information, which helps the model adapt to more data changes and new defect patterns, improving its generalization ability. The three-time defect analysis model of the brick layer is obtained through training and updating, so that the identification performance of defects of different types of brick layers can be effectively improved.
Positioning a plurality of partial images within a target swimming pool using image processing techniques or target detection algorithms to determine their location in the swimming pool, inputting each partial image into a respective tile defect recognition branch that is part of a model dedicated to identifying and analyzing defects on each tile, obtaining information about defects present on each tile by processing the plurality of partial images in each tile defect recognition branch, obtaining the plurality of defect information.
In summary, the method and the system for detecting defects of a brick layer of a swimming pool provided by the embodiment of the application have the following technical effects:
1. the comprehensive swimming pool brick layer defect detection device is provided by comprehensively utilizing the water level test module, the light detection module, the image detection module and the defect analysis module, so that the detection of the swimming pool brick layer defect can be completed in a short time, and the working efficiency is improved;
2. the minimum and maximum water level information is obtained through the water level testing module, and the optical detection and image processing technology is combined to perform multiple defect analysis and comprehensive consideration, so that the defect grade can be accurately judged, the defect information is obtained, and the accuracy of defect analysis is improved;
3. through multiple defect analysis, including light reflectivity information and local image acquisition and analysis, various defect information of the swimming pool brick layer, including obvious and tiny defects, can be comprehensively detected and obtained, and the comprehensiveness and accuracy of detection are improved.
In summary, by introducing the automation equipment and the image processing technology, the swimming pool brick layer defect detection method solves the technical problems of low detection efficiency, inaccurate defect analysis, lack of comprehensiveness and the like in the traditional method, and achieves the technical effects of high efficiency, high speed, accuracy, reliability, comprehensiveness and delicacy.
Example two
Based on the same inventive concept as the method for detecting defects of a swimming pool tile layer in the foregoing embodiments, as shown in fig. 2, the present application provides a swimming pool tile layer defect detection system, which is applied to a swimming pool tile layer defect detection device, the device includes a water level test module, a light detection module, an image detection module, and a defect analysis module, the system includes:
the water level information acquisition unit 10 is used for detecting the average water level information of the target swimming pool in a plurality of preset time windows through the water level test module, and extracting and obtaining the minimum water level information and the maximum water level information;
the defect primary prediction unit 20 is configured to perform primary prediction of a brick layer defect based on the minimum water level information and the maximum water level information, so as to obtain a primary defect analysis result, where the primary defect analysis result includes a defect level;
a regional optical detection unit 30, wherein the regional optical detection unit 30 is configured to match and obtain a defect detection scale when the defect level in the primary defect analysis result is greater than 0, and divide and optically detect a detection region in the target swimming pool according to the defect detection scale by using the optical detection module to obtain a plurality of light reflectivity information;
A secondary defect analysis unit 40, where the secondary defect analysis unit 40 is configured to perform secondary defect analysis on the brick layer according to the plurality of light reflectivity information, obtain a plurality of defect areas, and analyze and obtain an image detection scale according to the number of defect areas and the defect level of the plurality of defect areas;
the regional image acquisition unit 50 is configured to divide detected local areas and acquire images in the plurality of defect areas according to the image detection scale by using an image detection module, so as to obtain a plurality of local images;
and a third defect analysis unit 60, where the third defect analysis unit 60 is configured to perform third defect analysis on the brick layer of the plurality of local images through the defect analysis module, to obtain a plurality of defect information, and use the plurality of defect information as a brick layer defect detection result of the target swimming pool.
Further, the system further comprises a defect analysis result acquisition module for executing the following operation steps:
acquiring a sample minimum water level information record and a sample maximum water level information record based on the water level monitoring information in the target swimming pool history time;
according to the brick layer maintenance record in the history time of the target swimming pool, a sample primary defect analysis result record is obtained, and the sample primary defect analysis result is determined according to the number and the size of the brick layer defects in the target swimming pool;
The sample minimum water level information record and the sample maximum water level information record are adopted as input, the sample primary defect analysis result record is adopted as output, a defect primary classifier is constructed, and training and updating are carried out;
and classifying and identifying the minimum water level information and the maximum water level information by adopting the updated defect primary classifier to obtain the primary defect analysis result.
Further, the system further comprises a defect detection scale acquisition module for performing the following operation steps:
judging whether the defect grade in the primary defect analysis result is greater than 0;
if not, finishing the detection of the brick layer defect, and if so, carrying out mapping matching in a defect detection scale matcher according to the defect grade to obtain the defect detection scale, wherein the defect detection scale matcher comprises mapping relations of a plurality of sample defect grades and a plurality of sample defect detection scales, each defect detection scale comprises the size of a region for carrying out optical detection, and the size of the defect detection scale is inversely related to the size of the defect grade.
Further, the system further comprises a plurality of defect area acquisition modules to perform the following operation steps:
Obtaining a plurality of sample light reflectivity information records and a plurality of sample secondary defect detection result records, wherein the plurality of sample light reflectivity information records respectively comprise sample light reflectivity information of the inner bottom surface and different side surfaces of a target swimming pool, and the sample secondary defect detection result comprises whether brick layer defects are included or not;
respectively adopting the plurality of sample light reflectivity information records and a plurality of sample secondary defect detection result records as input and output, and constructing an optical defect detector based on machine learning and training and updating, wherein the optical defect detector comprises a plurality of optical defect classification branches;
adopting an updated optical defect detector to identify and classify the light reflectivity information to obtain a plurality of secondary defect detection results;
and extracting the region with the secondary defect detection result to obtain a plurality of defect regions.
Further, the system further comprises an image detection scale acquisition module for executing the following operation steps:
acquiring a sample defect area quantity set and a sample defect grade record according to a maintenance data record of a swimming pool brick layer, and setting an acquired sample image detection scale set, wherein the size of the image detection scale is inversely related to the size of the sample defect area quantity and the size of the sample defect grade;
Based on the decision tree, constructing an image detection decision device by adopting the sample defect area quantity set, the sample defect grade record and the sample image detection scale set;
and adopting an image detection decision maker to obtain the image detection scale by decision classification according to the defect area quantity and the defect grade of the plurality of defect areas.
Further, the system also comprises a plurality of local image acquisition modules for executing the following operation steps:
traversing and dividing the defect area information according to the image detection scale by the image detection module to obtain a plurality of image detection areas;
and acquiring the plurality of images to detect the local images, and obtaining a plurality of local images.
Further, the system further comprises a plurality of defect information acquiring modules for executing the following operation steps:
based on a defect analysis module, acquiring a brick layer maintenance data record of a swimming pool, acquiring a plurality of sample local image sets of a bottom surface and a plurality of side surfaces, and acquiring a plurality of sample defect information sets, wherein the sample defect information comprises arching, cracking or falling;
based on a convolutional neural network, constructing a brick layer three-time defect analysis model comprising a plurality of brick layer defect identification branches, and respectively performing supervision training update on the plurality of brick layer defect identification branches by adopting the plurality of sample local image sets and the plurality of sample defect information sets;
And based on the updated three-time defect analysis model of the brick layer, inputting corresponding brick layer defect identification branches according to the positions of the plurality of local images in the target swimming pool, and obtaining the plurality of defect information.
From the foregoing detailed description of a method for detecting defects of a tile layer of a swimming pool, it will be apparent to those skilled in the art that a system for detecting defects of a tile layer of a swimming pool in this embodiment is described more simply for the device disclosed in the embodiments, since it corresponds to the method disclosed in the embodiments, and the relevant points refer to the description of the method section.
The previous description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the application. Thus, the present application is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
Claims (8)
1. A method for detecting defects of a brick layer of a swimming pool, which is applied to a device for detecting defects of a brick layer of a swimming pool, wherein the device comprises a water level testing module, a light detecting module, an image detecting module and a defect analyzing module, and the method comprises the following steps:
detecting average water level information of a target swimming pool in a plurality of preset time windows through a water level testing module, and extracting and obtaining minimum water level information and maximum water level information;
performing primary prediction on the brick layer defects based on the minimum water level information and the maximum water level information to obtain primary defect analysis results, wherein the primary defect analysis results comprise defect grades;
when the defect grade in the primary defect analysis result is greater than 0, matching to obtain a defect detection scale, and dividing and optically detecting a detection area in the target swimming pool according to the defect detection scale by the optical detection module to obtain a plurality of light reflectivity information;
performing secondary defect analysis on the brick layer according to the light reflectivity information to obtain a plurality of defect areas, and analyzing and obtaining an image detection scale according to the defect area number and the defect grade of the defect areas;
Dividing detected local areas and acquiring images in the defect areas according to the image detection scale by an image detection module to obtain a plurality of local images;
and performing three-time defect analysis on the brick layers of the plurality of local images through the defect analysis module to obtain a plurality of defect information which is used as a brick layer defect detection result of the target swimming pool.
2. The method according to claim 1, characterized in that the method comprises:
acquiring a sample minimum water level information record and a sample maximum water level information record based on the water level monitoring information in the target swimming pool history time;
according to the brick layer maintenance record in the history time of the target swimming pool, a sample primary defect analysis result record is obtained, and the sample primary defect analysis result is determined according to the number and the size of the brick layer defects in the target swimming pool;
the sample minimum water level information record and the sample maximum water level information record are adopted as input, the sample primary defect analysis result record is adopted as output, a defect primary classifier is constructed, and training and updating are carried out;
and classifying and identifying the minimum water level information and the maximum water level information by adopting the updated defect primary classifier to obtain the primary defect analysis result.
3. The method according to claim 1, characterized in that the method comprises:
judging whether the defect grade in the primary defect analysis result is greater than 0;
if not, finishing the detection of the brick layer defect, and if so, carrying out mapping matching in a defect detection scale matcher according to the defect grade to obtain the defect detection scale, wherein the defect detection scale matcher comprises mapping relations of a plurality of sample defect grades and a plurality of sample defect detection scales, each defect detection scale comprises the size of a region for carrying out optical detection, and the size of the defect detection scale is inversely related to the size of the defect grade.
4. The method according to claim 1, characterized in that the method comprises:
obtaining a plurality of sample light reflectivity information records and a plurality of sample secondary defect detection result records, wherein the plurality of sample light reflectivity information records respectively comprise sample light reflectivity information of the inner bottom surface and different side surfaces of a target swimming pool, and the sample secondary defect detection result comprises whether brick layer defects are included or not;
respectively adopting the plurality of sample light reflectivity information records and a plurality of sample secondary defect detection result records as input and output, and constructing an optical defect detector based on machine learning and training and updating, wherein the optical defect detector comprises a plurality of optical defect classification branches;
Adopting an updated optical defect detector to identify and classify the light reflectivity information to obtain a plurality of secondary defect detection results;
and extracting the region with the secondary defect detection result to obtain a plurality of defect regions.
5. The method according to claim 1, characterized in that the method comprises:
acquiring a sample defect area quantity set and a sample defect grade record according to a maintenance data record of a swimming pool brick layer, and setting an acquired sample image detection scale set, wherein the size of the image detection scale is inversely related to the size of the sample defect area quantity and the size of the sample defect grade;
based on the decision tree, constructing an image detection decision device by adopting the sample defect area quantity set, the sample defect grade record and the sample image detection scale set;
and adopting an image detection decision maker to obtain the image detection scale by decision classification according to the defect area quantity and the defect grade of the plurality of defect areas.
6. The method according to claim 1, characterized in that the method comprises:
traversing and dividing the defect area information according to the image detection scale by the image detection module to obtain a plurality of image detection areas;
And acquiring the plurality of images to detect the local images, and obtaining a plurality of local images.
7. The method according to claim 6, characterized in that the method comprises:
based on a defect analysis module, acquiring a brick layer maintenance data record of a swimming pool, acquiring a plurality of sample local image sets of a bottom surface and a plurality of side surfaces, and acquiring a plurality of sample defect information sets, wherein the sample defect information comprises arching, cracking or falling;
based on a convolutional neural network, constructing a brick layer three-time defect analysis model comprising a plurality of brick layer defect identification branches, and respectively performing supervision training update on the plurality of brick layer defect identification branches by adopting the plurality of sample local image sets and the plurality of sample defect information sets;
and based on the updated three-time defect analysis model of the brick layer, inputting corresponding brick layer defect identification branches according to the positions of the plurality of local images in the target swimming pool, and obtaining the plurality of defect information.
8. A swimming pool tile layer defect detection system, wherein the system is applied to a swimming pool tile layer defect detection device, the device comprises a water level test module, a light detection module, an image detection module and a defect analysis module, and the swimming pool tile layer defect detection system is used for implementing the swimming pool tile layer defect detection method as claimed in any one of claims 1-7, and comprises the following steps:
The water level information acquisition unit is used for detecting the average water level information of the target swimming pool in a plurality of preset time windows through the water level test module, and extracting and obtaining the minimum water level information and the maximum water level information;
the defect primary prediction unit is used for carrying out primary prediction on the brick layer defects based on the minimum water level information and the maximum water level information to obtain primary defect analysis results, wherein the primary defect analysis results comprise defect grades;
the regional optical detection unit is used for matching and acquiring a defect detection scale when the defect grade in the primary defect analysis result is greater than 0, and dividing and optically detecting detection regions in the target swimming pool according to the defect detection scale by the optical detection module to acquire a plurality of light reflectivity information;
the secondary defect analysis unit is used for performing secondary defect analysis on the brick layer according to the light reflectivity information to obtain a plurality of defect areas, and analyzing and obtaining an image detection scale according to the defect area number and the defect grade of the defect areas;
The regional image acquisition unit is used for dividing detected local areas and acquiring images in the plurality of defect areas according to the image detection scale by the image detection module to obtain a plurality of local images;
and the third defect analysis unit is used for performing third defect analysis on the brick layers of the plurality of local images through the defect analysis module to obtain a plurality of defect information which is used as a brick layer defect detection result of the target swimming pool.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311473166.0A CN117250201B (en) | 2023-11-08 | 2023-11-08 | Swimming pool brick layer defect detection method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311473166.0A CN117250201B (en) | 2023-11-08 | 2023-11-08 | Swimming pool brick layer defect detection method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117250201A true CN117250201A (en) | 2023-12-19 |
| CN117250201B CN117250201B (en) | 2024-02-13 |
Family
ID=89126570
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311473166.0A Active CN117250201B (en) | 2023-11-08 | 2023-11-08 | Swimming pool brick layer defect detection method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117250201B (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2007315964A (en) * | 2006-05-26 | 2007-12-06 | Hitachi Ltd | Underwater defect inspection device and underwater defect inspection method |
| CN113640310A (en) * | 2021-10-18 | 2021-11-12 | 南京光衡科技有限公司 | Tile surface defect detection visual system and detection method |
| CN115356260A (en) * | 2022-07-29 | 2022-11-18 | 中水珠江规划勘测设计有限公司 | High-water-level running pipeline health condition efficient detection method |
| CN115511775A (en) * | 2021-06-23 | 2022-12-23 | 上海电力大学 | Light-weight ceramic tile surface defect detection method based on semantic segmentation |
| CN116245317A (en) * | 2023-01-18 | 2023-06-09 | 数智魔力(深圳)云计算技术有限公司 | Swimming pool water quality supervision system and method based on multivariate data |
| CN116840258A (en) * | 2023-07-04 | 2023-10-03 | 东南大学 | Pier disease detection method based on multifunctional underwater robot and stereoscopic vision |
-
2023
- 2023-11-08 CN CN202311473166.0A patent/CN117250201B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2007315964A (en) * | 2006-05-26 | 2007-12-06 | Hitachi Ltd | Underwater defect inspection device and underwater defect inspection method |
| CN115511775A (en) * | 2021-06-23 | 2022-12-23 | 上海电力大学 | Light-weight ceramic tile surface defect detection method based on semantic segmentation |
| CN113640310A (en) * | 2021-10-18 | 2021-11-12 | 南京光衡科技有限公司 | Tile surface defect detection visual system and detection method |
| CN115356260A (en) * | 2022-07-29 | 2022-11-18 | 中水珠江规划勘测设计有限公司 | High-water-level running pipeline health condition efficient detection method |
| CN116245317A (en) * | 2023-01-18 | 2023-06-09 | 数智魔力(深圳)云计算技术有限公司 | Swimming pool water quality supervision system and method based on multivariate data |
| CN116840258A (en) * | 2023-07-04 | 2023-10-03 | 东南大学 | Pier disease detection method based on multifunctional underwater robot and stereoscopic vision |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117250201B (en) | 2024-02-13 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| TWI603074B (en) | Optical film defect detection method and system thereof | |
| CN108711148A (en) | A kind of wheel tyre defect intelligent detecting method based on deep learning | |
| CN113592828B (en) | Nondestructive testing method and system based on industrial endoscope | |
| CN111652883B (en) | Glass surface defect detection method based on deep learning | |
| CN111402236B (en) | Hot rolled strip steel surface defect grading method based on image gray value | |
| US11521120B2 (en) | Inspection apparatus and machine learning method | |
| CN109741927A (en) | The equipment fault of miniature transformer production line and potential defective products intelligent predicting system | |
| CN111881970A (en) | Intelligent outer broken image identification method based on deep learning | |
| CN111611294A (en) | Star sensor data anomaly detection method | |
| CN117152119A (en) | Profile flaw visual detection method based on image processing | |
| CN117057644A (en) | Equipment production quality detection method and system based on characteristic matching | |
| CN114034772B (en) | Expert system for detecting potential failure of roller and predicting residual service life | |
| CN115035328A (en) | Converter image increment automatic machine learning system and establishment training method thereof | |
| CN117636073B (en) | Concrete defect detection method, device and storage medium | |
| CN117250201B (en) | Swimming pool brick layer defect detection method and system | |
| CN116380496B (en) | Automobile door fatigue endurance test method, system and medium | |
| CN110455370B (en) | Flood-prevention drought-resisting remote measuring display system | |
| CN116679653A (en) | Intelligent acquisition system for industrial equipment data | |
| CN113960700B (en) | Objective inspection, statistics and analysis system for regional numerical forecasting result | |
| CN115438547A (en) | Overall evaluation method and system based on pavement service state | |
| CN115546108A (en) | Intelligent detection method for appearance quality of automobile tire based on edge cloud cooperation and AR | |
| CN114565883A (en) | Graphic recognition algorithm for operation faults of equipment | |
| CN112508946A (en) | Cable tunnel abnormity detection method based on antagonistic neural network | |
| CN112766141A (en) | Method and system for detecting foreign matters in tobacco wrapping equipment | |
| CN117607019B (en) | Intelligent detection method and detection system for electric power fitting surface |
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 |