CN116071335A - Wall surface acceptance method, device, equipment and storage medium - Google Patents

Wall surface acceptance method, device, equipment and storage medium Download PDF

Info

Publication number
CN116071335A
CN116071335A CN202310123195.8A CN202310123195A CN116071335A CN 116071335 A CN116071335 A CN 116071335A CN 202310123195 A CN202310123195 A CN 202310123195A CN 116071335 A CN116071335 A CN 116071335A
Authority
CN
China
Prior art keywords
flaw
wall surface
checked
image
acceptance
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.)
Pending
Application number
CN202310123195.8A
Other languages
Chinese (zh)
Inventor
陈叶能
吴倩倩
陈丽燕
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
China Unicom Zhejiang Industrial Internet Co Ltd
Original Assignee
China Unicom Zhejiang Industrial Internet Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by China Unicom Zhejiang Industrial Internet Co Ltd filed Critical China Unicom Zhejiang Industrial Internet Co Ltd
Priority to CN202310123195.8A priority Critical patent/CN116071335A/en
Publication of CN116071335A publication Critical patent/CN116071335A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
    • G01N21/84—Systems specially adapted for particular applications
    • G01N21/88—Investigating the presence of flaws or contamination
    • G01N21/8851—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/0002—Inspection of images, e.g. flaw detection
    • G06T7/0004—Industrial image inspection
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/40—Extraction of image or video features
    • G06V10/44—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
    • 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
    • G06V10/774—Generating sets of training patterns; Bootstrap methods, e.g. bagging or boosting
    • 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
    • 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
    • Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
    • Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
    • Y02P90/30—Computing systems specially adapted for manufacturing

Abstract

The application provides a wall surface acceptance method, device, equipment and storage medium, at first gather and wait to accept the image, wait to accept the image and include the many images of waiting to accept the wall, then wait to accept the image and carry out feature extraction according to the feature extraction model, obtain the flaw information that waits to accept the wall, wherein, the feature extraction model is according to the training of many wall images that have the feature flaw, whether confirm to accept the wall and pass acceptance according to flaw information and acceptance criteria, thereby obtain the wall flaw feature through the image recognition technique and combine acceptance criteria to accept the wall and realize automatic acceptance, the subjectivity and the size deviation that exist of inspection wall have been overcome in the prior art naked eye observation and manual measurement, guarantee the accuracy of acceptance result, and then can promote completion acceptance result reliability and acceptance experience.

Description

Wall surface acceptance method, device, equipment and storage medium
Technical Field
The application relates to the technical field of detection, in particular to a wall surface acceptance method, a wall surface acceptance device, wall surface acceptance equipment and a wall surface acceptance storage medium.
Background
In the decoration engineering, the appearance of the engineering wall paint directly influences the appearance effect of the whole engineering, and the appearance acceptance of the engineering wall paint is an important link of engineering completion acceptance. The appearance inspection and acceptance of the engineering wall paint can comprise inspection and acceptance of flaw conditions of sand holes, sagging, rising knots and the like of the wall surface.
The current acceptance means mainly depend on information observed by naked eyes and a manual measurement mode to judge whether the engineering wall surface passes acceptance. For example, after the wall paint is dried, the defects such as sand holes, flow drops and rising knots are checked by adopting visual inspection and hand feeling along the wall surface at a certain position from the wall surface by means of a bulb under natural light or when the light is poor, and if the defects exist, the sizes of the defects such as sand holes, flow drops and rising knots are manually measured by means of a measuring tool such as a steel ruler. And then, whether the wall surface passes the acceptance inspection is determined by combining with the acceptance inspection standard through manual judgment.
However, visual inspection and manual measurement have problems of subjectivity and dimensional deviation, which results in inaccurate acceptance results and affects as-built acceptance results.
Disclosure of Invention
The application provides a wall surface acceptance method, device, equipment and storage medium, which are used for solving the technical problems that subjective performance and size deviation exist on an acceptance wall surface by adopting naked eye observation and manual measurement in the prior art, and the acceptance result is inaccurate and then the completion acceptance result is influenced.
In a first aspect, the present application provides a wall surface acceptance method, including:
collecting an image to be checked, wherein the image to be checked comprises a plurality of images of a wall surface to be checked;
Performing feature extraction on the to-be-inspected image according to a feature extraction model to obtain to-be-inspected flaw information of the to-be-inspected wall surface, wherein the feature extraction model is obtained by training a plurality of wall surface images with feature flaws;
and determining whether the wall surface to be checked is accepted or not according to the flaw information to be checked and the acceptance standard.
In one possible design, before the feature extraction is performed on the wall surface image to be inspected according to the feature extraction model, the method further includes:
acquiring a training data set, wherein the training data set comprises a plurality of wall surface images with characteristic flaws and design images corresponding to each wall surface image with the characteristic flaws;
and training the initial feature extraction model according to the training data set to obtain the feature extraction model.
In one possible design, the feature extraction of the image to be inspected according to the feature extraction model, to obtain defect information to be inspected of the wall to be inspected, includes:
inputting the image to be checked and the design image to be checked into the feature extraction model to perform feature extraction, and outputting an extraction result;
analyzing the extraction result according to a size deviation strategy, and determining the obtained analysis result as the flaw information to be checked;
The flaw information to be checked comprises flaw types and flaw grades of the images to be checked, and the design images to be checked are original design images of the wall surfaces to be checked.
In one possible design, the determining whether the wall surface to be inspected is acceptable according to the flaw information to be inspected and the inspection standard includes:
judging whether the flaw category of the image to be checked is consistent with at least one of preset flaw categories;
if not, determining that the wall surface to be checked is accepted through checking;
if yes, determining whether the wall surface to be checked is accepted or not according to the flaw grade corresponding to the flaw category with the same category and the preset flaw grade;
wherein the acceptance criteria include the preset flaw category and the preset flaw level.
In one possible design, the determining whether the wall surface to be inspected is accepted according to the flaw grade corresponding to the flaws with the same category and the preset flaw grade includes:
comparing the flaw grade corresponding to the flaw class with the same class with the preset flaw grade;
if the flaw grades corresponding to the flaw grades with the same grade are all first flaw grades of the preset flaw grade, determining that the wall surface to be checked is accepted through checking;
The preset flaw grade comprises a first flaw grade, a second flaw grade and a third flaw grade of each preset flaw class.
In one possible design, the predetermined flaw categories include one or more of a sand hole flaw category, a sink flaw category, and a pimple flaw category.
In one possible design, the parsing the extraction result according to a size deviation strategy includes:
generating a corresponding class label for the flaw class of the image to be checked according to the flaw class label, wherein the class label is used for representing the flaw class of the image to be checked;
determining flaw grades of the image to be inspected according to flaw size ranges, wherein the flaw size ranges comprise size ranges corresponding to the first flaw grade, the second flaw grade and the third flaw grade of each preset flaw class;
wherein the size deviation strategy comprises the flaw class label and the flaw size range.
In one possible design, the acquiring the image to be accepted includes:
shooting the wall surface to be checked by shooting equipment according to shooting standards so as to acquire the image to be checked, wherein the shooting standards comprise shooting distances and shooting angles.
In one possible design, before the feature extraction is performed on the image to be inspected according to the feature extraction model, the method further includes:
and carrying out identity authentication on the shooting equipment according to equipment identification, wherein the equipment identification is used for uniquely identifying the shooting equipment.
In a second aspect, the present application provides a wall surface acceptance device comprising:
the acquisition module is used for acquiring an image to be checked and accepted, wherein the image to be checked and accepted comprises a plurality of images of a wall surface to be checked and accepted;
the extraction module is used for carrying out feature extraction on the images to be inspected according to a feature extraction model to obtain defect information to be inspected of the walls to be inspected, and the feature extraction model is obtained by training according to a plurality of wall images with feature defects;
and the acceptance module is used for determining whether the wall surface to be accepted is accepted or not according to the flaw information to be accepted and the acceptance standard.
In one possible design, the wall surface acceptance device further includes: a training module; the training module is used for:
acquiring a training data set, wherein the training data set comprises a plurality of wall surface images with characteristic flaws and design images corresponding to each wall surface image with the characteristic flaws;
And training the initial feature extraction model according to the training data set to obtain the feature extraction model.
In one possible design, the extraction module includes:
the output module is used for inputting the image to be checked and the design image to be checked into the feature extraction model to perform feature extraction and outputting an extraction result;
the analysis module is used for analyzing the extraction result according to a size deviation strategy and determining the obtained analysis result as the flaw information to be checked;
the flaw information to be checked comprises flaw types and flaw grades of the images to be checked, and the design images to be checked are original design images of the wall surfaces to be checked.
In one possible design, the acceptance module is specifically configured to:
judging whether the flaw category of the image to be checked is consistent with at least one of preset flaw categories;
if not, determining that the wall surface to be checked is accepted through checking;
if yes, determining whether the wall surface to be checked is accepted or not according to the flaw grade corresponding to the flaw category with the same category and the preset flaw grade;
wherein the acceptance criteria include the preset flaw category and the preset flaw level.
In one possible design, the acceptance module is further configured to:
comparing the flaw grade corresponding to the flaw class with the same class with the preset flaw grade;
if the flaw grades corresponding to the flaw grades with the same grade are all first flaw grades of the preset flaw grade, determining that the wall surface to be checked is accepted through checking;
the preset flaw grade comprises a first flaw grade, a second flaw grade and a third flaw grade of each preset flaw class.
In one possible design, the predetermined flaw categories include one or more of a sand hole flaw category, a sink flaw category, and a pimple flaw category.
In one possible design, the parsing module is specifically configured to:
generating a corresponding class label for the flaw class of the image to be checked according to the flaw class label, wherein the class label is used for representing the flaw class of the image to be checked;
determining flaw grades of the image to be inspected according to flaw size ranges, wherein the flaw size ranges comprise size ranges corresponding to the first flaw grade, the second flaw grade and the third flaw grade of each preset flaw class;
Wherein the size deviation strategy comprises the flaw class label and the flaw size range.
In one possible design, the acquisition module is specifically configured to:
shooting the wall surface to be checked by shooting equipment according to shooting standards so as to acquire the image to be checked, wherein the shooting standards comprise shooting distances and shooting angles.
In one possible design, the wall surface acceptance device further includes: an authentication module; the authentication module is used for:
and carrying out identity authentication on the shooting equipment according to equipment identification, wherein the equipment identification is used for uniquely identifying the shooting equipment.
In a third aspect, the present application provides an electronic device, comprising: a processor, and a memory communicatively coupled to the processor;
the memory stores computer-executable instructions;
the processor executes computer-executable instructions stored in the memory to implement any one of the possible wall surface acceptance methods provided in the first aspect.
In a fourth aspect, the present application provides a computer readable storage medium having stored therein computer executable instructions which when executed by a processor are adapted to carry out any one of the possible wall surface acceptance methods provided in the first aspect.
In a fifth aspect, the present application provides a computer program product comprising computer-executable instructions for implementing any one of the possible wall surface acceptance methods provided in the first aspect when executed by a processor.
The application provides a wall surface acceptance method, device, equipment and storage medium, at first gather and wait to accept the image, wait to accept the image and include the many images of waiting to accept the wall, then wait to accept the image and carry out feature extraction according to the feature extraction model, obtain the flaw information that waits to accept the wall, wherein, the feature extraction model is according to the training of many wall images that have the feature flaw, whether confirm to accept the wall and pass acceptance according to flaw information and acceptance criteria, thereby obtain the wall flaw feature through the image recognition technique and combine acceptance criteria to accept the wall and realize automatic acceptance, the subjectivity and the size deviation that exist of inspection wall have been overcome in the prior art naked eye observation and manual measurement, guarantee the accuracy of acceptance result, and then can promote completion acceptance result reliability and acceptance experience.
Drawings
In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, a brief description will be given below of the drawings that are needed in the embodiments or the prior art descriptions, it being obvious that the drawings in the following description are some embodiments of the present application, and that other drawings may be obtained according to these drawings without inventive effort to a person skilled in the art.
Fig. 1 is a schematic view of an application scenario provided in an embodiment of the present application;
fig. 2 is a schematic flow chart of a wall surface acceptance method according to an embodiment of the present application;
fig. 3 is a flow chart of another wall surface acceptance method according to an embodiment of the present disclosure;
fig. 4 is a flow chart of another wall surface acceptance method according to an embodiment of the present disclosure;
fig. 5 is a schematic structural diagram of a wall inspection device according to an embodiment of the present application;
fig. 6 is a schematic structural diagram of another wall inspection device according to an embodiment of the present disclosure;
fig. 7 is a schematic structural diagram of an electronic device according to an embodiment of the present application.
Detailed Description
Reference will now be made in detail to exemplary embodiments, examples of which are illustrated in the accompanying drawings. When the following description refers to the accompanying drawings, the same numbers in different drawings refer to the same or similar elements, unless otherwise indicated. The implementations described in the following exemplary examples are not representative of all implementations consistent with the present application. Rather, they are merely examples of methods and apparatus consistent with aspects of the present application as detailed in the accompanying claims.
The terms "first," "second," "third," "fourth" and the like in the description and in the claims of this application and in the above-described figures, if any, are used for distinguishing between similar objects and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used may be interchanged where appropriate such that embodiments of the present application described herein may be capable of operation in sequences other than those illustrated or described herein, for example. Furthermore, the terms "comprises," "comprising," and "having," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.
At present, whether the engineering wall surface passes inspection or not is judged mainly by means of information observed by naked eyes and manual measurement. For example, after the wall paint is dried, the defects such as sand holes, flow drops and rising knots are checked by adopting visual inspection and hand feeling along the wall surface at a certain position from the wall surface by means of a bulb under natural light or when the light is poor, and if the defects exist, the sizes of the defects such as sand holes, flow drops and rising knots are manually measured by means of a measuring tool such as a steel ruler. And then, whether the wall surface passes the acceptance inspection is determined by combining with the acceptance inspection standard through manual judgment. However, visual inspection and manual measurement have problems of subjectivity and dimensional deviation, which results in inaccurate acceptance results and affects as-built acceptance results.
Aiming at the problems in the prior art, the application provides a wall surface acceptance method, a device, equipment and a storage medium. The wall surface acceptance method provided by the application has the following inventive conception: the method comprises the steps of collecting images to be checked and training to obtain a feature extraction model through a plurality of wall images with feature flaws, then carrying out feature extraction on the images to be checked through the feature extraction model to obtain information of the flaws to be checked of the walls to be checked, and further determining whether the walls to be checked pass through checking and accepting through combining checking and accepting standards, so that the wall flaw features are obtained through an image recognition technology and automatic checking and accepting are realized on the walls to be checked and accepting through combining checking and accepting standards, subjectivity and size deviation existing in visual observation and manual measurement of the checking and accepting walls in the prior art are overcome, accuracy of checking and accepting results is guaranteed, and reliability and checking and accepting experience of the checking and accepting results can be improved.
In the following, an exemplary application scenario of the embodiments of the present application is described.
Fig. 1 is a schematic view of an application scenario provided in the embodiment of the present application, as shown in fig. 1, the appearance of the engineering wall paint directly affects the appearance effect of the whole engineering, and the appearance acceptance of the engineering wall paint is an important link of the acceptance of the completion of the engineering. The photographing device 11 is used for photographing the wall surface 12 to be inspected so as to acquire a plurality of images of the wall surface 12 to be inspected, namely, acquire the images to be inspected. The electronic device 13 is configured to execute the wall surface inspection method provided in the embodiment of the present application, the electronic device 13 operates a feature extraction model trained according to a plurality of wall surface images with feature flaws, the image to be inspected is input to the feature extraction model operated by the electronic device 13 to perform feature extraction, the flaw information to be inspected of the wall surface 12 is obtained, and then the flaw information to be inspected is judged by combining with a preset inspection standard, so as to determine whether the wall surface 12 to be inspected passes inspection, and automatic inspection of the wall surface 12 to be inspected is realized.
It will be appreciated that the photographing device 11 may be, for example, a camera, a video camera, and other devices having photographing functions, such as a camera phone, a photographing wristwatch, and the like. The embodiment of the present application is not limited to a specific type of the photographing apparatus 11, and the photographing apparatus 11 in fig. 1 is illustrated as a camera. The electronic device 13 may be a computer, a smart phone, a server cluster, or other devices, and the embodiment of the present application does not limit the type of the electronic device 13, and the electronic device 13 in fig. 1 is illustrated by taking a computer as an example.
In addition, the wall surface 12 to be inspected may be any engineering wall surface, which is not limited in this embodiment of the present application.
It should be noted that the above application scenario is merely illustrative, and the wall surface acceptance method, device, equipment and storage medium provided in the embodiments of the present application include, but are not limited to, the above application scenario.
Fig. 2 is a schematic flow chart of a wall surface acceptance method according to an embodiment of the present application. As shown in fig. 2, the wall surface acceptance method provided in the embodiment of the application includes:
s101: and collecting an image to be checked.
The image to be checked comprises a plurality of images of the wall surface to be checked.
Shooting the wall surface to be checked according to shooting standards by shooting equipment so as to acquire a plurality of images of the wall surface to be checked, and defining the acquired images of the wall surface to be checked as the images to be checked.
For example, for a wall surface to be inspected, the wall surface to be inspected can be divided into a plurality of areas according to a preset area size, and each area of the divided wall surface to be inspected is photographed one by adopting photographing equipment according to photographing standards, so that a plurality of images of the wall surface to be inspected are obtained.
The photographing standard may include a photographing distance and a photographing angle. Specifically, the shooting distance may be the same horizontal distance between the camera of the shooting device and each region of the wall surface to be inspected at all times when the region is shot. The shooting angle may be the same angle that is always maintained between the camera of the shooting device and each region of the wall surface to be inspected when shooting the region of the wall surface to be inspected. The shooting standard is set to ensure that images of each wall surface to be checked are acquired under the same shooting environment. In the actual working condition, the specific values of the horizontal distance and the angle can be set according to the actual working condition.
Optionally, in order to ensure the reliability of the data of the image to be checked, before the image to be checked is subjected to feature extraction according to the feature extraction model, the method further comprises the step of carrying out identity authentication on the shooting equipment according to the equipment identification of the shooting equipment, wherein the equipment identification is identified by the equipment identification, and the image to be checked from the shooting equipment identified by the equipment identification can be input into the feature extraction model. It will be appreciated that the photographing device and the electronic device running the feature extraction model may be communicatively connected by wireless and/or wired means for input of the image to be accepted.
S102: and carrying out feature extraction on the image to be inspected according to the feature extraction model to obtain flaw information to be inspected of the wall surface to be inspected.
And after the image to be checked is obtained, extracting the characteristics of the flaws on the image to be checked by using the characteristic extraction model to obtain the flaw information to be checked of the wall surface to be checked.
The feature extraction model is obtained through training according to a plurality of wall surface images with feature flaws. In other words, model training is performed in advance by adopting the wall surface image with the characteristic flaws, so that a characteristic extraction model capable of carrying out characteristic recognition on flaws on the image is obtained, and the characteristic extraction model is used for obtaining flaw information to be checked of the wall surface to be checked. Specifically, the flaw information to be inspected may include a flaw class and a flaw level of the image to be inspected. The flaw category may be preset in advance, for example, may be one or more of a sand hole flaw category, a flow drop flaw category, and a knot flaw category.
In one possible design, training to obtain a feature extraction model may include:
firstly, a training data set is acquired, and then the training data set is utilized to carry out model training on the initial feature extraction model, so that a feature extraction model is obtained.
The training data set comprises a plurality of wall surface images with characteristic flaws and design images corresponding to each wall surface image with the characteristic flaws, wherein the categories of the characteristic flaws are one or more of sand hole flaw categories, sagging flaw categories and knot flaw categories. The design image corresponding to each wall surface image with the characteristic flaws refers to the image of the wall surface in the design manuscript of the wall surface corresponding to the wall surface image with the characteristic flaws, and the design manuscript is formed at the early stage of engineering construction and takes a picture form as a standard, a basis and an effect in inspection. The design image corresponding to the wall surface image with the characteristic flaws comprises the original design lines of the wall surface with the characteristic flaws and the line related size information. And (3) performing image shooting on the wall surface with the characteristic flaws to obtain the wall surface image with the characteristic flaws.
And after the training data set is obtained, training an initial feature extraction model by using the training data set. The initial feature extraction model may be a neural network model in machine learning, and the embodiment of the present application is not limited to a specific type of the neural network model. Each set of training data may specifically include: the method comprises the steps of inputting training data into an initial feature extraction model to carry out a model, outputting whether the feature defect exists on the wall surface with the feature defect, calculating the size information of the feature defect, comparing the calculated size information with the input size information of the feature defect, and carrying out next training if the comparison result meets preset conditions, such as that the error of the size information meets the corresponding error range. If the initial feature extraction model is not satisfied, correcting and adjusting the initial feature extraction model until the obtained output information and the input size information of the feature flaws on the wall surface image with the feature flaws satisfy preset conditions so as to reach a preset accuracy rate or a convergence state, and ending training. And determining the initial feature extraction model ending training as a feature extraction model.
It can be understood that the training may also be finished by setting a loss function in the model training process, which is not limited to the embodiment of the present application.
In one possible design, a possible implementation manner of step S102 is shown in fig. 3, and fig. 3 is a flow chart of another wall surface acceptance method provided in the embodiment of the present application. As shown in fig. 3, the embodiment of the present application includes:
s1021: inputting the image to be checked and the design image to be checked into a feature extraction model to perform feature extraction, and outputting an extraction result.
Inputting the image to be checked and the original design image of the wall surface to be checked into a feature extraction model, so that the feature extraction model performs feature extraction, and outputting the feature extraction result. The extraction result comprises whether flaws exist on the image to be checked or not, and flaw type and flaw size information of the existing flaws. The design image to be checked is an original design image of the wall surface to be checked.
S1022: and generating a corresponding class label for the flaw class of the image to be checked according to the flaw class label.
S1023: and determining the flaw grade of the image to be checked according to the flaw size range.
After the extraction result is output, the extraction result is analyzed according to a size deviation strategy, and the obtained analysis result is determined as flaw information to be checked. The flaw information to be checked includes flaw types and flaw grades of the images to be checked.
The size deviation strategy includes a flaw class label and a flaw size range. For example, if any of the predetermined flaw categories is present, then a flaw of the predetermined flaw category is present, indicated by "1", and if a flaw of the predetermined flaw category is not present, indicated by "0". Each preset flaw category has a corresponding flaw category label, for example, the flaw category label of the sand hole flaw category is "S", the flaw category label of the sink flaw category is "L", and the flaw category label of the starting flaw category is "Q". The flaw class label is used to distinguish each flaw class. The flaw size range is the size range of flaws corresponding to a plurality of deviation grades of various preset flaw types. For example, 3 bias levels are set, namely a first flaw level (denoted as a), a second flaw level (denoted as B), and a third flaw level (denoted as C). The first flaw scale corresponds to a size range of [0,3.00) mm, the second flaw scale ranges from [3.00,6.00) mm, and the third flaw scale ranges from [6.00, ++) mm. For example, the flaw class is a hole flaw class, that is, the flaw is a hole, if the diameter of the hole is [0,3.00 ] mm, the flaw grade of the hole is a first flaw grade, if the diameter of the hole is [3.00,6.00) mm, the flaw grade of the hole is a second flaw grade, and if the diameter of the hole is [6.00, +_mm), the flaw grade of the hole is a third flaw grade.
It can be understood that specific values of the flaw class label and the flaw size range in the size deviation policy may be set according to actual conditions, which is not limited in this embodiment of the present application. The size deviation strategy may also be set according to specific categories of preset flaw categories and the general size ranges of flaws of these categories. The predetermined flaw categories include, but are not limited to, one or more of a sand hole flaw category, a flow drop flaw category, and a knot flaw category. The flaw size range includes a size range corresponding to each of the first flaw level, the second flaw level, and the third flaw level of each preset flaw type.
Analyzing the extraction result through the defined size deviation strategy, and outputting the analysis result. Specifically, a corresponding class label is generated for the flaw class of the image to be checked according to the flaw class label, and the flaw class of the image to be checked is represented through the class label. For example, if any type of defect exists in the extracted result, the defect is represented by a corresponding preset type label, so as to generate a corresponding type label for the defect type of the image to be checked. For example, if there is a flaw of the pinhole flaw type in the extraction result, the analysis result is denoted as S1. If no flaws of the pinhole flaw type exist, the analysis result is denoted as S0. The analysis results of the flaws of other preset flaw categories are similar, if flaws of the falling flaw category exist, the analysis result is denoted as L1. If no flaws of the falling flaw class exist, the analysis result is expressed as L0; if there is a flaw of the classification of the pimple flaw, the analysis result is denoted as Q1. If there is no defect of the pimple defect type, the analysis result is denoted as Q0.
And after generating the class label for the flaw class of the image to be checked, further, determining the flaw grade of the image to be checked according to the flaw size range. For example, if the size range of the flaws of the sand holes in the sand hole flaw class existing in the extraction result belongs to the first flaw grade, the analysis result is denoted as S1A, if the flaws belong to the second flaw grade, the analysis result is denoted as S1B, if the flaws belong to the third flaw grade, the analysis result is denoted as S1C, and the analysis results of the flaws of other flaw classes are similar, and will not be repeated here.
And generating a corresponding class label for the flaw class of the image to be inspected and determining the flaw grade of the image to be inspected according to the flaw size range, and analyzing the extraction result according to a size deviation strategy to obtain an analysis result, namely obtaining the flaw information to be inspected.
S103: and determining whether the wall surface to be checked is accepted or not according to the flaw information to be checked and the checking standard.
After the flaw information to be checked is obtained, whether the wall surface to be checked passes the checking and accepting is determined by combining the checking and accepting standard, and the automatic checking and accepting of the wall surface to be checked and accepting is completed.
The acceptance criteria may be set in advance, for example, a corresponding preset flaw class and a preset flaw level may be set in advance. The preset flaw grade at least comprises a first flaw grade, a second flaw grade and a third flaw grade of each preset flaw class.
Specifically, the flaw class and the flaw grade of the image to be inspected in the flaw information to be inspected are compared with the preset flaw class and the preset flaw grade in the inspection standard, so that an inspection result is obtained, namely whether the wall to be inspected passes inspection or not, the inspection result is output, and automatic inspection of the wall to be inspected is completed.
According to the wall surface acceptance method, firstly, the image to be inspected is collected, the image to be inspected comprises a plurality of images of the wall surface to be inspected, then the image to be inspected is subjected to feature extraction according to the feature extraction model, to-be-inspected flaw information of the wall surface to be inspected is obtained, the feature extraction model is obtained through training according to the plurality of wall surface images with feature flaws, whether the wall surface to be inspected passes through acceptance is determined according to the to-be-inspected flaw information and acceptance criteria, so that wall surface flaw features are obtained through an image recognition technology and automatic acceptance is achieved by combining the acceptance criteria with the wall surface to be inspected, subjectivity and size deviation of the wall surface to be inspected by naked eyes and manual measurement in the prior art are overcome, accuracy of acceptance results is guaranteed, and reliability and acceptance experience of completion inspection results can be improved.
In one possible design, a possible implementation of step S103 is shown in fig. 4. Fig. 4 is a flow chart of another wall surface acceptance method according to an embodiment of the present application. As shown in fig. 4, the embodiment of the present application includes:
s201: and judging whether the flaw class of the image to be checked is consistent with at least one of preset flaw classes.
And comparing the flaw category in the flaw information to be inspected with the preset flaw category, judging whether the flaw category of the image to be inspected in the flaw information to be inspected is consistent with at least one of the preset flaw categories, if not, indicating that any flaw in the preset flaw category does not exist in the image to be inspected, and executing step S203, namely determining that the wall surface to be inspected passes inspection. If yes, that is, if at least one preset defect type exists in the defect types of the image to be checked, step S202 is performed.
S202: and comparing the flaw grade corresponding to the flaw class with the same class with a preset flaw grade.
If the defect levels corresponding to the defect types with the same type are the first defect level of the preset defect type, S203: and determining that the wall surface to be checked is accepted by checking.
If the defect levels corresponding to the defect levels with the same category are at least a first defect level of the non-preset defect category, S204: and determining that the wall surface to be checked is not accepted.
And comparing the flaw grade corresponding to the flaw grade with the preset flaw grade, and if the flaw grade of the flaw corresponding to the flaw grade with the same grade in the flaw information is the first flaw grade of the preset flaw grade, namely the flaw sizes of the flaws of the wall surface to be inspected all belong to the size range corresponding to the first flaw grade, determining that the wall surface to be inspected passes inspection. Otherwise, if at least one flaw grade of flaws corresponding to the flaw categories with the same categories is not the first flaw grade of the preset flaw category, the wall surface to be inspected is considered to be not inspected.
It can be understood that the above embodiments are all schematically listed one type of acceptance criteria, and in actual working conditions, the corresponding content of the acceptance criteria may be set according to actual situations, which is not limited in this embodiment of the present application.
According to the wall surface acceptance method, whether the wall surface to be accepted is inspected or not is output by comparing the flaw type in the flaw information to be inspected with the flaw grade and the preset flaw type in the acceptance standard, so that the wall surface to be inspected is automatically inspected according to the acceptance standard without manual comparison, subjectivity and size deviation of the wall surface to be inspected in the prior art are overcome, the intellectualization and accuracy of the inspection result are guaranteed, and the reliability and inspection experience of the completion inspection result can be improved.
Fig. 5 is a schematic structural diagram of a wall inspection device according to an embodiment of the present application. As shown in fig. 5, a wall surface acceptance device 400 provided in an embodiment of the present application includes:
the acquisition module 401 is configured to acquire an image to be checked, where the image to be checked includes a plurality of images of a wall surface to be checked;
the extracting module 402 is configured to perform feature extraction on the image to be inspected according to a feature extraction model, so as to obtain defect information to be inspected of the wall to be inspected, where the feature extraction model is obtained by training a plurality of wall images with feature defects;
and the acceptance module 403 is configured to determine whether the wall surface to be accepted passes acceptance according to the flaw information to be accepted and the acceptance criterion.
Fig. 6 is a schematic structural diagram of another wall inspection device according to an embodiment of the present disclosure on the basis of fig. 5. As shown in fig. 6, the wall surface acceptance device 400 provided in the embodiment of the present application further includes: training module 404. The training module 404 is configured to:
acquiring a training data set, wherein the training data set comprises a plurality of wall surface images with characteristic flaws and design images corresponding to each wall surface image with the characteristic flaws;
and training the initial feature extraction model according to the training data set to obtain a feature extraction model.
In one possible design, the extraction module 402 includes:
the output module is used for inputting the image to be checked and the design image to be checked into the feature extraction model to perform feature extraction and outputting an extraction result;
the analysis module is used for analyzing the extraction result according to the size deviation strategy and determining the obtained analysis result as defect information to be checked;
the flaw information to be checked comprises flaw types and flaw grades of images to be checked, and the design images to be checked are original design images of the wall surfaces to be checked.
In one possible design, acceptance module 403 is specifically configured to:
judging whether the flaw class of the image to be checked is consistent with at least one of preset flaw classes;
if not, determining that the wall surface to be checked is accepted through checking;
if yes, determining whether the wall surface to be checked and accepted passes the checking and accepting according to the flaw grade corresponding to the flaw class with the same class and the preset flaw grade;
the acceptance criteria include a preset flaw class and a preset flaw level.
In one possible design, acceptance module 403 is also to:
comparing the flaw grade corresponding to the flaw class with the same class with a preset flaw grade;
if the flaw grades corresponding to the flaw grades with the same grade are all the first flaw grade of the preset flaw grade, determining that the wall surface to be checked is accepted through checking;
The preset flaw grade comprises a first flaw grade, a second flaw grade and a third flaw grade of each preset flaw class.
In one possible design, the predetermined flaw categories include one or more of a sand hole flaw category, a sink flaw category, and a pimple flaw category.
In one possible design, the parsing module is specifically configured to:
generating a corresponding class label for the flaw class of the image to be checked according to the flaw class label, wherein the class label is used for representing the flaw class of the image to be checked;
determining flaw grades of the images to be inspected according to flaw size ranges, wherein the flaw size ranges comprise size ranges corresponding to the first flaw grade, the second flaw grade and the third flaw grade of each preset flaw class;
the size deviation strategy comprises a flaw type label and a flaw size range.
In one possible design, the acquisition module 401 is specifically configured to:
shooting the wall surface to be checked by shooting equipment according to shooting standards, so as to acquire an image to be checked, wherein the shooting standards comprise shooting distances and shooting angles.
In one possible design, wall surface inspection device 400 further includes: and an authentication module. The authentication module is used for:
And authenticating the identity of the shooting equipment according to the equipment identifier, wherein the equipment identifier is used for uniquely identifying the shooting equipment.
The wall surface acceptance device provided in the embodiment of the application can execute corresponding steps of the wall surface acceptance method in the embodiment of the method, and the implementation principle and the technical effect are similar and are not repeated here.
Fig. 7 is a schematic structural diagram of an electronic device according to an embodiment of the present application. As shown in fig. 7, the electronic device 500 may include: a processor 501, and a memory 502 communicatively coupled to the processor 501.
A memory 502 for storing a program. In particular, the program may include program code including computer-executable instructions.
The memory 502 may comprise high-speed RAM memory or may further comprise non-volatile memory (NoN-volatile memory), such as at least one disk memory.
The processor 501 is configured to execute computer-executable instructions stored in the memory 502 to implement a wall surface acceptance method.
The processor 501 may be a central processing unit (Central Processing Unit, abbreviated as CPU), or an application specific integrated circuit (Application Specific Integrated Circuit, abbreviated as ASIC), or one or more integrated circuits configured to implement embodiments of the present application.
Alternatively, the memory 502 may be separate or integrated with the processor 501. When the memory 502 is a device separate from the processor 501, the electronic device 500 may further include:
a bus 503 for connecting the processor 501 and the memory 502. The bus may be an industry standard architecture (industry standard architecture, abbreviated ISA) bus, an external device interconnect (peripheral component, PCI) bus, or an extended industry standard architecture (extended industry standard architecture, EISA) bus, among others. Buses may be divided into address buses, data buses, control buses, etc., but do not represent only one bus or one type of bus.
Alternatively, in a specific implementation, if the memory 502 and the processor 501 are integrated on a chip, the memory 502 and the processor 501 may complete communication through an internal interface.
The present application also provides a computer-readable storage medium, which may include: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a random access Memory (RAM, random AccessMemory), a magnetic disk, or an optical disk, or other various media capable of storing program codes, and specifically, the computer-readable storage medium stores therein computer-executable instructions for use in the wall surface inspection method in the above-described embodiment.
The present application also provides a computer program product comprising computer-executable instructions that when executed by a processor implement the wall surface acceptance method of the above embodiments.
Other embodiments of the present application will be apparent to those skilled in the art from consideration of the specification and practice of the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of the application following, in general, the principles of the application and including such departures from the present disclosure as come within known or customary practice within the art to which the application pertains. It is intended that the specification and examples be considered as exemplary only, with a true scope and spirit of the application being indicated by the following claims.
It is to be understood that the present application is not limited to the precise arrangements and instrumentalities shown in the drawings, which have been described above, and that various modifications and changes may be effected without departing from the scope thereof. The scope of the application is limited only by the appended claims.

Claims (13)

1. A wall surface acceptance method, comprising:
collecting an image to be checked, wherein the image to be checked comprises a plurality of images of a wall surface to be checked;
performing feature extraction on the to-be-inspected image according to a feature extraction model to obtain to-be-inspected flaw information of the to-be-inspected wall surface, wherein the feature extraction model is obtained by training a plurality of wall surface images with feature flaws;
And determining whether the wall surface to be checked is accepted or not according to the flaw information to be checked and the acceptance standard.
2. The wall surface acceptance method according to claim 1, further comprising, before said feature extraction of said wall surface image to be accepted according to a feature extraction model:
acquiring a training data set, wherein the training data set comprises a plurality of wall surface images with characteristic flaws and design images corresponding to each wall surface image with the characteristic flaws;
and training the initial feature extraction model according to the training data set to obtain the feature extraction model.
3. The wall surface acceptance method according to claim 1, wherein the feature extraction is performed on the image to be accepted according to a feature extraction model to obtain flaw information to be accepted of the wall surface to be accepted, and the method comprises the following steps:
inputting the image to be checked and the design image to be checked into the feature extraction model to perform feature extraction, and outputting an extraction result;
analyzing the extraction result according to a size deviation strategy, and determining the obtained analysis result as the flaw information to be checked;
the flaw information to be checked comprises flaw types and flaw grades of the images to be checked, and the design images to be checked are original design images of the wall surfaces to be checked.
4. A wall surface acceptance method according to claim 3, wherein said determining whether said wall surface to be accepted is accepted according to said flaw information to be accepted and acceptance criteria comprises:
judging whether the flaw category of the image to be checked is consistent with at least one of preset flaw categories;
if not, determining that the wall surface to be checked is accepted through checking;
if yes, determining whether the wall surface to be checked is accepted or not according to the flaw grade corresponding to the flaw category with the same category and the preset flaw grade;
wherein the acceptance criteria include the preset flaw category and the preset flaw level.
5. The wall surface acceptance method according to claim 4, wherein said determining whether the wall surface to be accepted is accepted according to the flaw level corresponding to the flaw with the same category and the preset flaw level comprises:
comparing the flaw grade corresponding to the flaw class with the same class with the preset flaw grade;
if the flaw grades corresponding to the flaw grades with the same grade are all first flaw grades of the preset flaw grade, determining that the wall surface to be checked is accepted through checking;
the preset flaw grade comprises a first flaw grade, a second flaw grade and a third flaw grade of each preset flaw class.
6. The wall acceptance method of claim 5, wherein the predetermined flaw categories include one or more of a sand hole flaw category, a sink flaw category, and a pimple flaw category.
7. The wall acceptance method of claim 6, wherein said parsing said extraction result according to a size deviation strategy comprises:
generating a corresponding class label for the flaw class of the image to be checked according to the flaw class label, wherein the class label is used for representing the flaw class of the image to be checked;
determining flaw grades of the image to be inspected according to flaw size ranges, wherein the flaw size ranges comprise size ranges corresponding to the first flaw grade, the second flaw grade and the third flaw grade of each preset flaw class;
wherein the size deviation strategy comprises the flaw class label and the flaw size range.
8. The wall surface acceptance method according to any one of claims 1 to 7, wherein said acquiring an image to be accepted comprises:
shooting the wall surface to be checked by shooting equipment according to shooting standards so as to acquire the image to be checked, wherein the shooting standards comprise shooting distances and shooting angles.
9. The wall surface acceptance method according to claim 8, further comprising, before said feature extraction of said image to be accepted according to a feature extraction model:
and carrying out identity authentication on the shooting equipment according to equipment identification, wherein the equipment identification is used for uniquely identifying the shooting equipment.
10. A wall inspection device, comprising:
the acquisition module is used for acquiring an image to be checked and accepted, wherein the image to be checked and accepted comprises a plurality of images of a wall surface to be checked and accepted;
the extraction module is used for carrying out feature extraction on the images to be inspected according to a feature extraction model to obtain defect information to be inspected of the walls to be inspected, and the feature extraction model is obtained by training according to a plurality of wall images with feature defects;
and the acceptance module is used for determining whether the wall surface to be accepted is accepted or not according to the flaw information to be accepted and the acceptance standard.
11. An electronic device, comprising: a processor, and a memory communicatively coupled to the processor;
the memory stores computer-executable instructions;
the processor executes computer-executable instructions stored in the memory to implement the wall surface acceptance method of any one of claims 1-9.
12. A computer readable storage medium having stored therein computer executable instructions which when executed by a processor are adapted to carry out the wall surface acceptance method of any one of claims 1 to 9.
13. A computer program product comprising computer-executable instructions for implementing the wall surface acceptance method of any one of claims 1-9 when executed by a processor.
CN202310123195.8A 2023-02-16 2023-02-16 Wall surface acceptance method, device, equipment and storage medium Pending CN116071335A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202310123195.8A CN116071335A (en) 2023-02-16 2023-02-16 Wall surface acceptance method, device, equipment and storage medium

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202310123195.8A CN116071335A (en) 2023-02-16 2023-02-16 Wall surface acceptance method, device, equipment and storage medium

Publications (1)

Publication Number Publication Date
CN116071335A true CN116071335A (en) 2023-05-05

Family

ID=86169691

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202310123195.8A Pending CN116071335A (en) 2023-02-16 2023-02-16 Wall surface acceptance method, device, equipment and storage medium

Country Status (1)

Country Link
CN (1) CN116071335A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116798030A (en) * 2023-08-28 2023-09-22 中国建筑第六工程局有限公司 Curved surface sightseeing radar high tower acceptance method, system, device and storage medium

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116798030A (en) * 2023-08-28 2023-09-22 中国建筑第六工程局有限公司 Curved surface sightseeing radar high tower acceptance method, system, device and storage medium
CN116798030B (en) * 2023-08-28 2023-11-14 中国建筑第六工程局有限公司 Curved surface sightseeing radar high tower acceptance method, system, device and storage medium

Similar Documents

Publication Publication Date Title
CN111340787B (en) Method and device for detecting and identifying wire defects of power transmission line and computer equipment
WO2019214309A1 (en) Model test method and device
CN111242899B (en) Image-based flaw detection method and computer-readable storage medium
CN111582359B (en) Image identification method and device, electronic equipment and medium
CN111598827A (en) Appearance flaw detection method, electronic device and storage medium
CN112365461A (en) Fastener loosening identification method, system, terminal and storage medium
CN105718931A (en) System And Method For Determining Clutter In An Acquired Image
CN116071335A (en) Wall surface acceptance method, device, equipment and storage medium
CN114813757A (en) Full-angle appearance detection method and device for plastic product
CN113077416A (en) Welding spot welding defect detection method and system based on image processing
CN109102486B (en) Surface defect detection method and device based on machine learning
CN111223078A (en) Method for determining defect grade and storage medium
CN113096121A (en) Pavement crack detection method and system based on cross fracture mechanics and image processing
CN111259887B (en) Intelligent quality inspection method, system and device for dumb resource equipment
CN112884018A (en) Power grid line fault recognition model training method and power grid line inspection method
CN109544514B (en) Sawn timber identity identification method, device and equipment integrating apparent features
CN111760292A (en) Method and device for detecting sampling data and electronic equipment
KR101944604B1 (en) Apparatus and method for inspection automation to count object quantity
CN111460198A (en) Method and device for auditing picture timestamp
CN115127856A (en) Method and device for sampling and identifying concrete test block compression test robot
CN110634124A (en) Method and equipment for area detection
CN111640096B (en) Method, device and terminal for detecting appearance of electronic product
CN110045354B (en) Method and device for evaluating radar performance
CN112967224A (en) Electronic circuit board detection system, method and medium based on artificial intelligence
CN114155522A (en) Point cloud data quality inspection repairing method and system

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