CN114897886A - Industrial defect detection optimization method, system, device, equipment and storage medium - Google Patents
Industrial defect detection optimization method, system, device, equipment and storage medium Download PDFInfo
- Publication number
- CN114897886A CN114897886A CN202210689839.5A CN202210689839A CN114897886A CN 114897886 A CN114897886 A CN 114897886A CN 202210689839 A CN202210689839 A CN 202210689839A CN 114897886 A CN114897886 A CN 114897886A
- Authority
- CN
- China
- Prior art keywords
- detection
- defect
- defect detection
- model
- original image
- 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
Links
- 238000001514 detection method Methods 0.000 title claims abstract description 210
- 230000007547 defect Effects 0.000 title claims abstract description 145
- 238000000034 method Methods 0.000 title claims abstract description 32
- 238000005457 optimization Methods 0.000 title claims abstract description 17
- 238000003860 storage Methods 0.000 title claims abstract description 9
- 238000012549 training Methods 0.000 claims abstract description 19
- 230000002708 enhancing effect Effects 0.000 claims abstract description 11
- 238000002372 labelling Methods 0.000 claims abstract description 10
- 238000004590 computer program Methods 0.000 claims description 17
- 238000012360 testing method Methods 0.000 claims description 17
- 238000007781 pre-processing Methods 0.000 claims description 10
- 238000006243 chemical reaction Methods 0.000 claims description 8
- 238000012800 visualization Methods 0.000 claims description 8
- 238000012795 verification Methods 0.000 claims description 3
- 230000008569 process Effects 0.000 abstract description 8
- 230000006872 improvement Effects 0.000 abstract description 3
- 238000010586 diagram Methods 0.000 description 9
- 230000005540 biological transmission Effects 0.000 description 4
- 238000013135 deep learning Methods 0.000 description 4
- 230000002950 deficient Effects 0.000 description 3
- 238000003384 imaging method Methods 0.000 description 3
- 230000006870 function Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 239000003086 colorant Substances 0.000 description 1
- 238000005520 cutting process Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 238000009776 industrial production Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000004806 packaging method and process Methods 0.000 description 1
- 238000012856 packing Methods 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
Images
Classifications
-
- 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
- G06T7/0008—Industrial image inspection checking presence/absence
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T3/00—Geometric image transformations in the plane of the image
- G06T3/60—Rotation of whole images or parts thereof
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/11—Region-based segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/22—Image preprocessing by selection of a specific region containing or referencing a pattern; Locating or processing of specific regions to guide the detection or recognition
- G06V10/225—Image preprocessing by selection of a specific region containing or referencing a pattern; Locating or processing of specific regions to guide the detection or recognition based on a marking or identifier characterising the area
-
- 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
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20112—Image segmentation details
- G06T2207/20132—Image cropping
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30108—Industrial image inspection
- G06T2207/30164—Workpiece; Machine component
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Multimedia (AREA)
- Computing Systems (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Evolutionary Computation (AREA)
- Databases & Information Systems (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Quality & Reliability (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses an industrial defect detection optimization method, a system, a device, equipment and a storage medium, wherein the method comprises the steps of obtaining an original image, and marking defect sample data in the original image; enhancing the defect sample data to obtain an original data set; performing cloud online learning according to the original data set, and training to obtain a defect detection model; detecting an industrial product sample by using a defect detection model to obtain a detection result; and randomly selecting a plurality of detection results, judging whether the detection of the defect detection model on the industrial product sample is correct, when the detection result is correct, continuously adopting the defect detection model to detect the industrial product sample, and otherwise, re-labeling the defect sample data in the original image. When the deviation appears in the model detection result, the improvement can be made, the new effective sample information in the detection process is fully utilized, the situation that the performance of the model is continuously reduced in the application process is ensured, and the detection performance is improved.
Description
Technical Field
The invention belongs to the technical field of industrial defect detection, and relates to an industrial defect detection optimization method, system, device, equipment and storage medium.
Background
The defect detection is an important link in the industrial production process, the traditional defect detection scheme based on manual work has low detection efficiency, and the risk of missing detection exists. In recent years, a target detection model based on deep learning is gradually applied to industrial defect detection, semi-automation of industrial defect detection is realized, and detection efficiency is improved to a certain extent. The defect detection system based on deep learning gradually gets the attention of related practitioners, different forms of defect detection systems appear in succession, and as the existing target detection model based on deep learning needs a large amount of training data, in practical application, a large amount of sample data is often collected firstly, model training is carried out, a model meeting the requirements is manually screened out, the model can be deployed into the actual industrial defect detection system, and the automatic operation of the whole flow cannot be realized; the existing industrial defect detection system usually defaults that the detection precision of a deployed detection model always meets the requirement, however, due to various uncertain factors, the detection precision of the model is reduced to an unacceptable degree in the actual operation process of the system, and the performance evaluation in the operation process cannot be realized; the existing defect detection system based on deep learning usually adopts a mode of 'one-time deployment and lifelong use', and does not use a new effective sample acquired by the system in the running process.
Disclosure of Invention
The invention aims to solve the problem that the detection precision of a model cannot be found to be reduced in time in actual operation of a defect detection system in the prior art, and provides an industrial defect detection optimization method, system, device, equipment and storage medium.
In order to achieve the purpose, the invention adopts the following technical scheme to realize the purpose:
an industrial defect detection optimization method comprises the following steps:
acquiring an original image, and marking defect sample data in the original image;
enhancing the defect sample data to obtain an original data set;
performing cloud online learning according to the original data set, and training to obtain a defect detection model;
detecting an industrial product sample by using a defect detection model to obtain a detection result;
and randomly selecting a plurality of detection results, judging whether the detection of the defect detection model on the industrial product sample is correct, when the detection result is correct, continuously adopting the defect detection model to detect the industrial product sample, and otherwise, re-labeling the defect sample data in the original image.
The invention is further improved in that:
the method for enhancing the defect sample data comprises the following steps:
acquiring a random number in [0,1] as a gamma conversion parameter, and carrying out gamma conversion on the original image;
rotating the original image by a random angle around the central point of the original image;
acquiring one random number r in [0.7,1] as a proportional parameter, and acquiring two independent random numbers a and b in [0,1] as position parameters;
in the original image, random cropping is performed in a rectangular region having (a, b) points and (min (a + r,1), min (b + r,1)) points as diagonal lines.
The method comprises the steps of conducting cloud online learning according to an original data set, training to obtain a defect detection model, testing the detection precision of the defect detection model, and storing the defect detection model after the detection precision is qualified; and if the detection precision is unqualified, the cloud online learning is continued.
An industrial defect detection optimization system, comprising:
the data acquisition module is used for acquiring an original image and marking defect sample data in the original image;
the data enhancement module is used for enhancing the defect sample data to obtain an original data set;
the cloud server is used for carrying out cloud online learning according to the original data set and training to obtain a defect detection model;
the detection module is used for detecting the industrial product sample according to the defect detection model to obtain a detection result;
and the result verification module is used for randomly selecting a plurality of detection results, judging whether the detection of the defect detection model on the industrial product sample is correct, when the detection result is correct, continuously adopting the defect detection model to detect the industrial product sample, and otherwise, re-labeling the defect sample data in the original image.
An industrial defect detection optimization apparatus, comprising:
the client is used for acquiring an original image, preprocessing the original image to obtain an original data set, transmitting the original image to the inference server, transmitting the original data set to the online learning terminal and verifying a detection result of the inference server;
the online learning end carries out cloud online learning according to the original data set, trains to obtain a defect detection model, and transmits the defect detection model to the inference server;
and the reasoning server calls the defect detection model after receiving the original image and sends the detection result to the client.
The client comprises an image acquisition unit, an image preprocessing unit, a result visualization unit and a manual re-detection unit;
the image acquisition unit acquires an original image by adopting an industrial camera, transmits the original image to the image preprocessing unit for enhancement to obtain an original data set, and the result visualization unit is used for receiving a detection result of the reasoning server and verifying the detection result by the manual re-detection unit.
The online learning end comprises a first request management unit, a dynamic database unit, a model training unit and a model testing unit;
the first request management unit receives an original data set transmitted by a client, stores the original data set into the dynamic database unit, trains data in the dynamic database unit by the model training unit to obtain a defect detection model, and transmits the defect detection model to the inference server after the model test unit tests the detection precision.
The reasoning server comprises a second request management unit, a version management unit and a model loading unit;
the second request management unit receives the original image transmitted by the client, calls the defect detection model from the version management unit, loads the called defect detection model by the model loading unit, detects the original image and sends the detection result to the client.
An apparatus comprising a memory, a processor and a computer program stored in the memory and executable on the processor, the processor implementing the steps of the method of any one of the preceding claims when executing the computer program.
A computer-readable storage medium, having stored thereon a computer program which, when being executed by a processor, carries out the steps of the method according to any of the preceding claims.
Compared with the prior art, the invention has the following beneficial effects:
in the application process of the defect detection model, the detection result is subjected to spot check, when the detection result is correct, the model is continuously adopted for detection, when the detection result is wrong, the defect sample data is marked and the cloud end is studied online, when the model detection result has deviation, improvement can be made, the effective sample information which appears newly in the detection process is fully utilized, the situation that the performance of the model is continuously reduced in the application process is ensured, and the detection performance is improved.
Drawings
In order to more clearly explain the technical solutions of the embodiments of the present invention, the drawings needed to be used in the embodiments will be briefly described below, it should be understood that the following drawings only illustrate some embodiments of the present invention, and therefore should not be considered as limiting the scope, and for those skilled in the art, other related drawings can be obtained according to the drawings without inventive efforts.
FIG. 1 is a schematic flow chart of an industrial defect detection optimization method according to the present invention;
FIG. 2 is a detailed flow chart of the industrial defect detection optimization method of the present invention;
FIG. 3 is a schematic flow chart of an industrial defect detection optimization system according to the present invention;
FIG. 4 is a schematic structural diagram of an industrial defect detection optimizing apparatus according to the present invention;
FIG. 5 is a schematic diagram of a specific structure of a client in the present invention;
FIG. 6 is a schematic diagram of a specific structure of an online learning end according to the present invention;
fig. 7 is a schematic diagram of a specific structure of the inference server in the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some, but not all, embodiments of the present invention. The components of embodiments of the present invention generally described and illustrated in the figures herein may be arranged and designed in a wide variety of different configurations.
Thus, the following detailed description of the embodiments of the present invention, presented in the figures, is not intended to limit the scope of the invention, as claimed, but is merely representative of selected embodiments of the invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
It should be noted that: like reference numbers and letters refer to like items in the following figures, and thus, once an item is defined in one figure, it need not be further defined and explained in subsequent figures.
In the description of the embodiments of the present invention, it should be noted that if the terms "upper", "lower", "horizontal", "inner", etc. are used for indicating the orientation or positional relationship based on the orientation or positional relationship shown in the drawings or the orientation or positional relationship which is usually arranged when the product of the present invention is used, the description is merely for convenience and simplicity, and the indication or suggestion that the referred device or element must have a specific orientation, be constructed and operated in a specific orientation, and thus, cannot be understood as limiting the present invention. Furthermore, the terms "first," "second," and the like are used merely to distinguish one description from another, and are not to be construed as indicating or implying relative importance.
Furthermore, the term "horizontal", if present, does not mean that the component is required to be absolutely horizontal, but may be slightly inclined. For example, "horizontal" merely means that the direction is more horizontal than "vertical" and does not mean that the structure must be perfectly horizontal, but may be slightly inclined.
In the description of the embodiments of the present invention, it should be further noted that unless otherwise explicitly stated or limited, the terms "disposed," "mounted," "connected," and "connected" should be interpreted broadly, and may be, for example, fixedly connected, detachably connected, or integrally connected; can be mechanically or electrically connected; they may be connected directly or indirectly through intervening media, or they may be interconnected between two elements. The specific meanings of the above terms in the present invention can be understood by those skilled in the art according to specific situations.
The invention is described in further detail below with reference to the accompanying drawings:
referring to fig. 1, a schematic flow chart of an industrial defect detection optimization method in the present invention includes the following steps:
s1, acquiring an original image, and labeling defect sample data in the original image;
s2, enhancing the defect sample data to obtain an original data set;
s2.1, acquiring a random number in [0,1] as a gamma conversion parameter, and carrying out gamma conversion on the original image;
s2.2, rotating the original image by a random angle around the central point of the original image;
s2.3, acquiring one random number r in [0.7,1] as a proportional parameter, and acquiring two independent random numbers a and b in [0,1] as position parameters;
and S2.4, randomly cutting the rectangular area with the (a, b) point and the (min (a + r,1), min (b + r,1)) point as opposite angles in the original image.
S3, performing cloud online learning according to the original data set, and training to obtain a defect detection model;
s4, detecting the industrial product sample according to the defect detection model to obtain a detection result;
s5, randomly selecting a plurality of detection results, judging whether the detection of the defect detection model on the industrial product sample is correct, when the detection result is correct, continuing to detect the industrial product sample by adopting the defect detection model, otherwise, re-labeling the defect sample data in the original image.
Referring to fig. 2, a specific flowchart of the industrial defect detection optimization method of the present invention includes the following steps:
s1, initializing the system, entering a manual detection mode, and marking the sample data of the defect;
s2, when the number of the marked defect sample data reaches the set scale, enhancing the defect sample data to obtain an original data set;
s3, performing cloud online learning based on the original data set, and training to obtain a defect detection model;
s4, testing the defect detection model, and when the test result is higher than the set standard, entering the subsequent step, applying and deploying the defect detection model, and detecting the industrial product sample; otherwise, returning to S1, and continuing to collect effective defect sample information;
s5, when detecting the industrial product sample, storing the detected sample information, when the detected sample information reaches a certain scale, randomly selecting a plurality of samples, and manually judging whether the result of the model detection is correct;
s6, manually judging whether the detection result of the model is correct, and continuously applying the defect detection model to detect the industrial product sample; otherwise, the system sends out an alarm signal, returns to S1, and continues to collect effective defect sample information.
When enhancing the defect sample data at S2, the following two data enhancement strategies are adopted. One type of data enhancement strategy is similar to the traditional data enhancement strategy: firstly, selecting a random number as a parameter in image gamma conversion in an interval [0,1] of an obtained original image, and carrying out gamma conversion on the original image; then, rotating the original image by a random angle around the central point of the original image; then, a random number r is selected from [0.7,1] as a proportion parameter, two independent random numbers a and b are selected from [0,1] as position parameters, the original image is randomly cropped, and the cropped area is a rectangular area with the (a, b) point and the (min (a + r,1), min (b + r,1)) point of the original image as opposite angles. Another class of data enhancement strategies is: an image of a defective area in an existing defective sample is randomly copied into a non-defective sample image.
In S5, the sample information to be stored for detection includes two types: one is the defect-free sample of the model test, and the other is the defect-free sample of the model test. Keeping the ratio between the two constant, randomly selecting a plurality of samples from the detected samples, and artificially judging whether the detection result of the model is correct or not.
Referring to FIG. 3, a schematic diagram of an industrial defect detection optimizing system according to the present invention includes
The data acquisition module is used for acquiring an original image and marking defect sample data in the original image;
the data enhancement module is used for enhancing the defect sample data to obtain an original data set;
the cloud server is used for carrying out cloud online learning according to the original data set and training to obtain a defect detection model;
the detection module is used for detecting the industrial product sample according to the defect detection model to obtain a detection result;
and the result verification module is used for randomly selecting a plurality of detection results, judging whether the detection of the defect detection model on the industrial product sample is correct, when the detection result is correct, continuously adopting the defect detection model to detect the industrial product sample, and otherwise, re-labeling the defect sample data in the original image.
Referring to fig. 4, a schematic structural diagram of an industrial defect detection optimizing apparatus of the present invention includes:
the client is used for acquiring an original image, preprocessing the original image to obtain an original data set, transmitting the original image to the inference server, transmitting the original data set to the online learning terminal and verifying a detection result of the inference server;
the online learning end carries out cloud online learning according to the original data set, trains to obtain a defect detection model, and transmits the defect detection model to the inference server;
and the reasoning server calls the defect detection model after receiving the original image and sends the detection result to the client.
Referring to fig. 5, which is a schematic diagram of a specific structure of the client in the present invention, the client includes an image acquisition unit, an image preprocessing unit, a result visualization unit, and a manual re-detection unit. The image acquisition unit acquires an original image by adopting an industrial camera, transmits the original image to the image preprocessing unit for enhancement to obtain an original data set, and the result visualization unit is used for receiving a detection result of the reasoning server and verifying the detection result by the manual re-detection unit.
The image acquisition unit comprises a light source, an industrial camera and a supporting device, wherein the industrial camera is placed right above an industrial product to be detected through the supporting device and acquires a product image on a production line; the light source is used for improving the imaging environment of the industrial camera, so that the image acquired by the industrial camera can clearly show the defects on the surface of the product; the supporting device is used for fixing the industrial camera and the light source so as to ensure that the industrial camera can clearly and completely acquire a product image. In this embodiment, the industrial camera is a CMOS area-array imaging camera, and has an optical resolution of 5120 × 5120, a frame rate of 15.1fps, a pixel size of 2.5um × 2.5um, a pixel depth of 8bit, and an exposure time of 10us to 1 s; the light source adopts an area array light source, the size is 0.7 multiplied by 0.7m, the power input voltage is AC 90-264V, the power input frequency range is 47-63 Hz, the output current adjustment range is 0-8A, the brightness adjustment method is manual key type adjustment or PC end remote adjustment, the brightness adjustment mode is a constant current type, and the brightness adjustment grade is 256. The image preprocessing unit is a gamma enhancer program and performs gamma enhancement on the defect sample data, and the coefficient of the gamma enhancement is set to be 0.7 so as to improve the imaging effect. The result visualization unit comprises a subprogram for visualizing the detection result output by the reasoning server, a square frame is marked at the corresponding position of the original image, different types of defects are distinguished by different colors, and the types of the defects comprise three types: scratch, spot, and rusty. The manual re-detection unit comprises a subprogram for manually re-labeling the detection result, judges whether the fault detection model has false detection/missing detection or not based on the output of the result visualization unit manually, and directly clicks a submission button to store the final detection result if the false detection/missing detection does not occur; and if the false detection/missing detection occurs, correcting the false detection, re-labeling the missing detection, and finally storing the corrected defect information. And storing the final defect information into a text document, wherein an independent text document contains all the defect information of one industrial product image. Each line of the text document corresponds to a defect, and specifically comprises the type of the defect and the position of the defect. The file name is consistent with the name of the collected corresponding industrial product image file.
The specific working steps of the client are as follows:
s1, calling a subprogram for controlling a camera to shoot images in the large constant image SDK to obtain an original image;
s2, transmitting the original image to a cloud server, storing the original image in an original image database, calling a defect detection model, detecting defects, and storing the detection result in a model detection result database;
and S3, judging whether the number of the samples detected by the model reaches a set scale, extracting a plurality of samples from the model detection result database when the number of the samples detected by the model reaches the set scale, manually judging whether the detection result of the defect detection model is correct, if so, continuing to detect, and otherwise, entering a manual detection mode.
Referring to fig. 6, which is a schematic diagram of a specific structure of an online learning end in the present invention, the online learning end includes a first request management unit, a dynamic database unit, a model training unit, and a model testing unit; the first request management unit receives an original data set transmitted by a client, stores the original data set into the dynamic database unit, trains data in the dynamic database unit by the model training unit to obtain a defect detection model, and transmits the defect detection model to the inference server after the model test unit tests the detection precision.
The online learning end comprises the following specific working steps:
and S1, judging whether the data received by the data transmission interface is legal or not, and eliminating data loss caused by network and other reasons. When the data is legal, adding the data to the dynamic database unit, otherwise, discarding the data;
s2, judging whether the data volume in the dynamic database unit reaches a certain scale, when reaching the certain scale, performing on-line training of the model, otherwise, continuing to receive data from the data transmission interface;
s3, testing the trained model, and when the test result reaches the set standard, storing the defect detection model, transmitting the defect detection model to the inference server, and applying the model; otherwise, data continues to be received from the data transmission interface.
Referring to fig. 7, which is a schematic diagram of a specific structure of the inference server in the present invention, the inference server includes a second request management unit, a version management unit, and a model loading unit; the second request management unit receives the original image transmitted by the client, calls the defect detection model from the version management unit, loads the called defect detection model by the model loading unit, detects the original image and sends the detection result to the client.
The specific working steps of the reasoning server side are as follows:
s1, the request management unit receives the image data to be processed, analyzes the data and obtains the information of the image to be processed, the called detection model, the priority and the like;
s2, the version management unit inquires whether the detection model requested to be called is started, if so, the detection model of the version is directly called to be processed; otherwise, searching the model of the version, loading the detection model by the model loading unit, and detecting the image to be processed;
and S3, the reasoning server sends the result to the client through the data transmission interface.
The online learning end and the reasoning server end are constructed based on a TorchServe framework of Pytrch, and the specific steps are as follows:
s1, acquiring a model file, a weight file and an operation handle file according to a guide provided by TorchServe;
s2, packing the model file, the weight file and the operation handle file into an MAR file by using a scaffold tool provided by TorchServe;
and S3, packaging the obtained MAR file into a service with a RESTful API interface by using a scaffold tool provided by TorchServe.
An embodiment of the present invention provides a terminal device. The terminal device of this embodiment includes: a processor, a memory, and a computer program stored in the memory and executable on the processor. The processor implements the steps in the various industrial defect detection method embodiments described above when executing the computer program. Alternatively, the processor implements the functions of the modules/units in the above device embodiments when executing the computer program.
The computer program may be partitioned into one or more modules/units that are stored in the memory and executed by the processor to implement the invention.
The industrial defect detection device/terminal equipment can be computing equipment such as a desktop computer, a notebook computer, a palm computer and a cloud server. The industrial defect detection device/terminal equipment may include, but is not limited to, a processor, a memory.
The Processor may be a Central Processing Unit (CPU), other general purpose Processor, a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), an off-the-shelf Programmable Gate Array (FPGA) or other Programmable logic device, discrete Gate or transistor logic device, discrete hardware component, etc.
The memory may be used to store the computer programs and/or modules, and the processor may implement various functions of the industrial defect detection apparatus/terminal device by executing or executing the computer programs and/or modules stored in the memory and calling data stored in the memory.
The module/unit integrated with the industrial defect detecting apparatus/terminal device may be stored in a computer readable storage medium if it is implemented in the form of a software functional unit and sold or used as a separate product. Based on such understanding, all or part of the flow of the method according to the embodiments of the present invention may also be implemented by a computer program, which may be stored in a computer-readable storage medium, and when the computer program is executed by a processor, the steps of the method embodiments may be implemented. Wherein the computer program comprises computer program code, which may be in the form of source code, object code, an executable file or some intermediate form, etc. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, usb disk, removable hard disk, magnetic disk, optical disk, computer Memory, Read-Only Memory (ROM), Random Access Memory (RAM), electrical carrier wave signals, telecommunications signals, software distribution medium, and the like. It should be noted that the computer readable medium may contain content that is subject to appropriate increase or decrease as required by legislation and patent practice in jurisdictions, for example, in some jurisdictions, computer readable media does not include electrical carrier signals and telecommunications signals as is required by legislation and patent practice.
The above is only a preferred embodiment of the present invention, and is not intended to limit the present invention, and various modifications and changes will occur to those skilled in the art. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (10)
1. An industrial defect detection optimization method is characterized by comprising the following steps:
acquiring an original image, and marking defect sample data in the original image;
enhancing the defect sample data to obtain an original data set;
performing cloud online learning according to the original data set, and training to obtain a defect detection model;
detecting an industrial product sample by using a defect detection model to obtain a detection result;
and randomly selecting a plurality of detection results, judging whether the detection of the defect detection model on the industrial product sample is correct, when the detection result is correct, continuously adopting the defect detection model to detect the industrial product sample, and otherwise, re-labeling the defect sample data in the original image.
2. The industrial defect detection optimization method of claim 1, wherein enhancing the defect sample data comprises:
acquiring a random number in [0,1] as a gamma conversion parameter, and carrying out gamma conversion on the original image;
rotating the original image by a random angle around the central point of the original image;
acquiring one random number r in [0.7,1] as a proportional parameter, and acquiring two independent random numbers a and b in [0,1] as position parameters;
in the original image, random cropping is performed in a rectangular region having (a, b) points and (min (a + r,1), min (b + r,1)) points as diagonal lines.
3. The industrial defect detection optimization method of claim 1, wherein after performing cloud online learning according to the original data set and training to obtain the defect detection model, the method further comprises testing the detection accuracy of the defect detection model, and storing the defect detection model after the detection accuracy is qualified; and if the detection precision is unqualified, the cloud online learning is continued.
4. An industrial defect detection optimization system, comprising:
the data acquisition module is used for acquiring an original image and marking defect sample data in the original image;
the data enhancement module is used for enhancing the defect sample data to obtain an original data set;
the cloud server is used for carrying out cloud online learning according to the original data set and training to obtain a defect detection model;
the detection module is used for detecting the industrial product sample according to the defect detection model to obtain a detection result;
and the result verification module is used for randomly selecting a plurality of detection results, judging whether the detection of the defect detection model on the industrial product sample is correct, when the detection result is correct, continuously adopting the defect detection model to detect the industrial product sample, and otherwise, re-labeling the defect sample data in the original image.
5. An industrial defect detection optimizing apparatus, comprising:
the client is used for acquiring an original image, preprocessing the original image to obtain an original data set, transmitting the original image to the inference server, transmitting the original data set to the online learning terminal and verifying a detection result of the inference server;
the online learning end carries out cloud online learning according to the original data set, trains to obtain a defect detection model, and transmits the defect detection model to the inference server;
and the reasoning server calls the defect detection model after receiving the original image and sends the detection result to the client.
6. The industrial defect detection optimizing device of claim 5, wherein the client comprises an image acquisition unit, an image preprocessing unit, a result visualization unit and a manual re-detection unit;
the image acquisition unit acquires an original image by adopting an industrial camera, transmits the original image to the image preprocessing unit for enhancement to obtain an original data set, and the result visualization unit is used for receiving a detection result of the reasoning server and verifying the detection result by the manual re-detection unit.
7. The industrial defect detection optimizing device as claimed in claim 5, wherein the online learning terminal comprises a first request management unit, a dynamic database unit, a model training unit and a model testing unit;
the first request management unit receives an original data set transmitted by a client, stores the original data set into the dynamic database unit, trains data in the dynamic database unit by the model training unit to obtain a defect detection model, and transmits the defect detection model to the inference server after the model test unit tests the detection precision.
8. The industrial defect detection optimizing device of claim 5, wherein the inference server comprises a second request management unit, a version management unit and a model loading unit;
the second request management unit receives the original image transmitted by the client, calls the defect detection model from the version management unit, loads the called defect detection model by the model loading unit, detects the original image and sends the detection result to the client.
9. An apparatus comprising a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that the processor implements the steps of the method according to any of claims 1-3 when executing the computer program.
10. A computer-readable storage medium, in which a computer program is stored which, when being executed by a processor, carries out the steps of the method according to any one of claims 1 to 3.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210689839.5A CN114897886A (en) | 2022-06-17 | 2022-06-17 | Industrial defect detection optimization method, system, device, equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210689839.5A CN114897886A (en) | 2022-06-17 | 2022-06-17 | Industrial defect detection optimization method, system, device, equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114897886A true CN114897886A (en) | 2022-08-12 |
Family
ID=82729350
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210689839.5A Pending CN114897886A (en) | 2022-06-17 | 2022-06-17 | Industrial defect detection optimization method, system, device, equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114897886A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115326814A (en) * | 2022-09-06 | 2022-11-11 | 上汽通用五菱汽车股份有限公司 | Online intelligent identification and detection method and medium for manufacturing defects of automobile stamping parts |
| CN116843625A (en) * | 2023-06-05 | 2023-10-03 | 广东粤桨产业科技有限公司 | Defect detection model deployment method, system and equipment for industrial quality inspection scene |
-
2022
- 2022-06-17 CN CN202210689839.5A patent/CN114897886A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115326814A (en) * | 2022-09-06 | 2022-11-11 | 上汽通用五菱汽车股份有限公司 | Online intelligent identification and detection method and medium for manufacturing defects of automobile stamping parts |
| CN116843625A (en) * | 2023-06-05 | 2023-10-03 | 广东粤桨产业科技有限公司 | Defect detection model deployment method, system and equipment for industrial quality inspection scene |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN114897886A (en) | Industrial defect detection optimization method, system, device, equipment and storage medium | |
| CN113096119B (en) | Method and device for classifying wafer defects, electronic equipment and storage medium | |
| CN113095438A (en) | Wafer defect classification method and device, system, electronic equipment and storage medium thereof | |
| WO2022105019A1 (en) | Snapshot quality evaluation method and apparatus for vehicle bayonet device, and readable medium | |
| CN108871185B (en) | Method, device and equipment for detecting parts and computer readable storage medium | |
| CN113610788B (en) | Fault monitoring method and device for image acquisition device, electronic equipment and storage medium | |
| CN116109635B (en) | Method, device, equipment and medium for detecting surface quality of composite suspension insulator | |
| WO2022088660A1 (en) | Method and apparatus for improving wafer detection sensitivity, and storage medium | |
| CN113888480A (en) | MES-based quality tracing method and system | |
| CN117589770A (en) | PCB patch board detection method, device, equipment and medium | |
| CN117455887A (en) | PCB incoming material detection method, device, equipment and medium based on machine vision | |
| EP4085843A1 (en) | Correction method and apparatus for ring artifact in ct image, and computer program medium | |
| CN117437200A (en) | Steel wire surface abnormality detection method, device, equipment and storage medium | |
| CN115326832B (en) | Automatic detection method and automatic detection device for wafer package | |
| US8184898B2 (en) | Analysis of leaded components | |
| JP2011232302A (en) | Image inspection method and image inspection device | |
| US8306311B2 (en) | Method and system for automated ball-grid array void quantification | |
| CN115266774A (en) | Weld ray detection and evaluation method based on artificial intelligence | |
| CN104267203B (en) | The method of testing of a kind of sample and device | |
| CN113610155A (en) | Wafer defect classification method and device based on similarity comparison model, electronic equipment and storage medium | |
| JP2021148720A (en) | Inspection system, learning device, learning program, learning method, inspection device, inspection program, and inspection method | |
| US10325361B2 (en) | System, method and computer program product for automatically generating a wafer image to design coordinate mapping | |
| CN113011297B (en) | Power equipment detection method, device, equipment and server based on edge cloud cooperation | |
| WO2021192627A1 (en) | Inspection system, learning device, learning program, learning method, inspection device, inspection program, and inspection method | |
| CN117667735B (en) | Image enhancement software response time calibration device and method |
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 |