CN112748120B - Defect detection system, defect detection method, defect detection device, defect detection equipment and storage medium - Google Patents

Defect detection system, defect detection method, defect detection device, defect detection equipment and storage medium Download PDF

Info

Publication number
CN112748120B
CN112748120B CN202011602352.6A CN202011602352A CN112748120B CN 112748120 B CN112748120 B CN 112748120B CN 202011602352 A CN202011602352 A CN 202011602352A CN 112748120 B CN112748120 B CN 112748120B
Authority
CN
China
Prior art keywords
detected
defect
defect detection
light source
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.)
Active
Application number
CN202011602352.6A
Other languages
Chinese (zh)
Other versions
CN112748120A (en
Inventor
林义闽
龚向锋
廉士国
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 United Network Communications Group Co Ltd
Unicom Big Data Co Ltd
Original Assignee
China United Network Communications Group Co Ltd
Unicom Big Data 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 United Network Communications Group Co Ltd, Unicom Big Data Co Ltd filed Critical China United Network Communications Group Co Ltd
Priority to CN202011602352.6A priority Critical patent/CN112748120B/en
Publication of CN112748120A publication Critical patent/CN112748120A/en
Application granted granted Critical
Publication of CN112748120B publication Critical patent/CN112748120B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

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
    • G01N21/8806—Specially adapted optical and illumination features
    • 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
    • G01N21/01—Arrangements or apparatus for facilitating the optical investigation
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
    • G01N21/84—Systems specially adapted for particular applications
    • G01N21/88—Investigating the presence of flaws or contamination
    • G01N21/8851—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
    • G01N21/01—Arrangements or apparatus for facilitating the optical investigation
    • G01N2021/0181—Memory or computer-assisted visual determination
    • 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
    • G01N2021/8854—Grading and classifying of flaws
    • 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
    • G01N2021/8854—Grading and classifying of flaws
    • G01N2021/8861—Determining coordinates of flaws
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
    • G01N21/84—Systems specially adapted for particular applications
    • G01N21/88—Investigating the presence of flaws or contamination
    • G01N21/8851—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges
    • G01N2021/8883—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges involving the calculation of gauges, generating models
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
    • G01N21/84—Systems specially adapted for particular applications
    • G01N21/88—Investigating the presence of flaws or contamination
    • G01N21/8851—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges
    • G01N2021/8887—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges based on image processing techniques
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N2201/00—Features of devices classified in G01N21/00
    • G01N2201/06—Illumination; Optics
    • G01N2201/061—Sources
    • 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

  • Physics & Mathematics (AREA)
  • Health & Medical Sciences (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Chemical & Material Sciences (AREA)
  • Analytical Chemistry (AREA)
  • Biochemistry (AREA)
  • General Health & Medical Sciences (AREA)
  • General Physics & Mathematics (AREA)
  • Immunology (AREA)
  • Pathology (AREA)
  • Engineering & Computer Science (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Signal Processing (AREA)
  • Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)

Abstract

The application provides a defect detection system, a method, a device, equipment and a storage medium. The defect detection system comprises a plurality of light sources, a plurality of industrial cameras and a data processing terminal, wherein the light sources are parallel to the arrangement positions of the industrial cameras. The light source is used for providing illumination for the object to be detected, the first light source and the second light source are arranged on the upper surface of the object to be detected, the third light source is arranged on the lower surface of the object to be detected, and a preset angle is formed between the first light source and the second light source. And shooting the object to be detected in the corresponding illumination area by the industrial camera to obtain an image to be detected. And the data processing terminal determines a defect detection result according to the image to be detected. Be between first light source and the second light source and predetermine the angle and can form the cross light source, can guarantee that the defect of different grade type can both form images for the defect that exists is all shot and is obtained, has avoided lou examining, has guaranteed detection effect. The defects are detected based on the computer vision technology, so that the detection efficiency is improved, and the requirements of an automatic production process are met.

Description

Defect detection system, defect detection method, defect detection device, defect detection equipment and storage medium
Technical Field
The present application relates to the field of computer vision technologies, and in particular, to a defect detection system, a method, an apparatus, a device, and a storage medium.
Background
In the production process of textiles, the defect detection of the textiles to be delivered from the factory is an important link of quality control, and the precision of the defect detection directly influences the delivery quality of the textiles. At present, the defect detection of textiles is often completed manually, but in the face of heavy and repeated detection work, the manual efficiency is low, and the labor cost is high, so that the auxiliary detection through computer vision is very important.
At present, the mainstream detection method is to use an industrial camera to firstly shoot a textile, and then perform image recognition and detection based on the shot picture to determine whether corresponding defects exist. However, in actual working conditions, because the type and the position of the defect are unknown, the defect cannot be shot by the camera due to improper illumination, missing detection is caused, and the detection effect is not ideal.
Therefore, the existing textile defect detection scheme has certain technical defects, and a solution is needed to solve the above problems.
Disclosure of Invention
The application provides a defect detection system, a method, a device, equipment and a storage medium, which are used for solving the technical problem of unsatisfactory detection effect caused by missed detection due to improper light source setting in the existing textile defect detection scheme.
In a first aspect, the present application provides a defect detection system, comprising: the system comprises a plurality of light sources, a plurality of industrial cameras and a data processing terminal, wherein the setting positions of the industrial cameras are parallel to the setting positions of the light sources;
the light source is used for providing illumination for an object to be detected, a first light source and a second light source in the light source are respectively arranged on the upper surface of the object to be detected, a third light source in the light source is arranged on the lower surface of the object to be detected, and a preset angle is formed between the first light source and the second light source;
the industrial camera is used for shooting the object to be detected in the corresponding illumination area so as to obtain a corresponding image to be detected;
and the data processing terminal is used for determining a defect detection result according to the image to be detected.
In one possible design, the defect detection system further includes: an encoder and a collection card;
the encoder is used for acquiring the motion state of the object to be detected and reporting the motion state to the acquisition card;
the acquisition card is used for controlling the shooting parameters of each industrial camera according to the motion state and transmitting the image to be detected shot by each industrial camera to the data processing terminal.
In one possible design, the defect detection system further includes: a display;
the display is used for displaying the defect detection result, and the defect detection result comprises the type of the defect and the position data of the defect.
In one possible design, the defect detection system further includes: an array of LED lights;
the LED lamp array is used for positioning the position data of the defects to the object to be detected in a lighting mode.
In one possible design, when the long axis of the light source is perpendicular to the moving direction of the object to be detected, the length of the long axis exceeds the distance between opposite sides of the object to be detected.
In a second aspect, the present application provides a defect detection method applied to the defect detection system of any one of the first aspect, the method including:
acquiring an image to be detected of an object to be detected shot by each industrial camera;
and determining a defect detection result according to the target image detection model and the image to be detected, wherein the defect detection result comprises the type of the defect and the position data of the defect.
In one possible design, the determining a defect detection result according to the target image detection model and the image to be detected includes:
splicing the images to be detected shot by each industrial camera based on a preset calibration plate to obtain initial images under the same coordinate;
carrying out noise pretreatment on the initial image to obtain a target image;
and carrying out image detection on the target image through the target image detection model so as to determine the defect detection result according to the output detection result.
In a possible design, before determining the defect detection result according to the target image detection model and the image to be detected, the method further includes:
acquiring a training sample set, wherein the training sample set comprises a plurality of training images with preset defects, and each training image is marked with the corresponding preset defect through a defect label;
taking the training image as a training sample to carry out sample training on a preset image detection model;
and when the loss function of the preset image detection model is smaller than a preset threshold value or the result of the loss function is converged, finishing sample training, and determining the trained preset image detection model as the target image detection model.
In a third aspect, the present application provides a defect detection apparatus, comprising:
the acquisition module is used for acquiring an image to be detected of the object to be detected shot by each industrial camera;
and the processing module is used for determining a defect detection result according to the target image detection model and the image to be detected, wherein the defect detection result comprises the type of the defect and the position data of the defect.
In one possible design, the processing module is specifically configured to:
splicing the images to be detected shot by each industrial camera based on a preset calibration plate to obtain initial images under the same coordinate;
carrying out noise pretreatment on the initial image to obtain a target image;
and carrying out image detection on the target image through the target image detection model so as to determine the defect detection result according to the output detection result.
In one possible design, the apparatus further includes: a training module; the training module is configured to:
acquiring a training sample set, wherein the training sample set comprises a plurality of training images with preset defects, and each training image is marked with the corresponding preset defect through a defect label;
taking the training image as a training sample to carry out sample training on a preset image detection model;
and when the loss function of the preset image detection model is smaller than a preset threshold value or the result of the loss function is converged, finishing sample training, and determining the trained preset image detection model as the target image detection model.
In a fourth aspect, the present application provides an electronic device comprising:
a processor; and the number of the first and second groups,
a memory for storing a computer program for the processor;
wherein the processor is configured to perform any one of the possible defect detection methods provided by the second aspect via execution of the computer program.
In a fifth aspect, the present application provides a computer-readable storage medium, in which a computer program is stored, the computer program being configured to execute any one of the possible defect detection methods provided in the second aspect.
In a sixth aspect, the present application further provides a computer program product comprising a computer program, which when executed by a processor, implements any one of the possible defect detection methods provided in the second aspect.
The application provides a defect detection system, a method, a device, equipment and a storage medium. The defect detection system comprises a plurality of light sources, a plurality of industrial cameras and a data processing terminal, wherein the light sources are parallel to the arrangement positions of the industrial cameras. The light source is used for providing illumination for waiting to detect the thing, and first light source and second light source in the light source set up respectively in waiting to detect the upper surface of thing, and the third light source sets up in waiting to detect the lower surface of thing, is preset angle between first light source and the second light source. The industrial camera can shoot the object to be detected in the corresponding illumination area to obtain an image to be detected. The data processing terminal can determine a defect detection result according to an image to be detected shot by the industrial camera, so that the defect of the object to be detected can be automatically detected. Because the first light source and the second light source in the light source form a cross light source, the defects of different types possibly existing on the object to be detected can be ensured to be subjected to light imaging, and then the defects of different types can be shot by corresponding industrial cameras, so that the existence of missed detection is avoided, and the detection effect is ensured. And the defects of the object to be detected are detected based on the computer vision technology, so that the detection efficiency is effectively improved, the labor cost is reduced, and the requirements of an automatic production process are met.
Drawings
In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings needed to be used in the description of the embodiments or the prior art will be briefly introduced below, and it is obvious that the drawings in the following description are some embodiments of the present application, and for those skilled in the art, other drawings can be obtained according to these drawings without inventive exercise.
FIG. 1 is a schematic illustration of a light comparison provided in an embodiment of the present application;
FIG. 2 is a schematic diagram of a comparison of an embodiment of the present application;
fig. 3 is a schematic view of an application scenario provided in an embodiment of the present application;
fig. 4 is a schematic structural diagram of a defect detection system according to an embodiment of the present application;
fig. 5 is a schematic view of a light source arrangement assembly according to an embodiment of the present disclosure;
fig. 6 is a schematic flowchart of a defect detection method according to an embodiment of the present application;
FIG. 7 is a schematic flowchart of another defect detection method according to an embodiment of the present application;
fig. 8 is a diagram of a defective object provided in an embodiment of the present application;
FIG. 9 is a schematic flowchart illustrating a further defect detection method according to an embodiment of the present application;
fig. 10 is a schematic structural diagram of a defect detection apparatus according to an embodiment of the present application;
fig. 11 is a schematic structural diagram of an electronic device provided in the present application.
Detailed Description
Reference will now be made in detail to the exemplary embodiments, examples of which are illustrated in the accompanying drawings. When the following description refers to the accompanying drawings, like numbers in different drawings represent the same or similar elements unless otherwise indicated. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present application. Rather, they are merely examples of methods and apparatus consistent with certain aspects of the present application, as detailed in the appended claims.
The terms "first," "second," "third," "fourth," and the like in the description and in the claims of the present application and in the above-described drawings (if any) are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used is interchangeable under appropriate circumstances such that the embodiments of the application described herein are, for example, capable of operation in sequences other than those illustrated or otherwise described herein. 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, textile defect detection is usually completed manually, but the manual mode has the defects of low detection speed, low efficiency, high labor cost and the like, and the defect of the manual visual inspection mode is increasingly highlighted along with the increase of yield. Thus, assisted detection by computer vision appears to be of paramount importance. Nowadays, the mainstream detection method is to first shoot the textile by using an industrial camera, and then perform image recognition and detection based on the shot picture to determine whether the corresponding defect exists. However, in actual conditions, the location and type of the defect on the textile is unknown, for example, the location of the defect may be on the surface of the textile or inside the textile, and the type of defect may be snagging, staining, breakage, bumps or dents, holes, and the like. Therefore, improper illumination exists easily, defects cannot be shot by the camera, missed detection is caused, and the detection effect is not ideal.
For example, fig. 1 is a schematic view of illumination comparison provided in the embodiment of the present application, and fig. 2 is a schematic view of object comparison provided in the embodiment of the present application. Referring to fig. 1 and 2, wherein fig. 2 (a and B) are physical diagrams of images captured by the camera 12 under illumination shown in fig. 1 (a and B), respectively, the defect 10 is a protrusion or indentation defect. As shown in fig. 1 a and 2 a, when the direction of the light emitted from the light source 11 is perpendicular to the direction of the defect 10, the camera 12 can capture an image object 13 (a part marked within a white dotted line) corresponding to the defect 10. On the other hand, as shown in fig. 1 (B) and 2 (B), when the direction of the light emitted from the light source 11 is parallel to the direction of the defect 10, the camera 12 fails to capture the defect 10, i.e., the image of the defect 10 in the camera 12 disappears. Labels 14 and 15 in fig. 2 are used to mark defects. Therefore, the illumination setting is improper, the camera can not shoot some defects, the missed detection is caused, and the detection effect is not ideal.
In order to overcome the technical defects existing in the defect detection of textiles in the prior art, embodiments of the present application provide a defect detection system, a method, an apparatus, a device, and a storage medium. The invention conception provided by the application is as follows: the industrial camera is arranged in parallel with the arrangement position of the light sources, and the light sources arranged on the surface of the object to be detected are arranged in a crossed mode, namely, the first light source and the second light source which are arranged on the surface of the object to be detected are at preset angles. Therefore, the possibility that illumination is parallel to the defect direction can not occur, and further, various types of defects on the object to be detected can be guaranteed to be imaged in the corresponding industrial camera so as to be shot by the industrial camera, missing detection is avoided, the detection effect is guaranteed, and the detection precision is improved. In addition, the defect detection system also comprises a data processing terminal, which can determine the defect detection result according to the image to be detected corresponding to the object to be detected shot by the industrial camera, and realize the defect detection based on the computer vision technology, thereby effectively improving the detection efficiency, reducing the labor cost and being beneficial to the requirements of the automatic production process.
An exemplary application scenario of the embodiments of the present application is described below.
Fig. 3 is a schematic view of an application scenario provided by an embodiment of the present application, and as shown in fig. 3, the defect detection method provided by the embodiment of the present application may be executed by the defect detection apparatus provided by the embodiment of the present application, an electronic device corresponding to the defect detection apparatus provided by the embodiment of the present application may be a terminal device, a server, or a server cluster, and the server is illustrated in fig. 3 as an example. The defect detection method provided in the embodiment of the present application is applied to the defect detection system 21 provided in the embodiment of the present application, and the server 20 is connected to the defect detection system 21, for example, electrically connected or communicatively connected, so that the processor in the server 20 is configured to execute the defect detection method provided in the embodiment of the present application, so as to implement defect detection on an object to be detected.
It should be noted that the above application scenarios are only exemplary, and the defect detection system, method, apparatus, device, and storage medium system provided in the embodiments of the present application include, but are not limited to, the above application scenarios.
The following describes the technical solutions of the present application and how to solve the above technical problems with specific embodiments. The following several specific embodiments may be combined with each other, and details of the same or similar concepts or processes may not be repeated in some embodiments. Embodiments of the present application will be described below with reference to the accompanying drawings.
Fig. 4 is a schematic structural diagram of a defect detection system according to an embodiment of the present application. As shown in fig. 4, a defect detection system 30 provided in the embodiment of the present application includes: a plurality of light sources, a plurality of industrial cameras 32, and a data processing terminal 33, the installation positions of the industrial cameras 32 being parallel to the installation positions of the light sources.
The light sources are used for providing illumination for the object to be detected 34, a first light source 311 and a second light source 312 of the light sources are respectively arranged on the upper surface of the object to be detected 34, a third light source 313 of the light sources is arranged on the lower surface of the object to be detected 34, and a preset angle is formed between the first light source 311 and the second light source 312.
The industrial camera 32 is used for shooting the object to be detected 34 in the corresponding illumination area to obtain a corresponding image to be detected.
The data processing terminal 33 is configured to determine a defect detection result from the image to be detected.
For example, textile products such as cloth, leather, or fabric need to be defect-detected on a conveyor belt before being delivered from a factory, and the defect detection system 30 provided in the embodiment of the present application may be disposed on the conveyor belt to detect the defect of the textile product to be delivered from the factory when the conveyor belt conveys the textile product. The placement position and the fixing manner between the defect detection system 30 and the conveyor belt may be set correspondingly according to the position and the movement direction of the conveyor belt in actual working conditions, which is not limited in this embodiment. It should be noted that the moving direction of the conveyor belt in fig. 4 is illustrated by taking a direction from a to b in the object to be detected 34 as an example, where the moving direction from a to b is a direction perpendicular to the actual physical horizontal direction. In addition, the positions of the plurality of industrial cameras 32 and the processing terminal 33 in the defect detection system 30 shown in fig. 4 are merely schematic and do not represent actual positions in an actual working condition.
Fig. 4 shows a schematic configuration of the defect detection system 30, and referring to fig. 4, the defect detection system 30 includes a plurality of light sources, a plurality of industrial cameras 32, and a data processing terminal 33. The light sources are correspondingly arranged at the parallel position of each industrial camera, and the number of the industrial cameras 32 is at least the same as or more than that of the light sources, and the light sources can be specifically arranged according to the requirement on the resolution in the actual working condition. For example, after the setting position of each light source is determined, the industrial cameras may be adaptively set at positions parallel to each light source, respectively, so that the setting positions of the two are parallel. Each light source is used for providing corresponding illumination for the object to be detected 34 on the conveyor belt, and the industrial camera arranged in parallel with the corresponding light source is used for shooting the object to be detected 34 in the corresponding illumination area so as to obtain a corresponding image to be detected. The data processing terminal 33 may be a server, a computer, or other terminal device, and is configured to perform corresponding data processing, for example, to determine a defect detection result, such as a type of a defect and position data of the defect, according to an image to be detected captured by each industrial camera, so as to detect a possible defect on the object to be detected 34 based on a computer vision technology.
In order not to miss possible defects on the object 34, the shape of the light source is generally arranged as a strip, so that each inch of the object 34 can be detected. The industrial camera 32 may be a line-array industrial camera or an area-array industrial camera, which may be specifically selected according to actual conditions, and this embodiment is not limited thereto.
Defects of the object 34 to be detected may be present on the surface, and defects such as dirt, snagging, breakage, protrusions and/or indentations of the surface, etc. may often be present on the surface of the object 34 to be detected. Defects such as pinches, holes and holes may often be present inside the object 34 to be inspected. Thus, the first light source 311 and the second light source 312 are disposed on the upper surface of the object 34 to be detected to form a low-angle illumination light source for detecting defects that may occur on the surface of the object 34 to be detected. For defects possibly existing in the object to be detected 34, the third light source 313 of the light sources is arranged on the lower surface of the object to be detected 34, wherein the number of the third light sources 313 is at least one, and the specific arrangement mode of the third light sources 313 can be correspondingly arranged according to the size of the object to be detected 34 and the moving direction of the conveyor belt, for example, the long axis of the third light source 313 can be perpendicular to the running direction of the conveyor belt.
The first light source 311 and the second light source 312 are previously at a predetermined angle, and the predetermined angle is greater than 0 degree and less than 180 degrees. In addition, it should be noted that the number of the first light sources 311 is at least one, the number of the second light sources 312 is at least one, and the number may be specifically set according to a specific size of the object to be detected 34 in an actual working condition, generally, the size of the object to be detected 34 is 2 meters, and the length of the object to be detected is 1 meter, and if the size of the object to be detected 34 is larger, the number of the first light sources 311 and the number of the second light sources 312 are set according to an actual situation, for this reason, this embodiment is not limited, and the number of the second light sources 312 in fig. 4 is illustrated by two.
The first light source 311 and the second light source 312 are at a predetermined angle to form a cross light source. The presence of the crossed light source ensures that defects on the object 34 to be inspected can also be imaged by the corresponding industrial camera when their defect direction is parallel to the light source. For specific principle explanation, reference may be made to the foregoing description of the embodiment shown in fig. 1 and 2, which is not repeated herein.
Fig. 5 is a schematic view of a light source arrangement combination provided in an embodiment of the present application, and as shown in fig. 5, in a to D of fig. 5, assuming that the first light source 311 is arranged in a horizontal direction, the number of the second light sources 312 may be set to be multiple, for example, 2, so as to cover the full size of the object 34 to be detected. Without the horizontally arranged light sources in E and F of fig. 5, the number of the first light sources 311 and the second light sources 312 may be set to be plural, for example, 2, so as to cover the full size of the object to be detected 34 as fully as possible. It should be noted that the placement combination of the first light source 311 and the second light source 312 shown in fig. 5, which is shown in fig. 5 and is at a preset angle, is referred to from top to bottom in the moving direction of the conveyor belt, that is, the placement combination of the first light source 311 and the second light source 312 shown in fig. 5, and is suitable for the working condition that the moving direction of the conveyor belt is perpendicular to the actual physical horizontal direction. And when the moving direction of the conveyor belt is parallel to the actual physical horizontal direction in the actual working condition, the placing combination of the first light source 311 and the second light source 312 at the preset angle should be adaptively adjusted according to the actual condition. However, no matter the running direction of the conveyor belt, it is necessary to ensure that the first light source 311 and the second light source 312 form a predetermined angle therebetween. Therefore, the placing combination of the first light source 311 and the second light source 312 at a predetermined angle includes, but is not limited to, the combination shown in fig. 5.
It should be noted that fig. 4 illustrates the first light source 311 and the second light source 312 as C in fig. 5, and the third light source 313 is disposed with its long axis perpendicular to the moving direction of the conveyor belt.
The defect detection system provided by the embodiment of the application comprises a plurality of light sources, a plurality of industrial cameras and a data processing terminal, wherein the industrial cameras are parallel to the light sources in arrangement positions. The light source is used for providing illumination for waiting to detect the thing, and first light source and second light source in the light source set up respectively in waiting to detect the upper surface of thing, and the third light source sets up in waiting to detect the lower surface of thing, is preset angle between first light source and the second light source. The industrial camera can shoot the object to be detected in the corresponding illumination area to obtain a corresponding image to be detected. The data processing terminal can determine a defect detection result according to an image to be detected shot by the industrial camera, so that the defect of the object to be detected can be detected. Because the first light source and the second light source in the light source form a cross light source, the defects of various types on the object to be detected can be ensured to form images, and then the defects of different types can be shot by the industrial camera arranged in parallel with the light source, so that the missing detection behavior is avoided, and the detection effect is ensured. And the defect detection is completed based on the computer vision technology, so that the labor cost is reduced, the detection efficiency is effectively improved, and the requirements of the automatic production process are met.
In one possible design, in order to enable the light source to better cover the object to be detected on the conveyor belt, the length of the long axis of the light source may exceed the distance between opposite sides of the object to be detected to achieve good illumination coverage when the long axis of the light source is perpendicular to the moving direction of the object to be detected, i.e. when the light source is placed, if the long axis of the light source is perpendicular to the moving direction of the conveyor belt, as shown by the first light source 311 and the third light source 313 in fig. 4, the size of the long axis is larger than the distance L between opposite sides of the object to be detected 34. Of course, the specific corresponding value of the length of the long axis may be set according to the size of the object to be detected in the actual working condition, and the portion of the length of the long axis exceeding the opposite side distance L of the object to be detected may also be set according to the actual working condition, which is not limited in the embodiment of the present application.
With continued reference to fig. 4, in one possible design, the defect detection system 30 provided in the embodiment of the present application may further include: an encoder 35 and an acquisition card 36.
The encoder 35 is configured to acquire a motion state of the object 34, such as a motion direction, a motion speed, and the like of the object 34 on the conveyor belt. The encoder 35 (encoder) may be specifically a device that compiles, converts, and/or encodes signals (e.g., a bit stream) or data into a form of signals that can be communicated, transmitted, and stored, and can convert angular or linear displacements into electrical signals. In other words, the encoder 35 may be mounted on a corresponding component that drives the conveyor belt to convey, and converts the displacement of the object to be detected 34 during the movement into an electric signal by the synchronous movement with the conveyor belt, thereby acquiring the movement state of the object to be detected 34 by means of the electric signal. The specific specification of the encoder 35 may be selected accordingly according to the actual working condition, which is not limited in this embodiment.
After the encoder 35 acquires the motion state of the object 34 to be detected by means of an electrical signal, the motion state can be reported to the acquisition card 36. The acquisition card 36 is used in cooperation with the industrial cameras 32, and the acquisition card 36 can regulate and control the shooting parameters of each industrial camera in the industrial cameras 32 according to the motion state reported by the encoder 35, and can transmit the image to be detected shot by each industrial camera to the data processing terminal 33 for corresponding data processing. When the industrial camera 32 is a line camera, the shooting parameters may be parameters used for controlling the camera to shoot, such as line frequency. When the industrial camera 32 is an area-array camera, the shooting parameters may be frame rate or some other parameters for controlling the camera to shoot. The specific content included in the shooting parameters may be set according to the actual working condition, which is not limited in this embodiment. It is understood that the acquisition card 36 is connected to the encoder 35, the industrial camera 32 and the data processing terminal 33 respectively, and the connection manner may be an electrical connection or a communication connection, and may be specifically set according to an actual operating condition, which is not limited in this embodiment.
When the data processing terminal 33 determines the defect detection result, in order to facilitate the staff to view, in one possible design, the defect detection system 30 provided in this embodiment of the present application may further include: a display 37.
The display 37 is used for displaying the defect detection result including the type of the defect and the position data of the defect.
As shown with continued reference to fig. 4, a display 37, such as a screen, may also be provided after the data processing terminal 33 to display the defect detection results determined by the data processing terminal 33 on the display 37. Accordingly, the display 37 can also display a physical map of the detected defects for the staff to conveniently view. In addition, a display 37 is communicatively connected to the data processing terminal 33.
Further, the defect detection system 30 may further include: an array of LED lights 38.
As shown in fig. 4, two rows of LED lamps in the transverse and longitudinal directions may be disposed in the space directly above the object to be detected 34, and the position data of the defect is mapped onto the LED lamp array 38, so that the position data of the defect is positioned on the object to be detected 34 by lighting the lamp. For example, the position data of the defect is an actual coordinate (X, Y) in an actual space, the coordinate is mapped to a horizontal X-th LED lamp and a vertical Y-th LED lamp in the LED lamp array 38, the horizontal X-th LED lamp and the vertical Y-th LED lamp are turned on, and an intersection point position of the two LED lamps in a lighting state is an area of the defect on the object to be detected 34, as shown by a defect 39 in fig. 4, so that a worker can be helped to realize quick positioning of the defect. The LED lamp array 38 and the data processing terminal 33 may be electrically connected, or may be set according to actual conditions, which is not limited in this embodiment.
It should be noted that, in the embodiment of the present application, in two rows of LED lamps arranged transversely and longitudinally, the spacing distance between each LED lamp may be set according to the size of the object to be detected 34, for example, the spacing distance may be 10 centimeters, and this embodiment is not limited thereto.
In addition, the two rows of LED lamps can be replaced by the marker pens, so that the defects can be automatically identified on the surface, the follow-up manual search is facilitated, and the defects can be quickly removed.
The following describes in detail a specific implementation of the defect detection method applied to the defect detection system provided in the above embodiments.
Fig. 6 is a schematic flowchart of a defect detection method according to an embodiment of the present application, and as shown in fig. 6, the defect detection method according to the embodiment includes:
s101: and acquiring an image to be detected of the object to be detected shot by each industrial camera.
Each industrial camera can shoot an image to be detected of an object to be detected, specifically, a shooting area of each industrial camera is an illumination area which is parallel to the industrial camera and is used for providing illumination for the object to be detected by a light source, and the industrial camera can shoot the object to be detected in the illumination area so as to obtain a corresponding image to be detected. It can be seen that the image to be detected shot by each industrial camera is a partial image of the object to be detected. After the industrial camera shoots the images to be detected, each image to be detected can be obtained through the acquisition card.
S102: and determining a defect detection result according to the target image detection model and the image to be detected.
The defect detection result includes the type of the defect and the position data of the defect.
After the image to be detected of the object to be detected shot by each industrial camera is obtained, image recognition is carried out on the image to be detected based on the target image detection model so as to determine a defect detection result. The defect detection result comprises the type of the defect and position data of the defect, wherein the type of the defect is such as crease, wrinkle, crack, stain, hole and the like, and the position data of the defect is the corresponding position of the defect on the surface and/or inside of the object to be detected and can be represented in a coordinate mode.
In addition, the target image detection model may be a corresponding detection model of a neural network class, typically a corresponding model determined after deep learning training.
In a possible design, a possible implementation manner of step S102 is shown in fig. 7, and fig. 7 is a schematic flow diagram of another defect detection method provided in the embodiment of the present application, and as shown in fig. 7, in the defect detection method provided in the embodiment, determining a defect detection result according to a target detection model and an image to be monitored includes:
s1021: and splicing the images to be detected shot by each industrial camera based on a preset calibration plate to obtain initial images under the same coordinate.
The image to be detected shot by each industrial camera is a partial image of the object to be detected, so that the images to be detected need to be spliced based on a preset calibration plate to obtain an initial image under the same coordinate.
Specifically, the preset calibration plate is set in advance according to the relative position between each industrial camera and the object to be detected, and can represent the mapping relation between the image to be detected and the complete image of the object to be detected, which are shot by each industrial camera. Therefore, after each image to be detected is obtained, the images to be detected can be spliced based on the preset calibration plate and the width and height of each image to be detected, so as to obtain an initial image under a coordinate, namely a finished image of the object to be detected.
S1022: and carrying out noise preprocessing on the initial image to obtain a target image.
After the initial image is obtained, noise preprocessing is carried out on the initial image to enhance or filter and remove corresponding noise according to the actual situation of the initial image, and the initial image after the noise preprocessing is determined to be a target image, so that the target image is obtained. It is understood that the noise pre-processing may be performed by using corresponding noise processing software, and the specific content of the software is not limited in this embodiment.
S1023: and carrying out image detection on the target image through the target image detection model so as to determine a defect detection result according to the output detection result.
The target image detection model is a corresponding detection model capable of detecting defects after deep learning of a large number of preset defects. Therefore, a target image is input to the target image detection model as input data for image detection, and a corresponding detection result is output. If there is a defect in the target image, the output detection result includes the type of the detected defect and the position data of the defect, so that the defect detection result can be determined according to the detection result output by the target detection model. It will be appreciated that if no defects are present in the target image, the output detection results will not include any relevant data characterizing the defects.
Further, if a defect is detected in the target image, the defect may be labeled by a label frame, such as the black label frame of fig. 8, to label the defect. Fig. 8 is a defect object diagram provided in an embodiment of the present application.
The defect detection method provided by the embodiment of the application is applied to the defect detection system provided by the embodiment of the application, and the defect detection system comprises a plurality of light sources, a plurality of industrial cameras and a data processing terminal. A first light source and a second light source in the light sources are respectively arranged on the upper surface of the object to be detected, and a preset angle is formed between the first light source and the second light source, so that a cross light source is formed. The cross light source can ensure that the defects of different types can be shot by the industrial camera arranged in parallel with the light source, and the occurrence of missed detection is avoided, so that the detection effect is ensured. Based on this, the defect detection method provided by the embodiment of the application firstly obtains the image to be detected of the object to be detected shot by each industrial camera, and then determines the defect detection result according to the target image detection model and the image to be detected, wherein the determined defect detection result comprises the type of the detected defect and the position data of the defect, so that the textile defect detection is realized based on the computer vision technology, the labor cost is effectively reduced, the detection efficiency is improved, and the method is favorable for meeting the requirements of the automatic production process.
As described in the above embodiments, the target image detection model is a corresponding detection model capable of performing defect detection after deep learning of a large number of preset defects, and thus, in one possible design, before step S102, determining the target image detection model is further included. Fig. 9 is a schematic flowchart of another defect detection method provided in an embodiment of the present application, and as shown in fig. 9, before determining a defect detection result according to a target image detection model and an image to be detected, the defect detection method provided in this embodiment further includes:
s201: a training sample set is obtained.
The training sample set comprises a plurality of training images with preset defects, and each training image is marked with the corresponding preset defect through a defect label.
The method comprises the steps of collecting a large number of images of textile fabrics with preset defects through a camera, and determining the obtained images as a training sample set, wherein the images of the textile fabrics with the preset defects are training images, and the preset defects exist in each training image. In addition, the corresponding preset defects on the training image can be labeled through defect labels, such as a labeling frame. In order to enrich the data contained in the training sample set, the number of training images of the same type of preset defect may be at least 500. The type of the preset defect in this embodiment may be a crease, a fold, a crack, an offset, a raw material defect, a knot, a snag, an air hole, an eyelet, and the like, and the position of the preset defect on the training image is not limited.
S202: and taking the training image as a training sample to carry out sample training on the preset image detection model.
After the training sample set is obtained, the training images in the training sample set are used as training samples, and sample training is carried out on the preset image detection model through the training samples. The preset image detection model may be a convolutional neural network model, such as a neural network model like fast-RCNN or YOLO. Specifically, a training image is imported into a preset image detection model as a training sample, so that sample training is performed through deep learning of the preset image detection model. And after the sample is trained to a certain degree, determining the trained preset image detection model as a target image detection model.
S203: and when the loss function of the preset image detection model is smaller than the preset threshold value or the result of the loss function is converged, finishing the sample training, and determining the trained preset image detection model as the target image detection model.
And performing sample training on the preset image detection model by taking the training image as a training sample, and finishing the sample training process when the loss function of the preset image detection model is smaller than a preset threshold value or the result of the loss function is converged. And determining the trained preset image detection model as a target image detection model.
The loss function can represent the difference degree between the prediction and the actual data, and when the loss function is smaller than a preset threshold value or the result of the loss function is converged, it is indicated that the robustness of the preset image detection model reaches an expected effect, and the expected effect can be fed back through the preset threshold value. Specifically, the specific value corresponding to the preset threshold may be set according to the actual operating condition, and the specific condition when the result of the loss function reaches convergence may be determined according to the actual operating condition, which is not limited in this embodiment. And after the target image detection model is obtained, performing image detection on the image to be detected according to the target image detection model to determine a defect detection result. Therefore, the defects of the object to be detected which possibly exist are detected based on the computer vision technology, compared with a manual detection mode, the detection efficiency is improved, the detection cost is reduced, and the requirements of an automatic production process are met.
The following are embodiments of the apparatus of the present application that may be used to perform corresponding method embodiments of the present application. For details which are not disclosed in the embodiments of the apparatus of the present application, reference is made to the embodiments of the method corresponding to the present application.
Fig. 10 is a schematic structural diagram of a defect detection apparatus according to an embodiment of the present application, and as shown in fig. 10, the defect detection apparatus 100 according to the embodiment includes:
the acquiring module 101 is configured to acquire an image to be detected of the object to be detected, which is captured by each industrial camera.
And the processing module 102 is configured to determine a defect detection result according to the target image detection model and the image to be detected.
Wherein the defect detection result includes a type of the defect and position data of the defect.
In one possible design, the processing module 102 is specifically configured to:
splicing the images to be detected shot by each industrial camera based on a preset calibration plate to obtain initial images under the same coordinate;
carrying out noise pretreatment on the initial image to obtain a target image;
and carrying out image detection on the target image through the target image detection model so as to determine a defect detection result according to the output detection result.
In one possible design, the defect detection apparatus 100 provided in this embodiment further includes: and a training module. A training module to:
acquiring a training sample set, wherein the training sample set comprises a plurality of training images with preset defects, and each training image is marked with the corresponding preset defect through a defect label;
taking the training image as a training sample to carry out sample training on a preset image detection model;
and when the loss function of the preset image detection model is smaller than the preset threshold value or the result of the loss function is converged, finishing the sample training, and determining the trained preset image detection model as the target image detection model.
The defect detection apparatus provided in the foregoing embodiments of the present application may be configured to perform corresponding steps of the defect detection method provided in the foregoing embodiments, and specific implementation manners, principles, and technical effects are similar to those of the foregoing method embodiments, and are not described herein again.
It should be noted that the above device embodiments provided in this application are merely illustrative, and the module division is only one logic function division, and there may be other division ways in actual implementation. For example, multiple modules may be combined or may be integrated into another system. The coupling of the various modules to each other may be through interfaces that are typically electrical communication interfaces, but mechanical or other forms of interfaces are not excluded. Accordingly, modules illustrated as separate components may or may not be physically separate, may be located in one place, or may be distributed in different locations on the same or different devices.
Fig. 11 is a schematic structural diagram of an electronic device provided in the present application, and as shown in fig. 11, an electronic device 200 provided in this embodiment includes: at least one processor 201 and a memory 202, wherein fig. 11 illustrates an electronic device as an example of one processor.
A memory 202 for storing a computer program for the processor 201. In particular, the computer program may comprise program code comprising computer operating instructions.
Memory 202 may comprise high-speed RAM memory and may also include non-volatile memory (non-volatile memory), such as at least one disk memory.
The processor 201 is configured to execute the computer program stored in the memory 202 to implement the defect detection method provided by the above method embodiments.
The processor 201 may be a Central Processing Unit (CPU), an Application Specific Integrated Circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of the present application.
Alternatively, the memory 202 may be separate or integrated with the processor 201. When the memory 202 is a device independent from the processor 201, the electronic device 200 may further include:
a bus 203 for connecting the processor 201 and the memory 202. The bus may be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, an Extended ISA (EISA) bus, or the like. Buses may be classified as address buses, data buses, control buses, etc., but do not represent only one bus or type of bus.
Alternatively, in a specific implementation, if the memory 202 and the processor 201 are integrated on a chip, the memory 202 and the processor 201 may communicate through an internal interface.
The present application also provides a computer-readable storage medium, which may include: various media capable of storing program codes, such as a usb disk, a removable hard disk, a read-only memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and in particular, a computer program is stored in the computer-readable storage medium, and the computer program can be executed by a processor in the electronic device to implement the defect detection method in the embodiments.
The present application also provides a computer program product comprising a computer program which, when executed by a processor, implements the defect detection method in the embodiments described above.
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 invention 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 invention 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 will be understood that the present application is not limited to the precise arrangements described above and shown in the drawings and that various modifications and changes may be made without departing from the scope thereof. The scope of the application is limited only by the appended claims.

Claims (11)

1. A textile defect detection system, comprising: the system comprises a plurality of light sources, a plurality of industrial cameras and a data processing terminal, wherein the setting positions of the industrial cameras are parallel to the setting positions of the light sources;
the light source is used for providing illumination for an object to be detected, a first light source and a second light source of the light sources are respectively arranged on the upper surface of the object to be detected, a third light source of the light sources is arranged on the lower surface of the object to be detected, and a preset angle is formed between the first light source and the second light source;
the industrial camera is used for shooting the object to be detected in the corresponding illumination area so as to obtain a corresponding image to be detected;
the data processing terminal is used for determining a defect detection result of the textile according to the image to be detected, the defect detection result of the textile comprises the type of the defect and position data of the defect, wherein the type of the defect comprises a crease, a fold, a crack, stain, a raw material flaw, a knot, a crochet, an air hole and an eyelet, and the position data of the defect is the corresponding position of the defect on the surface and/or inside of the object to be detected;
the textile defect detection system is arranged on a conveyor belt and used for detecting the defects of the object to be detected when the object to be detected is conveyed on the conveyor belt; when the long axis of the light source is perpendicular to the moving direction of the object to be detected, the length of the long axis exceeds the distance between opposite sides of the object to be detected.
2. The defect detection system of claim 1, wherein the defect detection system of the textile further comprises: an encoder and a collection card;
the encoder is used for acquiring the motion state of the object to be detected and reporting the motion state to the acquisition card;
the acquisition card is used for controlling the shooting parameters of each industrial camera according to the motion state and transmitting the image to be detected shot by each industrial camera to the data processing terminal.
3. The defect detection system of claim 1, wherein the defect detection system of the textile further comprises: a display;
the display is used for displaying the defect detection result of the textile, and the defect detection result of the textile comprises the type of the defect and the position data of the defect.
4. The defect detection system of claim 3, wherein the defect detection system of the textile further comprises: an array of LED lights;
the LED lamp array is used for positioning the position data of the defects to the object to be detected in a lighting mode.
5. A method for detecting defects of a textile, the method being applied to a system for detecting defects of a textile according to any one of claims 1 to 4, the method comprising:
acquiring an image to be detected of an object to be detected, which is shot by each industrial camera arranged in parallel with each corresponding light source; when the long axis of the light source is vertical to the movement direction of the object to be detected, the length of the long axis exceeds the distance between opposite sides of the object to be detected;
determining a defect detection result of the textile according to the target image detection model and the image to be detected, wherein the defect detection result of the textile comprises the type of the defect and the position data of the defect, the type of the defect comprises a crease, a fold, a crack, fouling, a raw material flaw, a knot, a crochet, an air hole and an eyelet, and the position data of the defect is the corresponding position of the defect on the surface and/or inside of the object to be detected.
6. The method for detecting the defects of the textile according to claim 5, wherein the step of determining the defect detection result of the textile according to the target image detection model and the image to be detected comprises the following steps:
splicing the images to be detected shot by each industrial camera based on a preset calibration plate to obtain initial images under the same coordinate;
carrying out noise pretreatment on the initial image to obtain a target image;
and carrying out image detection on the target image through the target image detection model so as to determine a defect detection result of the textile according to the output detection result.
7. The method for detecting the defects of the textile according to claim 6, wherein before determining the defect detection result of the textile according to the target image detection model and the image to be detected, the method further comprises the following steps:
acquiring a training sample set, wherein the training sample set comprises a plurality of training images with preset defects, and each training image is marked with the corresponding preset defect through a defect label;
taking the training image as a training sample to carry out sample training on a preset image detection model;
and when the loss function of the preset image detection model is smaller than a preset threshold value or the result of the loss function is converged, finishing sample training, and determining the trained preset image detection model as the target image detection model.
8. A defect detection apparatus for a textile, comprising:
the acquisition module is used for acquiring an image to be detected of the object to be detected, which is shot by each industrial camera arranged in parallel with each corresponding light source; the textile defect detection system is arranged on the conveying belt and used for detecting the defects of the object to be detected when the object to be detected is conveyed on the conveying belt; when the long axis of the light source is vertical to the movement direction of the object to be detected, the length of the long axis exceeds the distance of the opposite sides of the object to be detected;
the processing module is used for determining a defect detection result of the textile according to the target image detection model and the image to be detected, wherein the defect detection result of the textile comprises the type of the defect and position data of the defect, the type of the defect comprises a crease, a fold, a crack, stain, a raw material flaw, a knot, a hooked yarn, an air hole and an eye, and the position data of the defect is the corresponding position of the defect on the surface and/or inside of the object to be detected.
9. An electronic device, comprising:
a processor; and (c) a second step of,
a memory for storing a computer program for the processor;
wherein the processor is configured to perform the defect detection method of any of claims 5 to 7 via execution of the computer program.
10. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the defect detection method according to any one of claims 5 to 7.
11. A computer program product comprising a computer program, characterized in that the computer program, when being executed by a processor, implements the defect detection method of any one of claims 5 to 7.
CN202011602352.6A 2020-12-29 2020-12-29 Defect detection system, defect detection method, defect detection device, defect detection equipment and storage medium Active CN112748120B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202011602352.6A CN112748120B (en) 2020-12-29 2020-12-29 Defect detection system, defect detection method, defect detection device, defect detection equipment and storage medium

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202011602352.6A CN112748120B (en) 2020-12-29 2020-12-29 Defect detection system, defect detection method, defect detection device, defect detection equipment and storage medium

Publications (2)

Publication Number Publication Date
CN112748120A CN112748120A (en) 2021-05-04
CN112748120B true CN112748120B (en) 2023-03-14

Family

ID=75647090

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202011602352.6A Active CN112748120B (en) 2020-12-29 2020-12-29 Defect detection system, defect detection method, defect detection device, defect detection equipment and storage medium

Country Status (1)

Country Link
CN (1) CN112748120B (en)

Families Citing this family (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113935995B (en) * 2021-12-16 2022-07-19 成都数之联科技股份有限公司 Image processing method and device, electronic equipment and computer readable storage medium
CN114708266A (en) * 2022-06-07 2022-07-05 青岛通产智能科技股份有限公司 Tool, method and device for detecting card defects and medium
CN115272340B (en) * 2022-09-29 2023-04-07 江苏智云天工科技有限公司 Industrial product defect detection method and device
CN115457034B (en) * 2022-11-09 2023-03-24 南京云创大数据科技股份有限公司 Method and device for detecting surface defects of mirror-like workpiece
CN115661137B (en) * 2022-12-12 2023-12-22 宁德时代新能源科技股份有限公司 Detection device, detection method, terminal device, and computer-readable storage medium
CN115656197A (en) * 2022-12-14 2023-01-31 宁德时代新能源科技股份有限公司 Optical detection device, detection method and device, electronic device and storage medium
CN115876784B (en) * 2023-01-31 2023-05-26 眉山博雅新材料股份有限公司 Workpiece defect detection method, system and equipment
CN117078677B (en) * 2023-10-16 2024-01-30 江西天鑫冶金装备技术有限公司 Defect detection method and system for starting sheet

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108108768A (en) * 2017-12-29 2018-06-01 清华大学 Photovoltaic glass defect classification method and device based on convolutional neural networks
CN108240990A (en) * 2018-03-22 2018-07-03 深圳市永光神目科技有限公司 A kind of scolding tin defect detecting system
CN110404816A (en) * 2019-07-29 2019-11-05 海南梯易易智能科技有限公司 A kind of 3D bend glass defect detecting device and method based on mechanical arm
CN110567977A (en) * 2019-10-11 2019-12-13 湖南讯目科技有限公司 Curved glass defect detection system and method
CN110632087A (en) * 2019-11-06 2019-12-31 合肥泰禾光电科技股份有限公司 Surface defect detection device
CN111896550A (en) * 2020-03-31 2020-11-06 广西师范大学 Surface defect detection device and method

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109490313B (en) * 2018-11-09 2022-08-23 中国科学院光电技术研究所 Automatic detection device and method for surface defects of large-caliber curved surface optical element
CN210572007U (en) * 2019-08-16 2020-05-19 研祥智能科技股份有限公司 LED support defect image capturing device and detection equipment

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108108768A (en) * 2017-12-29 2018-06-01 清华大学 Photovoltaic glass defect classification method and device based on convolutional neural networks
CN108240990A (en) * 2018-03-22 2018-07-03 深圳市永光神目科技有限公司 A kind of scolding tin defect detecting system
CN110404816A (en) * 2019-07-29 2019-11-05 海南梯易易智能科技有限公司 A kind of 3D bend glass defect detecting device and method based on mechanical arm
CN110567977A (en) * 2019-10-11 2019-12-13 湖南讯目科技有限公司 Curved glass defect detection system and method
CN110632087A (en) * 2019-11-06 2019-12-31 合肥泰禾光电科技股份有限公司 Surface defect detection device
CN111896550A (en) * 2020-03-31 2020-11-06 广西师范大学 Surface defect detection device and method

Also Published As

Publication number Publication date
CN112748120A (en) 2021-05-04

Similar Documents

Publication Publication Date Title
CN112748120B (en) Defect detection system, defect detection method, defect detection device, defect detection equipment and storage medium
US11449980B2 (en) System and method for combined automatic and manual inspection
JP5947169B2 (en) Appearance inspection apparatus, appearance inspection method and program
CN108465648A (en) A kind of magnetic core Automated Sorting System based on machine vision
CN102393397B (en) System and method for detecting surface defects of magnetic shoe
JP2017096750A (en) Positioning method, positioning apparatus, program and computer readable recording medium
CN202770781U (en) Fabric defect online detection device based on machine vision
CN110102511A (en) A kind of vision detection system and method for product appearance
US20180276811A1 (en) Automatic optical inspection system and operating method thereof
CN109416327B (en) Machine vision method and system for monitoring a manufacturing process
CN102175692A (en) System and method for detecting defects of fabric gray cloth quickly
US8401272B2 (en) Patterned wafer defect inspection system and method
CN103785622A (en) Machine-vision-based part sorting device and method
CN102879404B (en) System for automatically detecting medical capsule defects in industrial structure scene
WO2017071406A1 (en) Method and system for detecting pin of gold needle element
CN104568985A (en) AOI (automatic optic inspection) equipment
CN114022436B (en) Detection method, device and detection equipment for printed circuit board
CN108144865A (en) Automobile oil pipe Rough Inspection system and its detection method
EP3699578A1 (en) System for improved fabric inspection
CN104422694A (en) Processing device and processing method of measured data as well as optical measurement system
CN206832700U (en) The defects of one kind is based on infrared distance measuring sensor detection means
KR20130128923A (en) Method and apparatus for tracking mounting error of chip mounter
KR20190061660A (en) Device and method of product inspection based on foreground image
KR101742260B1 (en) auto conveying type vision inspection device for a harness
CN109115780A (en) For monitoring the method and device thereof of multiple automated optical detection equipments

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant