CN113267506A - Wood board AI visual defect detection device, method, equipment and medium - Google Patents

Wood board AI visual defect detection device, method, equipment and medium Download PDF

Info

Publication number
CN113267506A
CN113267506A CN202110619136.0A CN202110619136A CN113267506A CN 113267506 A CN113267506 A CN 113267506A CN 202110619136 A CN202110619136 A CN 202110619136A CN 113267506 A CN113267506 A CN 113267506A
Authority
CN
China
Prior art keywords
wood board
defect detection
visual defect
assembly line
wood
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN202110619136.0A
Other languages
Chinese (zh)
Inventor
杨龙
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.)
Individual
Original Assignee
Individual
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 Individual filed Critical Individual
Priority to CN202110619136.0A priority Critical patent/CN113267506A/en
Publication of CN113267506A publication Critical patent/CN113267506A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
    • G01N21/84—Systems specially adapted for particular applications
    • G01N21/88—Investigating the presence of flaws or contamination
    • G01N21/8806—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
    • G01N21/84—Systems specially adapted for particular applications
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
    • G01N21/84—Systems specially adapted for particular applications
    • G01N21/88—Investigating the presence of flaws or contamination
    • G01N21/8851—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
    • G01N21/84—Systems specially adapted for particular applications
    • G01N2021/8411—Application to online plant, process monitoring
    • 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
    • G01N2021/845—Objects on a conveyor
    • 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
    • G01N2021/8841—Illumination and detection on two sides of object
    • 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
    • G01N2021/8858—Flaw counting
    • 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
    • G01N2021/8867—Grading and classifying of flaws using sequentially two or more inspection runs, e.g. coarse and fine, or detecting then analysing
    • G01N2021/887—Grading and classifying of flaws using sequentially two or more inspection runs, e.g. coarse and fine, or detecting then analysing the measurements made in two or more directions, angles, positions
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
    • G01N21/84—Systems specially adapted for particular applications
    • G01N21/88—Investigating the presence of flaws or contamination
    • G01N21/8851—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges
    • G01N2021/8887—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges based on image processing techniques

Landscapes

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

Abstract

The invention discloses a wood board AI visual defect detection device, a method, equipment and a medium, wherein the device comprises a front detection assembly line and a back detection assembly line, the front detection assembly line and the back detection assembly line are connected through a turnover mechanism, a swinging disc feeding mechanism is arranged at the inlet of the front detection assembly line, the front detection assembly line is provided with a front visual defect detection device, and the back detection assembly line is provided with a back visual defect detection device. According to the invention, through the integration of the automatic system and the vision system, the system omits the process of manually classifying the wood boards, improves the production efficiency and the classification accuracy, and simultaneously provides guarantee for quality tracing.

Description

Wood board AI visual defect detection device, method, equipment and medium
Technical Field
The invention belongs to the technical field of machine vision detection, and particularly relates to a device, a method, equipment and a medium for detecting AI visual defects of a wood board.
Background
With the rapid development of automated production in modern manufacturing industry, many industries have higher requirements for inspection and measurement in industrial production. For example, inspection of a print packaging process, inspection of a package of a semiconductor chip, inspection of a product quality in a production line of a factory, inspection of high-precision parts, and the like. In these applications, most automated plants require high volume production, with part of the assembly accuracy requirements being very high. The traditional manual detection method can not meet the current process requirements, and the development and progress of the manufacturing industry are limited to a great extent. On the one hand, the traditional manual detection method has low efficiency, high error rate and high labor cost; on the other hand, the physiological limits of the human eye also result in human being unable to achieve the accuracy of computer detection techniques in this respect. The rapidity, reliability and accuracy of the computer are combined with the intellectualization of human vision, so that the machine vision is more and more widely applied to industrial detection.
Machine vision mainly uses a computer to simulate the visual function of a human, but not only is the simple extension of human eyes, but also has a part of functions of human brain, namely, information is extracted from the image of an objective object, processed and understood, and finally the information is used for actual detection, measurement and control.
Wood defects refer to various defects present in wood that can affect the quality and use value of the wood. Including wood natural defects, drying defects, processing defects. The manual detection method is adopted to detect the processing defects, so that the efficiency is low, the error rate is high, and a large amount of labor cost is consumed.
In the process of cutting the wood board, N different types of defective wood boards usually appear on the production line of a factory. Because each wood board has special texture, the traditional factory usually adopts a manual experience judgment method when classifying the defects of the wood boards, the method has low efficiency, the defects are difficult to trace, and the like, and the streamlined operation cannot be realized.
Disclosure of Invention
The invention aims to overcome the defects of the prior art and provides a device, a method, equipment and a medium for detecting the AI visual defects of the wood board.
The purpose of the invention is realized by the following technical scheme:
plank AI visual defect detection device includes: the front side detection assembly line and the back side detection assembly line are connected through the turnover mechanism, the inlet of the front side detection assembly line is provided with the swing disc feeding mechanism, the front side detection assembly line is provided with the front side visual defect detection device, and the back side detection assembly line is provided with the back side visual defect detection device.
Furthermore, the outlet of the reverse side detection production line is also connected with a rejection production line, and the rejection production line is provided with one or more rejection stations.
Further, the front visual defect detection device comprises a front line scanning camera, a light source and a bracket for fixing the front line scanning camera and the light source on the production line; the reverse side visual defect detection device comprises a reverse side scanning camera, a light source and a bracket for fixing the reverse side scanning camera and the light source on the production line; the front side line scanning camera and the back side line scanning camera are in communication connection with the data processing center.
On the other hand, the invention also provides a wood board AI visual defect detection method, which comprises the following steps:
s11: placing the wood board to be detected on a swinging plate feeding mechanism, and performing high-definition imaging when the wood board passes through a front visual defect detection device of a front detection assembly line;
s12: preprocessing the high-definition image, detecting and positioning, and sending the high-definition image into a board visual defect detection model for AI algorithm defect identification;
s13: if the wood board is defective, recording is carried out, the defective wood board is removed when passing through a removal assembly line, if the wood board is not defective, the wood board enters a reverse side detection assembly line through a turnover mechanism, and high-definition imaging is carried out when the wood board passes through a reverse side visual defect detection device of the reverse side detection assembly line;
s14: step S12 is executed, if the wood board is defective, recording is carried out, the defective wood board is removed when passing through the removing assembly line, and if the wood board is not defective, the step S15 is executed;
s15: when the front surface and the back surface of the wood board are not detected with defects, the wood board enters a production line.
Further, the specific construction method of the wood board visual defect detection model comprises the following steps:
s21: collecting the data of the picture of the surface of the shot wood board by a front visual defect detection device or a back visual defect detection device;
s22: carrying out data annotation on the picture data on the surface of the wood board;
s23: and training the marked data to obtain a wood board visual defect detection model.
Further, the wood board surface picture data comprise a wood board defect picture and a normal wood board picture.
Further, the method also comprises an accuracy verification step:
s24: deploying the model to a data processing center;
s25: carrying out defect detection on the plurality of wood boards by using the wood board visual defect detection model;
s26: the detection accuracy of the wood board visual defect detection model is rechecked, the absolute value of the error is qualified within 1%, otherwise, the absolute value is unqualified;
s27: and judging whether the detection accuracy of the wood board visual defect detection model reaches 95%, if so, determining that the wood board visual defect detection model is qualified, otherwise, determining that the wood board visual defect detection model is unqualified.
Further, the AI algorithm defect identification performed by the wood board visual defect detection model specifically comprises:
s121: inputting the preprocessed wood floor picture into a pre-trained ResNet50 neural network to obtain a corresponding feature map;
s122: setting an ROI corresponding to the wood floor picture for each point in the obtained feature map, so that each wood floor picture can obtain a plurality of candidate ROIs;
s123: sending the candidate ROI into an RPN network for binary classification and BB regression, and filtering out partial candidate ROI;
s124: ROIAlign operation is carried out on the rest ROIs;
s125: and classifying the rest ROI, BB regression and MASK generation to obtain a final defect identification result.
In another aspect, the present invention further provides a computer apparatus comprising a processor and a memory, wherein the memory stores a computer program, and the computer program is loaded and executed by the processor to implement the wood board AI visual defect detection method according to any one of the above.
In another aspect, the present invention further provides a computer-readable storage medium, in which a computer program is stored, the computer program being loaded and executed by a processor to implement the wood board AI visual defect detection method according to any one of the above.
The invention has the beneficial effects that:
according to the invention, through the integration of the automatic system and the vision system, the system omits the process of manually classifying the wood boards, improves the production efficiency and the classification accuracy, and simultaneously provides guarantee for quality tracing.
The method has low implementation cost, strong flexibility and expandability, and the accuracy of the whole set of system can be continuously improved in the operation process. Enterprises can realize self-growth on the system according to own actual conditions, and the economic value brought by a production line is improved while the labor cost is reduced.
Drawings
FIG. 1 is a schematic projection diagram of a wood board AI visual defect inspection device provided in embodiment 1 of the present invention;
FIG. 2 is a schematic side view of a wood board AI visual defect inspection device provided in example 1 of the present invention;
FIG. 3 is a block diagram of a flow chart of a wood board AI visual defect detection method provided in embodiment 2 of the invention;
FIG. 4 is a block diagram of a flow chart of an algorithm model design in the wood board AI visual defect detection method provided in embodiment 2 of the present invention;
fig. 5 is a block diagram of a flow of AI algorithm defect identification performed by a board visual defect detection model in the board AI visual defect detection method provided in embodiment 2 of the present invention.
Reference numerals: 1-a swing plate feeding mechanism, 2-a front visual defect detection device, 3-a front detection assembly line, 4-a turnover mechanism, 5-a back visual defect detection device and 6-a back detection assembly line.
Detailed Description
The embodiments of the present invention are described below with reference to specific embodiments, and other advantages and effects of the present invention will be easily understood by those skilled in the art from the disclosure of the present specification. The invention is capable of other and different embodiments and of being practiced or of being carried out in various ways, and its several details are capable of modification in various respects, all without departing from the spirit and scope of the present invention. It is to be noted that the features in the following embodiments and examples may be combined with each other without conflict.
It should be noted that, 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 are clearly and completely described below, and it is obvious that the described embodiments are some embodiments of the present invention, but not all embodiments.
Thus, the following detailed description of the embodiments of the present invention 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.
Example 1
As shown in the attached drawings 1 and 2, the device is a projection schematic view and a side schematic view of the wood board AI visual defect detection device provided by the embodiment, and specifically comprises a front detection assembly line and a back detection assembly line, wherein the front detection assembly line and the back detection assembly line are connected through a turnover mechanism, a swinging disc feeding mechanism is arranged at the inlet of the front detection assembly line, the front detection assembly line is provided with a front visual defect detection device, and the back detection assembly line is provided with a back visual defect detection device. The outlet of the reverse side detection production line is also connected with a rejection production line, the rejection production line is provided with one or more rejection stations, and the rejection production line is not shown in the figure.
The front visual defect detection device comprises a front line scanning camera, a light source and a bracket for fixing the front line scanning camera and the light source on the production line; the reverse side visual defect detection device comprises a reverse side scanning camera, a light source and a bracket for fixing the reverse side scanning camera and the light source on the production line; the front side line scanning camera and the back side line scanning camera are in communication connection with the data processing center.
The wood board to be detected is placed in the swing disc feeding mechanism, passes through the lower portion of the visual lens through the conveying belt, is subjected to high-definition imaging firstly, is subjected to image preprocessing, and is sent to the neural network model for AI algorithm defect identification through detection and positioning. If the board is defective, the algorithm will record that the board is defective and reject it as it passes through the reject mechanism. If the wood board has no defect, the wood board enters the reverse side detection mechanism of the wood board through the turnover mechanism, and if the wood board also has no defect on the reverse side, the wood board enters the next production line through the material distribution production line. If the reverse side is defective, then the station is removed again and the next production line is not entered.
The plank AI visual defect detection device that this embodiment provided, through automatic system and visual system's integration, this system has saved the process of artifical categorised plank, has improved production efficiency and categorised accuracy, also can provide the guarantee for the quality is traceed back simultaneously.
Example 2
As shown in fig. 3, is a flow chart of the wood board AI visual defect detection method provided in this embodiment, and the method specifically includes:
s11: placing the wood board to be detected on a swinging plate feeding mechanism, and performing high-definition imaging when the wood board passes through a front visual defect detection device of a front detection assembly line;
s12: preprocessing the high-definition image, detecting and positioning, and sending the high-definition image into a board visual defect detection model for AI algorithm defect identification;
s13: if the wood board is defective, recording is carried out, the defective wood board is removed when passing through a removal assembly line, if the wood board is not defective, the wood board enters a reverse side detection assembly line through a turnover mechanism, and high-definition imaging is carried out when the wood board passes through a reverse side visual defect detection device of the reverse side detection assembly line;
s14: step S12 is executed, if the wood board is defective, recording is carried out, the defective wood board is removed when passing through the removing assembly line, and if the wood board is not defective, the step S15 is executed;
s15: when the front surface and the back surface of the wood board are not detected with defects, the wood board enters a production line.
The specific construction method of the wood board visual defect detection model comprises the following steps:
s21: collecting the data of the picture of the surface of the shot wood board by a front visual defect detection device or a back visual defect detection device; and (4) preparing equipment in advance to collect the picture data of the surface of the shot wood board. Collecting a large number of wood board pictures according to requirements, wherein the pictures comprise wood board defect pictures and normal wood board pictures.
S22: carrying out data annotation on the picture data on the surface of the wood board;
s23: and training the marked data to obtain a wood board visual defect detection model. The difficulty of the wood board visual identification algorithm is to accurately find the positions of the defects of the wood board. In an actual scene, under the influence of illumination and shielding, the algorithm model cannot be well positioned at the position of the wood board, and the conditions of false detection and missed detection also exist. Therefore, in order to accurately predict the number of the defects on each wood board, two different algorithms of detecting and dividing the defects of the wood board are respectively tested to position the defects of the wood board. The detection algorithm can detect the specific position of the board in the image, and the segmentation algorithm can calculate which pixels belong to the defects and which pixels belong to the background. The example segmentation algorithm can not only be used for positioning the wood board, but also be used for calculating the part of the pixel area belonging to the defect. And finally, in order to improve the accuracy of identifying the defects of the wood board, taking 10 pictures as a group of test data, and then taking the mode of the prediction results of the ten pictures as the inventory number of the group of pictures. The test result proves that the force segmentation algorithm can well complete the detection task of the defects of the wood board, and all indexes meet the requirements.
After the wood board visual defect detection model is specifically constructed, the method further comprises the following accuracy verification steps:
s24: deploying the model to a data processing center;
s25: carrying out defect detection on the plurality of wood boards by using the wood board visual defect detection model; and taking 2000 pictures as a test set, and performing defect prediction reasoning. The 2000 boards are classified and selected manually, and after the selection is finished, the 2000 boards are also processed by the vision equipment. And (4) one person is selected for detailed review, and the absolute value of the error between the machine detection accuracy and the manual detection accuracy is qualified within 1% in the same time.
S26: the detection accuracy of the wood board visual defect detection model is rechecked, the absolute value of the error is qualified within 1%, otherwise, the absolute value is unqualified;
s27: and judging whether the detection accuracy of the wood board visual defect detection model reaches 95%, if so, determining that the wood board visual defect detection model is qualified, otherwise, determining that the wood board visual defect detection model is unqualified.
And packaging and delivering the AI model which meets the requirements, and using the AI model on line.
The board AI visual defect detection method provided by the embodiment can be combined with the board AI visual defect detection device provided by the embodiment 1 to obtain a whole set of board automatic visual recognition machine equipment. The wood board to be detected is placed in the swing disc feeding mechanism, passes through the lower portion of the visual lens through the conveying belt, is subjected to high-definition imaging firstly, is subjected to image preprocessing, and is sent to the neural network model for AI algorithm defect identification through detection and positioning. If the board is defective, the algorithm will record that the board is defective and reject it as it passes through the reject mechanism. If the wood board has no defect, the wood board enters the reverse side detection mechanism of the wood board through the turnover mechanism, and if the wood board also has no defect on the reverse side, the wood board enters the next production line through the material distribution production line. If the reverse side is defective, then the station is removed again and the next production line is not entered.
The whole set of equipment has strong expansibility and self-optimization capability in the operation process, the identification accuracy of the algorithm can be continuously improved in the continuous operation process, and even if accidental identification failure or identification errors occur in a certain special case of a certain defect type wood board, the wood board characteristics can be accurately identified in the next special condition after being added into a training sample. And because enough sample training is carried out before the algorithm is obtained, the occurrence probability of the situation is low, and the system can stably run in the initial stage.
Example 3
The preferred embodiment provides a computer device, which can implement the steps in any embodiment of the wood board AI visual defect detection method provided in the embodiment of the present application, and therefore, the beneficial effects of the wood board AI visual defect detection method provided in the embodiment of the present application can be achieved, which are detailed in the foregoing embodiment and will not be described herein again.
Example 4
It will be understood by those skilled in the art that all or part of the steps of the methods of the above embodiments may be performed by instructions or by associated hardware controlled by the instructions, which may be stored in a computer readable storage medium and loaded and executed by a processor. To this end, the present invention provides a storage medium, in which a plurality of instructions are stored, and the instructions can be loaded by a processor to execute the steps of any one of the wood board AI visual defect detection methods provided by the embodiments of the present invention.
Wherein the storage medium may include: read Only Memory (ROM), Random Access Memory (RAM), magnetic or optical disks, and the like.
Since the instructions stored in the storage medium can execute the steps in any wood board AI visual defect detection method provided by the embodiment of the present invention, the beneficial effects that can be achieved by any wood board AI visual defect detection method provided by the embodiment of the present invention can be achieved, which are detailed in the foregoing embodiments and will not be described herein again.
The foregoing basic embodiments of the invention and their various further alternatives can be freely combined to form multiple embodiments, all of which are contemplated and claimed herein. In the scheme of the invention, each selection example can be combined with any other basic example and selection example at will. Numerous combinations will be known to those skilled in the art.
The above description is only for the purpose of illustrating the preferred embodiments of the present invention and is not to be construed as limiting the invention, and any modifications, equivalents and improvements made within the spirit and principle of the present invention are intended to be included within the scope of the present invention.

Claims (10)

1. Plank AI visual defect detection device, its characterized in that includes: the front side detection assembly line and the back side detection assembly line are connected through the turnover mechanism, the inlet of the front side detection assembly line is provided with the swing disc feeding mechanism, the front side detection assembly line is provided with the front side visual defect detection device, and the back side detection assembly line is provided with the back side visual defect detection device.
2. The wood board AI visual defect detection apparatus of claim 1 wherein the reverse side detection assembly line outlet is further connected to a rejection assembly line, the rejection assembly line having one or more rejection stations.
3. The wood board AI visual defect detecting device according to claim 1, wherein the front visual defect detecting device includes a front line scanning camera, a light source and a bracket for fixing the front line scanning camera and the light source on a production line; the reverse side visual defect detection device comprises a reverse side scanning camera, a light source and a bracket for fixing the reverse side scanning camera and the light source on the production line; the front side line scanning camera and the back side line scanning camera are in communication connection with the data processing center.
4. The wood board AI visual defect detection method is characterized by comprising the following steps:
s11: placing the wood board to be detected on a swinging plate feeding mechanism, and performing high-definition imaging when the wood board passes through a front visual defect detection device of a front detection assembly line;
s12: preprocessing the high-definition image, detecting and positioning, and sending the high-definition image into a board visual defect detection model for AI algorithm defect identification;
s13: if the wood board is defective, recording is carried out, the defective wood board is removed when passing through a removal assembly line, if the wood board is not defective, the wood board enters a reverse side detection assembly line through a turnover mechanism, and high-definition imaging is carried out when the wood board passes through a reverse side visual defect detection device of the reverse side detection assembly line;
s14: step S12 is executed, if the wood board is defective, recording is carried out, the defective wood board is removed when passing through the removing assembly line, and if the wood board is not defective, the step S15 is executed;
s15: when the front surface and the back surface of the wood board are not detected with defects, the wood board enters a production line.
5. The wood board AI visual defect detection method of claim 4, wherein the specific construction method of the wood board visual defect detection model comprises:
s21: collecting the data of the picture of the surface of the shot wood board by a front visual defect detection device or a back visual defect detection device;
s22: carrying out data annotation on the picture data on the surface of the wood board;
s23: and training the marked data to obtain a wood board visual defect detection model.
6. The wood board AI visual defect detection method of claim 4, wherein said wood board surface picture data comprises a wood board defect picture and a normal wood board picture.
7. The wood board AI visual defect detection method according to claim 5, further comprising an accuracy verification step:
s24: deploying the model to a data processing center;
s25: carrying out defect detection on the plurality of wood boards by using the wood board visual defect detection model;
s26: the detection accuracy of the wood board visual defect detection model is rechecked, the absolute value of the error is qualified within 1%, otherwise, the absolute value is unqualified;
s27: and judging whether the detection accuracy of the wood board visual defect detection model reaches 95%, if so, determining that the wood board visual defect detection model is qualified, otherwise, determining that the wood board visual defect detection model is unqualified.
8. The wood board AI visual defect detection method of claim 4, wherein said wood board AI visual defect detection model performing AI algorithm defect recognition specifically comprises:
s121: inputting the preprocessed wood floor picture into a pre-trained ResNet50 neural network to obtain a corresponding feature map;
s122: setting an ROI corresponding to the wood floor picture for each point in the obtained feature map, so that each wood floor picture can obtain a plurality of candidate ROIs;
s123: sending the candidate ROI into an RPN network for binary classification and BB regression, and filtering out partial candidate ROI;
s124: ROIAlign operation is carried out on the rest ROIs;
s125: and classifying the rest ROI, BB regression and MASK generation to obtain a final defect identification result.
9. A computer device, characterized in that the computer device comprises a processor and a memory, in which a computer program is stored, which computer program is loaded and executed by the processor to implement the wood board AI visual defect detection method according to any one of claims 1 to 6.
10. A computer-readable storage medium, in which a computer program is stored, which is loaded and executed by a processor to implement the wood board AI visual defect detection method according to any one of claims 1 to 6.
CN202110619136.0A 2021-06-03 2021-06-03 Wood board AI visual defect detection device, method, equipment and medium Pending CN113267506A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202110619136.0A CN113267506A (en) 2021-06-03 2021-06-03 Wood board AI visual defect detection device, method, equipment and medium

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202110619136.0A CN113267506A (en) 2021-06-03 2021-06-03 Wood board AI visual defect detection device, method, equipment and medium

Publications (1)

Publication Number Publication Date
CN113267506A true CN113267506A (en) 2021-08-17

Family

ID=77234107

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202110619136.0A Pending CN113267506A (en) 2021-06-03 2021-06-03 Wood board AI visual defect detection device, method, equipment and medium

Country Status (1)

Country Link
CN (1) CN113267506A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114577816A (en) * 2022-01-18 2022-06-03 广州超音速自动化科技股份有限公司 Hydrogen fuel bipolar plate detection method

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114577816A (en) * 2022-01-18 2022-06-03 广州超音速自动化科技股份有限公司 Hydrogen fuel bipolar plate detection method

Similar Documents

Publication Publication Date Title
US10964004B2 (en) Automated optical inspection method using deep learning and apparatus, computer program for performing the method, computer-readable storage medium storing the computer program, and deep learning system thereof
CN111325713B (en) Neural network-based wood defect detection method, system and storage medium
CN110659660B (en) Automatic optical detection classification equipment using deep learning system and training equipment thereof
CN109829914B (en) Method and device for detecting product defects
CN111189837B (en) Cigarette appearance online detection method and device
CN112053318A (en) Two-dimensional PCB defect real-time automatic detection and classification device based on deep learning
CN109840900A (en) A kind of line detection system for failure and detection method applied to intelligence manufacture workshop
CN110956191B (en) Classifier, classification method and classification system based on probability fusion
JP2021515885A (en) Methods, devices, systems and programs for setting lighting conditions and storage media
CN110415214A (en) Appearance detecting method, device, electronic equipment and the storage medium of camera module
CN111062961A (en) Contact lens edge defect detection method based on deep learning
CN113240673B (en) Defect detection method, defect detection device, electronic equipment and storage medium
CN108889635B (en) Online visual inspection method for manufacturing defects of ring-pull cans
CN116978834B (en) Intelligent monitoring and early warning system for wafer production
CN116385430A (en) Machine vision flaw detection method, device, medium and equipment
CN117392042A (en) Defect detection method, defect detection apparatus, and storage medium
US20240046617A1 (en) Machine Learning-Based Generation of Rule-Based Classification Recipes for Inspection System
CN114612418A (en) Method, device and system for detecting surface defects of mouse shell and electronic equipment
CN114549414A (en) Abnormal change detection method and system for track data
CN113267506A (en) Wood board AI visual defect detection device, method, equipment and medium
CN115719326A (en) PCB defect detection method and device
CN117788400A (en) Binocular vision-based photovoltaic module defect detection method and system
CN117372770A (en) PCB micro defect detection and identification method based on photometric stereo and deep learning
CN217332186U (en) Plank AI visual defect detection device
CN117011216A (en) Defect detection method and device, electronic equipment and storage medium

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