Intelligent labeling machine based on multispectral machine vision and detection method thereof

Abstract

The invention discloses a multispectral machine vision intelligent labeling machine and a detection method thereof, and relates to the technical field of labeling machines. The placing assembly is provided with a visual calibration reference plate, and materials are conveyed to the visual calibration reference plate. The positioning detection assembly comprises a machine vision analysis control system and a multispectral camera, the machine vision system is used for sorting qualified and unqualified materials, and the multispectral camera photographs the materials and transmits the photographs to the machine vision analysis control system; the method comprises the steps that S10, materials are placed on the visual calibration reference plate; s20, taking a picture of the material by the multispectral camera, transmitting the picture to a machine vision analysis control system, and focusing the camera; s30, the materials are photographed again, the photographs are sent back to the machine vision analysis control system, and the machine vision analysis control system conducts material detection; and S40, the materials are moved out through the material moving assembly, and the qualified materials are labeled through the labeling assembly. The illumination influence of the external environment on materials is reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of labeling machines, and in particular to an intelligent labeling machine with multi-spectral machine vision. Background Art

[0002] When the current labeling machine takes pictures to detect and label materials, it is affected by uneven external light, resulting in brightness differences in the image and low image clarity, which in turn causes errors in the labeling machine's detection and sorting of materials, causing the labeling machine to label unqualified materials, thereby reducing the labeling machine's detection and sorting accuracy and wasting labels. Summary of the invention

[0003] The purpose of the present invention is to provide a multi-spectral machine vision intelligent labeling machine and a detection method thereof, which can reduce the impact of the external environment on materials, enhance the detection and sorting accuracy of the labeling machine, and automatically label qualified materials.

[0004] To achieve the above object, the solution of the present invention is:

[0005] A multi-spectral machine vision intelligent labeling machine, which is used to be installed on a conveyor belt, has a feed port and a discharge port, and includes a placement component, a positioning detection component, a labeling component, and a material transfer component;

[0006] The placement component includes a placement platform, on which a visual calibration reference plate is arranged, and a conveyor belt conveys the material to the visual calibration reference plate through a feed port;

[0007] The positioning detection component is located above the placement platform. The positioning detection component includes a machine vision analysis control system and a multispectral camera. The machine vision system is used to sort qualified and unqualified materials, and the machine vision analysis control system and the multispectral camera are connected in communication. The lens of the multispectral camera is downward. The multispectral camera is used to take pictures of the materials on the visual calibration reference plate and upload them to the machine vision analysis control system to detect the position of the materials while enabling the machine vision analysis control system to sort qualified and unqualified materials.

[0008] The labeling component is located above the placement platform. The labeling component has a label and is connected to the machine vision analysis control system. The labeling component can receive the signal of the machine vision analysis control system and can attach the label to the qualified material.

[0009] The material moving assembly is used to move the labeled qualified materials and unqualified materials out of the discharge port. The material moving assembly includes a three-axis linear displacement mechanism and a material moving part. The three-axis linear displacement mechanism is located above the placement assembly. The three-axis linear displacement mechanism includes a moving end, and the moving end can move in three degrees of freedom of the X, Y, and Z axes. The multispectral camera, the labeling assembly, and the material moving part are arranged side by side on the moving end to follow the moving end to move in three degrees of freedom.

[0010] Furthermore, a light source assembly for irradiating downwards is arranged around the lens of the multi-spectral camera;

[0011] The labeling component is located on one side of the multispectral camera, and includes a labeling suction cup and a lifting cylinder, and the lifting cylinder is connected to and drives the labeling suction cup to move up or down;

[0012] The material moving part is arranged on the other side of the multi-spectral camera, and the material moving part comprises a mounting seat and at least one downward material suction cup, and the material suction cup is arranged at the bottom of the mounting seat.

[0013] Furthermore, the placement assembly also includes a longitudinally extending lifting frame, on which a lifting drive mechanism is provided, the placement platform is provided on the lifting frame, and the lifting drive mechanism is connected to and drives the placement platform to rise or descend along the lifting frame.

[0014] Furthermore, a conveying component is also provided on the placement platform, and the top side of the conveying component protrudes from the top surface of the placement platform. The conveying component extends from one end of the placement platform close to the conveyor belt to the other end of the placement platform away from the conveyor belt. The conveyor belt is used to convey the material to the conveying component, and the conveying component is used to convey the material in a direction away from the conveyor belt.

[0015] Furthermore, the visual calibration reference plate is in the shape of a black and white chessboard and is disposed at one end of the conveying component away from the conveyor belt. The conveying component can convey the material thereon to the visual calibration reference plate.

[0016] Furthermore, the top side of the lifting frame extends to the bottom side of the three-axis linear displacement mechanism, and when the conveying component conveys the material to the extreme position, the lifting drive mechanism drives the placement platform to rise to the top side of the lifting frame.

[0017] Furthermore, it also includes a label printer, which is used to print labels. The labeling component can pick up the labels printed by the label printer and stick them on the material.

[0018] A detection method for a multi-spectral machine vision intelligent labeling machine, using the multi-spectral machine vision intelligent labeling machine, the intelligent labeling machine also involves a motion control system, including the following steps:

[0019] Step S10: The conveyor belt places the material to be inspected on the visual calibration reference plate, with the surface of the material to be inspected facing upward and located below the lens of the multispectral camera;

[0020] Step S20: starting the machine vision analysis control system connected to the multispectral camera, triggering the multispectral camera at the machine vision analysis control system, so that the multispectral camera takes a picture of the material for the first time, transmitting the taken picture to the machine vision analysis control system, and the machine vision analysis control system controls the multispectral camera to perform a focusing process, and the multispectral camera moves with three degrees of freedom so that the material is located in the center of the field of view of the multispectral camera;

[0021] Step S30: After the multispectral camera is focused, the material is photographed for the second time, and the second photograph is transmitted back to the machine vision analysis control system, and the machine vision analysis control system performs material inspection;

[0022] Step S40: The machine vision analysis control system transmits the material inspection result to the motion control system, and the motion control system controls the material moving component to remove the material with unqualified inspection result from the discharge port and place it in the unqualified material placement area, and the motion control system controls the labeling component to label the material with qualified inspection result.

[0023] Furthermore, in step S20, the following steps are also included:

[0024] Step S21, when the material is within the field of view of the multispectral camera, the distance from the rear edge of the material to the rear boundary of the field of view of the multispectral camera is set to B1, the distance from the front edge of the material to the front boundary of the field of view of the multispectral camera is set to B2, the distance from the left edge of the material to the left boundary of the field of view of the multispectral camera is set to A1, and the distance from the right edge of the material to the right boundary of the field of view of the multispectral camera is set to A2;

[0025] Step S22, when the material is in the center of the multi-spectral camera field of view, the distance from the rear edge of the material to the rear boundary of the multi-spectral camera field of view is set to B1, which is the minimum setting value L1min, and the distance from the left edge of the material to the left boundary of the multi-spectral camera field of view is set to A1, which is the minimum setting value L2min;

[0026] Step S23, measure and calculate the difference between B1 and the minimum setting value L1min. If B1-L1min>0, the multispectral camera moves forward, B1-L1min<0, the multispectral camera moves backward, if B1-L1min=0, the multispectral camera does not move forward or backward. The judgment method for the left and right movement of the multispectral camera is the same as the judgment method for the front and back movement. If A1-L2min>0, the multispectral camera moves to the right, if A1-L2min<0, the multispectral camera moves to the left, if A1-L2min=0, the multispectral camera does not move left or right.

[0027] After adopting the above scheme, the beneficial effects of the present invention are:

[0028] (1) The present invention adopts multi-spectral machine vision technology, so that the multi-spectral camera and the visual calibration reference plate are used in combination, and the multi-spectral camera is used to take pictures of the materials on the visual calibration reference plate, thereby reducing the light impact of the external environment on the materials, improving the clarity of the image, and balancing the brightness difference in the image. The machine vision analysis control system is used to separate qualified and unqualified materials, and the signal is transmitted to the labeling component so that the labeling component labels the qualified materials. The multi-spectral machine vision technology is combined with the automation equipment to realize the automation of the detection process, and improve the accuracy of the detection, improve the work efficiency and accuracy;

[0029] (2) The multispectral camera of the present invention can automatically move and focus in the three degrees of freedom of the X, Y, and Z axes to obtain the best photographic image, perform more accurate image analysis, and improve the adaptability and switching efficiency of detecting different materials without the need for manual focus adjustment. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] Figure 1 It is a schematic diagram of the overall structure of the present invention;

[0031] Figure 2 yes Figure 1 A magnified view of part A;

[0032] Figure 3 It is a structural diagram of the present invention in cooperation with a conveyor belt;

[0033] Figure 4 It is a schematic diagram of the internal structure of the present invention;

[0034] Figure 5 It is a schematic diagram of the structure of the placement component of the present invention;

[0035] Figure 6 It is a relationship diagram between the field of view of the multi-spectral camera of the present invention and the material position.

[0036] Explanation of reference numbers: 10. Placement component; 11. Placement platform; 111. Conveying component; 112. Visual calibration reference plate; 12. Lifting frame; 20. Positioning detection component; 21. Multispectral camera; 211. Light source component; 30. Labeling component; 31. Labeling suction cup; 32. Lifting cylinder; 40. Material transfer component; 41. Three-axis linear displacement mechanism; 411. Moving end; 42. Material transfer part; 421. Mounting seat; 422. Material suction cup; 50. Feed inlet; 60. Discharge outlet; 70. Labeling printer; 80. Material; 90. Conveyor belt. DETAILED DESCRIPTION

[0037] The present invention is described in detail below with reference to the accompanying drawings and specific embodiments.

[0038] like Figures 1 to 6 As shown, the present invention provides a multi-spectral machine vision intelligent labeling machine, which is used to be installed on a conveyor belt 90. The labeling machine has a feed port 50 and a discharge port 60, and includes a placement component 10, a positioning detection component 20, a labeling component 30, and a material moving component 40; the placement component 10 includes a placement platform 11, and a visual calibration reference plate 112 is arranged on the placement platform 11. The conveyor belt 90 transfers the material 80 to the visual calibration reference plate 112 through the feed port 50; the positioning detection component 20 is located above the placement platform 11, and includes a machine vision analysis control system The machine vision system is used to sort qualified and unqualified materials 80, and the machine vision analysis control system is connected to the multispectral camera 21, the lens of the multispectral camera 21 is downward, and the multispectral camera 21 is used to take pictures of the material 80 on the visual calibration reference plate 112, and upload them to the machine vision analysis control system to detect the position of the material 80 while enabling the machine vision analysis control system to sort qualified and unqualified materials 80; the labeling component 30 is located above the placement platform 11, and the labeling component 30 has a label and communicates with the machine vision analysis control system. The control system is connected in communication, the labeling component 30 can receive the signal of the machine vision analysis control system, and can label the qualified material 80; the material moving component 40 is used to move the labeled qualified material 80 and the unqualified material 80 out of the discharge port 60, and the material moving component 40 includes a three-axis linear displacement mechanism 41 and a material moving part 42. The three-axis linear displacement mechanism 41 includes a moving end 411, and the moving end 411 can move in three degrees of freedom of X, Y, and Z axes. The multispectral camera 21, the labeling component 30, and the material moving part 42 are arranged side by side on the moving end. The multi-spectral camera 21, the labeling component 30 and the material moving part 42 are arranged side by side on the moving end 411 of the three-axis linear displacement mechanism 41, so that a three-axis linear displacement mechanism 41 can simultaneously drive the multi-spectral camera 21, the labeling component 30 and the material moving part 42 to move, so that the internal structure of the labeling machine of the present invention is simple, uncomplicated and compact. Compared with the prior art, the number of parts inside the labeling machine is reduced, and the coordination and concession between the various components inside the labeling machine are also reduced.

[0039] Specifically, the multispectral camera 21 can provide the machine vision system with richer data than ordinary RGB images by capturing spectral information in different bands, which enables the system to more accurately identify and analyze target objects and maintain high stability and accuracy even when external light conditions change. Even in complex lighting environments, there may be problems such as large brightness differences, shadows and reflections in the image. The multispectral camera 21 can use spectral information in different bands to reduce these effects. For example, by adjusting the weights of different bands or performing image fusion processing, the brightness differences in the image can be balanced to improve the clarity and consistency of the image.

[0040] Specifically, a downward-irradiating light source assembly 211 is arranged around the lens of the multi-spectral camera 21, and the light source assembly 211 is used to provide light to the material 80; the labeling assembly 30 is located on one side of the multi-spectral camera 21, and includes a labeling suction cup 31 and a lifting cylinder 32, and the lifting cylinder 32 is connected to and drives the labeling suction cup 31 to perform an upward or downward movement; the material moving part 42 is arranged on the other side of the multi-spectral camera 21, and the material moving part 42 includes a mounting seat 421 and at least one downward material suction cup 422, and the material suction cup 422 is arranged at the bottom of the mounting seat 421.

[0041] Furthermore, the placement assembly 10 further includes a longitudinally extending lifting frame 12, on which a lifting drive mechanism is disposed, and the placement platform 11 is disposed on the lifting frame 12, and a lifting drive mechanism (not shown in the figure) is connected to and drives the placement platform 11 to rise or fall along the lifting frame 12; the specific lifting drive mechanism may be a structure in the prior art that can drive the placement platform 11 to rise and fall, and the placement platform 11 rises and falls, so that the placement platform 11 can adjust its height according to the height of the conveyor belt 90;

[0042] Furthermore, a conveying component 111 is also provided on the placement platform 11, the top side of the conveying component 111 protrudes from the top surface of the placement platform 11, and the conveying component 111 extends from one end of the placement platform 11 close to the conveyor belt 90 to the other end of the placement platform 11 away from the conveyor belt 90, and the conveyor belt 90 is used to convey the material 80 to the conveying component 111, and the conveying component 111 is used to convey the material 80 in a direction away from the conveyor belt 90; specifically, the conveying component 111 is two parallel conveying chains, the conveying chains are embedded in the placement platform 11, the top side of the conveying chains protrudes from the top surface of the placement platform 11, the bottom side of the conveying chains is located below the bottom surface of the placement platform 11, and the conveying chains are driven by a driving device to transmit in a direction away from the conveyor belt 90, and the conveyor belt 90 described here is the conveyor belt 90 located at the feeding port 50 of the labeling machine;

[0043] Specifically, the visual calibration reference plate 112 is in the shape of a black and white chessboard, and is disposed at one end of the conveying component 111 away from the conveyor belt, and the conveying component 111 can convey the material 80 thereon to the visual calibration reference plate 112;

[0044] Specifically, when inspecting materials, step (1) obtains spectral comparison benchmark information: the multi-spectral camera 21 takes a picture, analyzes and calculates the spectral characteristics of each black and white square, records the energy values ​​of the different characteristic bands, and uses this spectral characteristic as the comparison benchmark for the detection and identification of the actual material 80; step (2) corrects the spectral comparison benchmark information: the qualified material 80 is placed on the visual calibration benchmark plate, the multi-spectral camera takes a picture 21, analyzes and calculates the spectral characteristics of each black and white square, records the energy values ​​of the different characteristic bands, and uses this spectral characteristic to correct the benchmark information obtained in step (1). In addition, for reflective objects, the high reflectivity of their surfaces may cause the information in the image to be submerged, making visual detection difficult. The different effects of light in different bands on reflectivity can be used to select the band for material detection, suppress the interference of reflective objects, and improve detection accuracy. Step (3) actual operation: the multi-spectral camera 21 takes a photo of the material 80 to be detected and labeled, obtains the actual spectral characteristics, extracts and analyzes the band information determined by step (2), and compares and corrects it with the reference information obtained in step (2), balances the spectral value and brightness differences of the image caused by differences in lighting conditions, and improves the clarity and consistency of the image.

[0045] Specifically, the top side of the lifting frame 12 extends to the bottom side of the three-axis linear displacement mechanism 41. When the conveying component 111 conveys the material 80 to the extreme position, the lifting drive mechanism drives the placement platform 11 to rise to the top side of the lifting frame 12. That is, when the conveying component 111 conveys the material 80 to the extreme position, the material 80 will rise to the bottom side of the three-axis linear displacement mechanism 41 along with the placement platform 11 to reduce the descending stroke of the moving end 411 of the three-axis linear displacement mechanism 41.

[0046] Furthermore, it also includes a label printer 70, which is used to print labels. The label component 30 can pick up the labels printed by the label printer 70 and stick them on the material 80.

[0047] Specifically, the present invention also includes a central control device, which is communicatively connected to the placement component 10, the positioning detection component 20, the labeling component 30, the material transfer component 40 and the labeling printer 70 for controlling the above-mentioned devices.

[0048] The present invention also provides a detection method for a multi-spectral machine vision intelligent labeling machine, wherein the multi-spectral machine is a visual intelligent labeling machine, and the intelligent labeling machine also involves a motion control system, comprising the following steps:

[0049] Step S10: The conveyor belt 90 places the material 80 to be inspected on the visual calibration reference plate 112, with the surface of the material 80 to be inspected facing upward and located below the lens of the multi-spectral camera 21;

[0050] Step S20: starting the machine vision analysis control system connected to the multispectral camera 21, triggering the multispectral camera 21 at the machine vision analysis control system, so that the multispectral camera 21 takes a picture of the material 80 for the first time, and transmitting the taken picture to the machine vision analysis control system, the machine vision analysis control system controls the multispectral camera 21 to perform a focusing process, and the multispectral camera 21 moves with three degrees of freedom so that the material 80 is located in the center of the field of view of the multispectral camera 21;

[0051] Step S30: After the multi-spectral camera 21 has completed focusing, the material 80 is photographed for the second time, and the second photograph is transmitted back to the machine vision analysis control system, and the machine vision analysis control system performs inspection on the material 80;

[0052] Step S40: The machine vision analysis control system transmits the material inspection result to the motion control system, and the motion control system controls the material moving component 40 to move the material 80 with an unqualified inspection result out of the discharge port 60 and place it in the unqualified material 80 placement area, and the motion control system controls the labeling component 30 to label the material 80 with a qualified inspection result.

[0053] The method of the present invention can maintain high stability and accuracy even when the external light conditions change. Even in a complex lighting environment, there may be problems such as large brightness differences, shadows and reflections in the image. The multispectral camera 21 can also use the spectral information of different bands to reduce these effects. For example, by adjusting the weights of different bands or performing image fusion processing, the brightness differences in the image can be balanced and the clarity and consistency of the image can be improved.

[0054] Furthermore, in step S20, the following steps are also included:

[0055] Step S21, when the material 80 is within the field of view of the multispectral camera 21, the distance from the rear edge of the material 80 to the rear boundary of the field of view of the multispectral camera 21 is set to B1, the distance from the front edge of the material 80 to the front boundary of the field of view of the multispectral camera 21 is set to B2, the distance from the left edge of the material 80 to the left boundary of the field of view of the multispectral camera 21 is set to A1, and the distance from the right edge of the material 80 to the right boundary of the field of view of the multispectral camera 21 is set to A2;

[0056] Step S22, when the material 80 is set in the center of the field of view of the multi-spectral camera 21, the distance from the rear edge of the material 80 to the rear boundary of the field of view of the multi-spectral camera 21 is B1, which is the minimum setting value L1min, and the distance from the left edge of the material 80 to the left boundary of the field of view of the multi-spectral camera 21 is A1, which is the minimum setting value L2min;

[0057] Step S23, measure and calculate the difference between B1 and the minimum setting value L1min. If B1-L1min>0, the multispectral camera 21 moves forward, B1-L1min<0, the multispectral camera 21 moves backward, if B1-L1min=0, the multispectral camera 21 does not move forward or backward. The judgment method for the left and right movement of the multispectral camera 21 is the same as the judgment method for the front and back movement. If A1-L2min>0, the multispectral camera 21 moves to the right, if A1-L2min<0, the multispectral camera 21 moves to the left, if A1-L2min=0, the multispectral camera 21 does not move left or right. Through this focus judgment method, the material 80 is located in the center of the field of view of the multispectral camera 21.

[0058] Specifically, the multispectral camera 21 can also be adaptively moved up and down for adjustment.

[0059] Specifically, the multispectral camera 21 can automatically move and focus in three degrees of freedom of the X, Y, and Z axes to obtain the best photographic image, perform more accurate image analysis, and improve the adaptability and switching efficiency of detecting different materials without manual focus adjustment.

[0060] To further illustrate the various embodiments, the present invention provides drawings. These drawings are part of the disclosure of the present invention, which are mainly used to illustrate the embodiments and can be used in conjunction with the relevant descriptions in the specification to explain the operating principles of the embodiments. With reference to these contents, a person of ordinary skill in the art should be able to understand other possible implementations and advantages of the present invention. The components in the figures are not drawn to scale, and similar component symbols are generally used to represent similar components.

[0061] Meanwhile, the directions such as front, back, left, and right involved in this embodiment are only used as a reference for directions and do not represent directions in actual use. In addition, the terms "first", "second", and "third" are only used for descriptive purposes and cannot be understood as indicating or implying relative importance.

[0062] The above description is only a preferred embodiment of the present invention and is not a limitation on the design of this case. Any equivalent changes made based on the design key of this case shall fall within the protection scope of this case.

Claims

1. A multi-spectral machine vision intelligent labeling machine, used to be installed on a conveyor belt, the labeling machine has a feed inlet and a discharge outlet, characterized in that: Including placement components, positioning detection components, labeling components and material transfer components; The placement component includes a placement platform, on which a visual calibration reference plate is arranged, and a conveyor belt conveys the material to the visual calibration reference plate through a feed port; The positioning detection component is located above the placement platform. The positioning detection component includes a machine vision analysis control system and a multispectral camera. The machine vision system is used to sort qualified and unqualified materials, and the machine vision analysis control system and the multispectral camera are connected in communication. The lens of the multispectral camera is downward. The multispectral camera is used to take pictures of the materials on the visual calibration reference plate and upload them to the machine vision analysis control system to detect the position of the materials while enabling the machine vision analysis control system to sort qualified and unqualified materials. The labeling component is located above the placement platform. The labeling component has a label and is connected to the machine vision analysis control system. The labeling component can receive the signal of the machine vision analysis control system and can attach the label to the qualified material. The material moving assembly is used to move the labeled qualified materials and unqualified materials out of the discharge port. The material moving assembly includes a three-axis linear displacement mechanism and a material moving part. The three-axis linear displacement mechanism is located above the placement assembly. The three-axis linear displacement mechanism includes a moving end, and the moving end can move in three degrees of freedom of the X, Y, and Z axes. The multispectral camera, the labeling assembly, and the material moving part are arranged side by side on the moving end to follow the moving end to move in three degrees of freedom.

2. The multi-spectral machine vision intelligent labeling machine according to claim 1, characterized in that: A light source assembly for irradiating downwards is arranged around the lens of the multispectral camera; The labeling component is located on one side of the multispectral camera, and includes a labeling suction cup and a lifting cylinder, and the lifting cylinder is connected to and drives the labeling suction cup to move up or down; The material moving part is arranged on the other side of the multi-spectral camera, and the material moving part comprises a mounting seat and at least one downward material suction cup, and the material suction cup is arranged at the bottom of the mounting seat.

3. The multi-spectral machine vision intelligent labeling machine according to claim 2, characterized in that: The placement assembly also includes a longitudinally extending lifting frame, on which a lifting drive mechanism is arranged, and the placement platform is arranged on the lifting frame, and the lifting drive mechanism is connected to and drives the placement platform to rise or fall along the lifting frame.

4. The multi-spectral machine vision intelligent labeling machine according to claim 3, characterized in that: The placement platform is also provided with a conveying component, the top side of which protrudes from the top surface of the placement platform. The conveying component extends from one end of the placement platform close to the conveyor belt to the other end of the placement platform away from the conveyor belt. The conveyor belt is used to convey materials to the conveying component, and the conveying component is used to convey materials in a direction away from the conveyor belt.

5. The multi-spectral machine vision intelligent labeling machine according to claim 4, characterized in that: The visual calibration reference plate is in the shape of a black and white chessboard and is arranged at one end of the conveying component away from the conveyor belt. The conveying component can convey the material thereon to the visual calibration reference plate.

6. The multi-spectral machine vision intelligent labeling machine according to claim 4, characterized in that: The top side of the lifting frame extends to the bottom side of the three-axis linear displacement mechanism. When the conveying component conveys the material to the extreme position, the lifting drive mechanism drives the placement platform to rise to the top side of the lifting frame.

7. The multi-spectral machine vision intelligent labeling machine according to claim 1, characterized in that: It also includes a label printer, which is used to print labels. The labeling component can pick up the labels printed by the label printer and stick them on the material.

8. A detection method for a multi-spectral machine vision intelligent labeling machine, using the multi-spectral machine vision intelligent labeling machine of claim 1, the intelligent labeling machine also involves a motion control system, characterized in that: The following steps are involved: Step S10: The conveyor belt places the material to be inspected on the visual calibration reference plate, with the surface of the material to be inspected facing upward and located below the lens of the multispectral camera; Step S20: starting the machine vision analysis control system connected to the multispectral camera, triggering the multispectral camera at the machine vision analysis control system, so that the multispectral camera takes a picture of the material for the first time, transmitting the taken picture to the machine vision analysis control system, and the machine vision analysis control system controls the multispectral camera to perform a focusing process, and the multispectral camera moves with three degrees of freedom so that the material is located in the center of the field of view of the multispectral camera; Step S30: After the multispectral camera is focused, the material is photographed for the second time, and the second photograph is transmitted back to the machine vision analysis control system, and the machine vision analysis control system performs material inspection; Step S40: The machine vision analysis control system transmits the material inspection result to the motion control system, and the motion control system controls the material moving component to remove the material with unqualified inspection result from the discharge port and place it in the unqualified material placement area, and the motion control system controls the labeling component to label the material with qualified inspection result.

9. The detection method of a multi-spectral machine vision intelligent labeling machine as claimed in claim 8, characterized in that: In step S20, the following steps are also included: Step S21, when the material is within the field of view of the multispectral camera, the distance from the rear edge of the material to the rear boundary of the field of view of the multispectral camera is set to B1, the distance from the front edge of the material to the front boundary of the field of view of the multispectral camera is set to B2, the distance from the left edge of the material to the left boundary of the field of view of the multispectral camera is set to A1, and the distance from the right edge of the material to the right boundary of the field of view of the multispectral camera is set to A2; Step S22, when the material is in the center of the multi-spectral camera field of view, the distance from the rear edge of the material to the rear boundary of the multi-spectral camera field of view is set to B1, which is the minimum setting value L1min, and the distance from the left edge of the material to the left boundary of the multi-spectral camera field of view is set to A1, which is the minimum setting value L2min; Step S23, measure and calculate the difference between B1 and the minimum setting value L1min. If B1-L1min>0, the multispectral camera moves forward, B1-L1min<0, the multispectral camera moves backward, if B1-L1min=0, the multispectral camera does not move forward or backward. The judgment method for the left and right movement of the multispectral camera is the same as the judgment method for the front and back movement. If A1-L2min>0, the multispectral camera moves to the right, if A1-L2min<0, the multispectral camera moves to the left, if A1-L2min=0, the multispectral camera does not move left or right.

Images