Spring roll quality detection system based on machine vision and artificial intelligence

Abstract

The invention discloses a spring roll quality detection system based on machine vision and artificial intelligence, the spring roll quality detection system comprises a mounting rack, an intermittent feeding mechanism, an image acquisition mechanism, a material pushing mechanism, a turnover mechanism and a control box, the intermittent feeding mechanism is used for feeding spring rolls on a feeding conveyor belt to the image acquisition mechanism, the turnover mechanism is arranged on the intermittent feeding mechanism, and the image acquisition mechanism is arranged on the turnover mechanism. The overturning mechanism is used for overturning the spring rolls on the intermittent feeding mechanism, the image collecting mechanism sends collected appearance pictures of the spring rolls to the AI processor, the AI processor analyzes the pictures sent by the image collecting mechanism and obtains a judgment result, and the AI processor sends the obtained judgment result to the PLC. According to the spring roll appearance detection system, the intermittent feeding mechanism, the turnover mechanism and the image acquisition mechanism are matched with an AI processor, automatic detection of the appearance of spring rolls is achieved, and compared with manual detection, the spring roll appearance detection system can remarkably improve the detection efficiency and reduce the labor cost.

Description

Technical Field

[0001] The invention relates to the technical field of machine vision detection, and in particular to a spring roll quality detection system based on machine vision and artificial intelligence. Background Art

[0002] Spring rolls, also known as spring pancakes, spring pancakes, and pancakes, are a traditional food for Chinese folk festivals and are popular throughout China.

[0003] The Chinese patent with announcement number CN109548815B discloses a fully automatic spring roll wrapping device and a fully automatic spring roll wrapping method thereof, wherein the spring roll skin is sent to a supporting platform by a spring roll skin conveying unit, and the jet unit is used to blow the spring roll skin placed on the supporting platform off from the skin dropping hole. Since the spring roll skin always falls from one skin dropping hole, it can be ensured that the spring roll skin can always fall accurately to the specified position, and there is no need to flip the spring roll skin through other mechanical structures, but the spring roll skin is blown off by the jet unit, thereby greatly improving efficiency, reducing failure rate, and better ensuring that the spring roll skin will not be damaged; the semi-finished spring roll is wrapped in the spring roll skin placement area, and there is no need to convey it to multiple positions for wrapping, and the semi-finished spring roll is rolled into a finished spring roll by a rolling device, which greatly improves the wrapping efficiency; by setting a paste dotting machine to dot paste on the spring roll skin, when the semi-finished spring roll is subsequently wrapped into a finished spring roll, the spring roll will not fall apart through the adhesion of the dough paste, and the spring roll will be better formed, which greatly ensures the quality of the spring roll after forming.

[0004] However, the above-mentioned equipment may be damaged or the filling may be exposed during the rolling of the spring rolls. In order to ensure that the packaged spring rolls have a good appearance, it is necessary to manually inspect the appearance of the rolled spring rolls. However, manual inspection not only consumes a lot of manpower, but also has low efficiency.

[0005] Therefore, a spring roll quality inspection system based on machine vision and artificial intelligence is proposed. Summary of the invention

[0006] The object of the present invention is to provide a spring roll quality detection system based on machine vision and artificial intelligence, so as to solve or at least alleviate one or more of the above-mentioned problems and other problems existing in the prior art.

[0007] To achieve the above object, the present invention provides the following technical solution: a spring roll quality detection system based on machine vision and artificial intelligence, comprising:

[0008] Mounting frame;

[0009] An intermittent feeding mechanism, the intermittent feeding mechanism is used to feed the spring rolls on the feeding conveyor belt to the image acquisition mechanism, and the intermittent feeding mechanism is arranged on the mounting frame;

[0010] The image acquisition mechanism is used to collect appearance data of spring rolls, and the image acquisition mechanism is arranged on the mounting frame;

[0011] A turning mechanism, the turning mechanism being arranged on the intermittent feeding mechanism, and the turning mechanism being used to turn over the spring rolls on the intermittent feeding mechanism, so that the image acquisition mechanism can perform an all-round detection on the surface of the spring rolls;

[0012] A control box, wherein an AI processor and a PLC controller are arranged in the control box, the image acquisition mechanism sends the collected appearance picture of the spring roll to the AI ​​processor, the AI ​​processor analyzes the picture sent by the image acquisition mechanism and obtains a determination result, and the AI ​​processor sends the obtained determination result to the PLC controller, and the control box is arranged on one side of the mounting frame;

[0013] A pushing mechanism, when the result obtained by the AI ​​processor is bad, the PLC controller controls the pushing mechanism to push the bad spring rolls to the bad product conveyor belt, and when the result obtained by the AI ​​processor is good, the PLC controller controls the pushing mechanism to push the good spring rolls to the good product conveyor belt, and the pushing mechanism is arranged on the mounting frame;

[0014] The turning mechanism and the intermittent feeding mechanism are both controlled by the PLC controller.

[0015] In a spring roll quality inspection system based on machine vision and artificial intelligence according to the present invention, optionally, the intermittent feeding mechanism includes a feeding tray and a support seat, the feeding tray is provided with a plurality of feeding grooves at equal angles, the support seat is fixedly mounted on the top of the mounting frame, a cross roller bearing is arranged between the top of the support seat and the bottom of the feeding tray, the top of the support seat is fixedly connected to the outer ring of the cross roller bearing, the bottom of the feeding tray is fixedly connected to the inner ring of the cross roller bearing, an inner gear annular disk is fixedly mounted on the bottom of the feeding tray, the inner gear annular disk is located on the inner side of the inner ring of the cross roller bearing, a first driving motor is fixedly mounted on the top of the mounting frame, a first gear is fixedly mounted on the rotating shaft of the first driving motor, the first gear is meshed with the inner gear annular disk, and the first driving motor is controlled by the PLC controller.

[0016] In a spring roll quality inspection system based on machine vision and artificial intelligence according to the present invention, optionally, the flipping mechanism includes a flipping assembly, a gear disk and a second motor;

[0017] The flipping assembly is arranged in the feeding trough, and the flipping assembly includes a first flipping roller and a second flipping roller. The first flipping roller and the second flipping roller are both rotatably connected to the inner lower end of the feeding trough. The toothed disc drives the first flipping roller and the second flipping roller to rotate synchronously through a transmission member. The toothed disc is arranged on the inner side of the feeding tray. The second motor is fixedly connected to the top of the mounting frame. The toothed disc is fixedly connected to the rotating shaft of the second motor. By controlling the rotation of the rotating shaft of the second motor, the toothed disc is driven to rotate. The rotating toothed disc drives the first flipping roller and the second flipping roller to rotate through the transmission member. The rotating first flipping roller and the second flipping roller drive the spring roll to flip, so that the image acquisition mechanism can take all-round photos of the spring roll.

[0018] In a spring roll quality inspection system based on machine vision and artificial intelligence according to the present invention, optionally, the transmission member includes a transmission rod, a second gear and a fixed seat, the fixed seat is fixedly connected to the inner top of the feeding plate, the transmission rod is rotatably mounted on the fixed seat, a first synchronous wheel and a second synchronous wheel are fixedly connected to a rotating shaft at one end of the first flipping roller, a third synchronous wheel is fixedly connected to a rotating shaft at one end of the second flipping roller, one end of the transmission rod is fixedly connected to a fourth synchronous wheel, the fourth synchronous wheel is transmission-connected to the second synchronous wheel via a first synchronous belt, the first synchronous wheel is transmission-connected to the third synchronous wheel via a second synchronous belt, the second gear is fixedly connected to the other end of the transmission rod, and the second gear is meshed with the toothed disc.

[0019] In a spring roll quality inspection system based on machine vision and artificial intelligence according to the present invention, optionally, guide inclined plates are fixedly connected to the inner walls on both sides of the opposite sides of the feeding trough, and the guide inclined plates are located above the first turning roller and the second turning roller. The guide inclined plates can place the spring rolls in the feeding trough between the first turning roller and the second turning roller.

[0020] In a spring roll quality inspection system based on machine vision and artificial intelligence according to the present invention, optionally, the surface of the first turning roller and the surface of the second turning roller are both provided with anti-slip grooves.

[0021] In a spring roll quality inspection system based on machine vision and artificial intelligence according to the present invention, optionally, the image acquisition mechanism includes a first support frame, the first support frame is fixedly mounted on the top of the mounting frame, a CCD camera is fixedly mounted on the bottom of the upper end of the first support frame, and a signal output end of the CCD camera is electrically connected to a signal input end of the AI ​​processor.

[0022] In a spring roll quality inspection system based on machine vision and artificial intelligence according to the present invention, optionally, a plurality of fill lights are fixedly mounted on the bottom of the upper end of the first support frame.

[0023] In a spring roll quality inspection system based on machine vision and artificial intelligence according to the present invention, optionally, the pushing mechanism includes a second support frame, and a qualified product pushing member and a defective product pushing member are provided at the bottom of the upper end of the second support frame;

[0024] The qualified product pushing member comprises a first mounting seat, a first pushing electric push rod is fixedly mounted on the first mounting seat, and a first push plate is fixedly connected to the end of the piston rod of the first pushing electric push rod;

[0025] The defective product pushing member includes a second mounting seat, on which a second pushing electric push rod is fixedly mounted, a piston rod end of the second pushing electric push rod is fixedly connected to a second push plate, and the first pushing electric push rod and the second pushing electric push rod are both controlled by the PLC controller.

[0026] In a spring roll quality inspection system based on machine vision and artificial intelligence according to the present invention, optionally, an annular groove is provided on the top of the feeding tray, the annular groove is connected to the feeding trough, and the lower end of the first push plate and the lower end of the second push plate are both slidably arranged inside the annular groove.

[0027] Compared with the prior art, the present invention has the following beneficial effects:

[0028] The present invention adopts an intermittent feeding mechanism, a turning mechanism, an image acquisition mechanism and an AI processor to realize automatic detection of the appearance of spring rolls. Compared with manual detection, the system can significantly improve detection efficiency and reduce labor costs.

[0029] Through the design of the flipping mechanism, the image acquisition mechanism can perform all-round inspection of the surface of the spring roll, ensuring that each spring roll can be comprehensively and carefully inspected, thereby effectively reducing the risk of missed inspection and false inspection. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] Figure 1 This is a structural schematic diagram of a spring roll quality detection system based on machine vision and artificial intelligence of the present invention;

[0031] Figure 2 It is a structural schematic diagram of an intermittent feeding mechanism in a spring roll quality detection system based on machine vision and artificial intelligence of the present invention;

[0032] Figure 3 It is a schematic diagram of a partial explosion structure of an intermittent feeding mechanism in a spring roll quality detection system based on machine vision and artificial intelligence of the present invention;

[0033] Figure 4 It is a schematic cross-sectional structure diagram of an intermittent feeding mechanism in a spring roll quality detection system based on machine vision and artificial intelligence of the present invention;

[0034] Figure 5 This is one of the partial structural schematic diagrams of a turning mechanism in a spring roll quality inspection system based on machine vision and artificial intelligence of the present invention;

[0035] Figure 6 This is a second partial structural schematic diagram of a turning mechanism in a spring roll quality inspection system based on machine vision and artificial intelligence of the present invention;

[0036] Figure 7 It is a structural schematic diagram of a turning mechanism turning over spring rolls in a spring roll quality inspection system based on machine vision and artificial intelligence of the present invention;

[0037] Figure 8 It is a structural schematic diagram of a sub-image acquisition mechanism in a spring roll quality detection system based on machine vision and artificial intelligence of the present invention;

[0038] Fig. 9 The present invention is a structural schematic diagram of a material pushing structure in a spring roll quality detection system based on machine vision and artificial intelligence.

[0039] In the figure: 1. Mounting frame;

[0040] 2. Intermittent feeding mechanism; 201. Feeding plate; 2011. Feeding trough; 2012. Annular groove; 2013. Guide inclined plate; 202. Support seat; 203. Internally toothed annular plate; 204. Cross roller bearing; 205. First driving motor; 206. First gear;

[0041] 3. Image acquisition mechanism; 301. First support frame; 302. CCD camera; 303. Fill light;

[0042] 4. Pushing mechanism; 401. Second support frame; 402. First mounting seat; 403. First pusher; 404. First push plate; 405. Second mounting seat; 406. Second pusher; 407. Second push plate

[0043] 5. Spring rolls;

[0044] 6. Feeding conveyor belt;

[0045] 7. Conveyor belt for qualified products;

[0046] 8. Defective product conveyor belt;

[0047] 9. Turning mechanism; 901. first turning roller; 9011. first synchronous wheel; 9012. second synchronous wheel; 9013. first synchronous belt; 902. second turning roller; 9021. third synchronous wheel; 9022. second synchronous belt; 903. transmission rod; 9031. fourth synchronous wheel; 904. second gear; 905. fixing seat; 906. toothed disc; 907. second motor;

[0048] 10. Control box. DETAILED DESCRIPTION

[0049] The technical solution of the present invention is further described below with reference to the accompanying drawings and through specific implementation methods.

[0050] Among them, the drawings are only used for illustrative explanations, and they only represent schematic diagrams rather than actual pictures, and should not be understood as limitations on this patent; in order to better illustrate the embodiments of the present invention, some parts of the drawings may be omitted, enlarged or reduced, and do not represent the size of the actual product; for those skilled in the art, it is understandable that some well-known structures and their descriptions in the drawings may be omitted.

[0051] The same or similar numbers in the drawings of the embodiments of the present invention correspond to the same or similar parts; in the description of the present invention, it should be understood that if the terms "upper", "lower", "left", "right", "inner", "outer", etc. appear, the orientation or position relationship indicated is based on the orientation or position relationship shown in the drawings, which is only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation. Therefore, the terms describing the position relationship in the drawings are only used for illustrative purposes and cannot be understood as a limitation on this patent. For ordinary technicians in this field, the specific meanings of the above terms can be understood according to specific circumstances.

[0052] In the description of the present invention, unless otherwise clearly specified and limited, if the term "connection" or the like appears to indicate the connection relationship between components, the term should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium, it can be the internal connection of two components or the interaction relationship between two components. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0053] Example

[0054] See also Figures 1 to 9 This embodiment provides the following technical solution: a spring roll quality inspection system based on machine vision and artificial intelligence, including a mounting frame 1, an intermittent feeding mechanism 2, an image acquisition mechanism 3, a pushing mechanism 4, a turning mechanism 9 and a control box 10.

[0055] The intermittent feeding mechanism 2 is used to feed the spring rolls 5 on the feeding conveyor belt 6 to the image acquisition mechanism 3, and the intermittent feeding mechanism 2 is arranged on the mounting frame 1;

[0056] The turning mechanism 9 is arranged on the intermittent feeding mechanism 2, and is used to turn over the spring rolls 5 on the intermittent feeding mechanism 2, so that the image acquisition mechanism 3 can perform a full range of detection on the surface of the spring rolls 5;

[0057] The control box 10 is provided with an AI processor and a PLC controller. The image acquisition mechanism 3 sends the collected appearance picture of the spring roll 5 to the AI ​​processor. The AI ​​processor analyzes the picture sent by the image acquisition mechanism 3 and obtains a determination result. The AI ​​processor sends the obtained determination result to the PLC controller. The control box 10 is provided on one side of the mounting frame 1.

[0058] When the result obtained by the AI ​​processor is bad, the PLC controller controls the pushing mechanism 4 to push the bad spring roll 5 onto the bad product conveyor belt 8. When the result obtained by the AI ​​processor is good, the PLC controller controls the pushing mechanism 4 to push the good spring roll 5 onto the good product conveyor belt 7. The pushing mechanism 4 is arranged on the mounting frame 1.

[0059] The turning mechanism 9 and the intermittent feeding mechanism 2 are both controlled by a PLC controller.

[0060] Specifically, in the present embodiment, the intermittent feeding mechanism 2 comprises a feeding disc 201 and a support seat 202, a plurality of feeding grooves 2011 are provided at equal angles on the feeding disc 201, the support seat 202 is fixedly mounted on the top of the mounting frame 1, a cross roller bearing 204 is arranged between the top of the support seat 202 and the bottom of the feeding disc 201, the top of the support seat 202 is fixedly connected to the outer ring of the cross roller bearing 204, the bottom of the feeding disc 201 is fixedly connected to the inner ring of the cross roller bearing 204, an inner toothed annular disc 203 is fixedly mounted on the bottom of the feeding disc 201, the inner toothed annular disc 203 is located on the inner side of the inner ring of the cross roller bearing 204, a first driving motor 205 is fixedly mounted on the top of the mounting frame 1, a first gear 206 is fixedly mounted on the rotating shaft of the first driving motor 205, the first gear 206 is meshed with the inner toothed annular disc 203, and the first driving motor 205 is controlled by a PLC controller.

[0061] When the intermittent feeding mechanism 2 adopting the above technical solution is in use, the spring rolls 5 on the feeding conveyor belt 6 can be transported to the feeding trough 2011 of the feeding tray 201 by controlling the intermittent rotation of the rotating shaft of the first driving motor 205, so as to realize the intermittent feeding operation of the spring rolls 5.

[0062] In this embodiment, the flip mechanism 9 includes a flip assembly, a gear plate 906 and a second motor 907;

[0063] The flip assembly is arranged in the feeding trough 2011, and the flip assembly includes a first flip roller 901 and a second flip roller 902. The first flip roller 901 and the second flip roller 902 are both rotatably connected to the inner lower end of the feeding trough 2011. The toothed disc 906 drives the first flip roller 901 and the second flip roller 902 to rotate synchronously through a transmission member, wherein the transmission member includes a transmission rod 903, a second gear 904 and a fixed seat 905. The fixed seat 905 is fixedly connected to the inner top of the feeding disc 201, and the transmission rod 903 is rotatably installed on the fixed seat 905. A first synchronous wheel 9011 and a second synchronous wheel 9012 are fixedly connected to a rotating shaft at one end of the first flip roller 901. A third synchronous wheel 9021 is fixedly connected to the rotating shaft at one end of the second flipping roller 902, a fourth synchronous wheel 9031 is fixedly connected to one end of the transmission rod 903, the fourth synchronous wheel 9031 is transmission-connected to the second synchronous wheel 9012 via the first synchronous belt 9013, the first synchronous wheel 9011 is transmission-connected to the third synchronous wheel 9021 via the second synchronous belt 9022, the second gear 904 is fixedly connected to the other end of the transmission rod 903, the second gear 904 is meshed with a toothed disc 906, the toothed disc 906 is arranged on the inner side of the feeding tray 201, the second motor 907 is fixedly connected to the top of the mounting frame 1, and the toothed disc 906 is fixedly connected to the rotating shaft of the second motor 907.

[0064] When the flipping mechanism 9 adopting the above technical solution is in use, the rotation of the shaft of the second motor 907 is controlled, thereby driving the toothed disc 906 to rotate. The rotating toothed disc 906 drives the first flipping roller 901 and the second flipping roller 902 to rotate through the transmission member. The rotating first flipping roller 901 and the second flipping roller 902 drive the spring roll 5 to flip, so that the image acquisition mechanism 3 can take a full-range photo of the spring roll 5.

[0065] In order to ensure that the spring rolls 5 falling into the feeding trough 2011 are accurately placed between the first turning roller 901 and the second turning roller 902, the inner walls on both sides of the opposite sides of the feeding trough 2011 are fixedly connected with guide inclined plates 2013, and the guide inclined plates 2013 are located above the first turning roller 901 and the second turning roller 902. The guide inclined plates 2013 can place the spring rolls 5 in the feeding trough 2011 between the first turning roller 901 and the second turning roller 902.

[0066] In this embodiment, the surfaces of the first turning roller 901 and the second turning roller 902 are both provided with anti-slip grooves. This arrangement can ensure that the rotating first turning roller 901 and the second turning roller 902 can drive the spring roll 5 to turn over smoothly and avoid slipping.

[0067] In this embodiment, the image acquisition mechanism 3 includes a first support frame 301, which is fixedly installed on the top of the mounting frame 1. A CCD camera 302 is fixedly installed on the upper bottom of the first support frame 301. The signal output end of the CCD camera 302 is electrically connected to the signal input end of the AI ​​processor. In order to ensure the image shooting effect, a number of fill lights 303 are fixedly installed on the upper bottom of the first support frame 301.

[0068] In this embodiment, the material pushing mechanism 4 includes a second support frame 401, and a qualified product pushing member and a defective product pushing member are arranged at the bottom of the upper end of the second support frame 401;

[0069] The qualified product pusher comprises a first mounting seat 402, on which a first pusher electric push rod 403 is fixedly mounted, and a first push plate 404 is fixedly connected to the piston rod end of the first pusher electric push rod 403;

[0070] The defective product pushing part includes a second mounting seat 405, on which a second pushing electric push rod 406 is fixedly mounted, and the piston rod end of the second pushing electric push rod 406 is fixedly connected to a second push plate 407, and the first pushing electric push rod 403 and the second pushing electric push rod 406 are both controlled by a PLC controller.

[0071] By adopting the pushing mechanism 4 of the above-mentioned technical solution, when in use, when the AI ​​processor determines that the current spring roll 5 is a qualified product, the PLC controller controls the piston rod of the first pushing electric push rod 403 to extend, and then pushes the spring roll 5 to the qualified product conveyor belt 7. When the AI ​​processor determines that the current spring roll 5 is a defective product, the PLC controller first controls the rotating shaft of the first drive motor 205 to continue rotating, and continues to detect the spring roll 5. When the spring roll 5 determined to be a defective product moves to the feed port of the defective product conveyor belt 8, the PLC controller controls the piston rod of the second push plate 407 to extend, and then conveys the defective spring roll 5 to the defective product conveyor belt 8.

[0072] In order to facilitate the first push plate 404 and the second push plate 407 to smoothly push the spring rolls 5 out of the feeding trough 2011, an annular groove 2012 is opened on the top of the feeding tray 201, and the annular groove 2012 is connected to the feeding trough 2011, and the lower end of the first push plate 404 and the lower end of the second push plate 407 are both slidably arranged inside the annular groove 2012.

[0073] The spring roll quality inspection system based on machine vision and artificial intelligence in this embodiment realizes automated and efficient inspection of spring roll quality by integrating advanced machine vision technology and artificial intelligence technology.

[0074] 1. The intermittent feeding mechanism works based on the intermittent feeding algorithm

[0075] The intermittent feeding mechanism realizes the intermittent feeding operation of the spring rolls by controlling the intermittent rotation of the rotating shaft of the first driving motor 205. This process can be simplified into the following steps:

[0076] Set the intermittent time: Set the appropriate intermittent time T according to the speed of the production line and the detection requirements.

[0077] Control the motor rotation: Under the control of the PLC controller, the shaft of the first drive motor 205 rotates at a set intermittent time T.

[0078] Feeding trough positioning: The meshing of the inner-toothed annular disk 203 and the first gear 206 ensures that the feeding trough 2011 on the feeding disk 201 can be accurately positioned directly below the CCD camera 302 .

[0079] 2. The flip mechanism works based on the flip mechanism algorithm

[0080] The flipping mechanism controls the rotation of the shaft of the second motor 907 to drive the first flipping roller 901 and the second flipping roller 902 to rotate synchronously, thereby realizing the flipping operation of the spring rolls. This process involves the following algorithms:

[0081] Synchronous transmission algorithm: through the transmission of the transmission rod 903, the second gear 904, the synchronous wheel and the synchronous belt, it is ensured that the first flip roller 901 and the second flip roller 902 can rotate synchronously.

[0082] Flip angle control: According to the shape of the spring roll and the detection requirements, a suitable flip angle θ is set, which is achieved by controlling the rotation angle of the shaft of the second motor 907 .

[0083] 3. Image acquisition mechanism 3 works based on machine vision algorithm

[0084] The machine vision algorithm is the core part of the system, which mainly includes steps such as image acquisition, preprocessing, feature extraction and classification recognition.

[0085] Image acquisition: Use CCD camera 302 to shoot the spring rolls to obtain high-quality image data.

[0086] Image preprocessing: Enhance, denoise, and perform geometric correction on the collected images to improve image quality.

[0087] Feature extraction: Extract the shape, color, texture and other feature information of the spring rolls from the preprocessed image.

[0088] Classification and recognition: Use classifiers such as support vector machines and neural networks to classify and recognize the extracted features to obtain the quality judgment results of the spring rolls.

[0089] 4.PLC controller works based on PLC controller algorithm

[0090] The PLC controller is responsible for coordinating the operation of the entire system, including the control of the intermittent feeding mechanism, the flipping mechanism, the pushing mechanism, etc. Its algorithm mainly includes:

[0091] Status monitoring: Real-time monitoring of the operating status and location information of each mechanism.

[0092] Logical control: Execute the corresponding control logic according to the judgment results of the machine vision algorithm and the operating status of each mechanism.

[0093] Fault handling: When a fault occurs, an alarm is issued and appropriate fault handling measures are taken.

[0094] 5. Key technical points

[0095] Machine vision technology: realizes the collection, preprocessing, feature extraction and classification recognition of spring roll images.

[0096] Artificial intelligence technology: Use deep learning algorithms to train large amounts of image data to improve the accuracy and robustness of classification and recognition.

[0097] Automation control technology: Coordinated operation and fault handling of various mechanisms are achieved through PLC controller.

[0098] High-precision transmission technology: ensures accurate positioning and synchronous transmission of the turning mechanism and intermittent feeding mechanism.

[0099] In summary, the system realizes the automatic and efficient detection of spring roll quality by integrating machine vision technology and artificial intelligence technology. The algorithms and technical points involved in its workflow reflect the advancement and practicality of the system.

[0100] In order to facilitate understanding of the above technical solutions of the present invention, the working principle or operation mode of the present invention in the actual process is described in detail below.

[0101] Working principle: When in use, the spring rolls 5 on the feeding conveyor belt 6 can be transported to the feeding slot 2011 of the feeding tray 201 by controlling the rotating shaft of the first driving motor 205 to rotate intermittently. When the spring rolls 5 rotate to the position directly below the CCD camera 302, the PLC controller controls the first driving motor 205 to stop, and then the CCD camera 302 takes a picture of the spring rolls 5. At the same time, the PLC controller controls the rotating shaft of the second motor 907 to rotate, thereby rotating the first turning roller 901 and the second turning roller 902. The rotating first turning roller 901 and the second turning roller 902 turn the spring rolls 5, so that the CCD camera 302 takes a full-scale picture of the spring rolls 5. The CCD camera 302 takes the captured picture. The film is sent to the AI ​​processor, and the AI ​​processor analyzes and processes the image transmitted by the CCD camera 302 to obtain a judgment result. When the AI ​​processor determines that the current spring roll 5 is a qualified product, the PLC controller controls the piston rod of the first pushing electric push rod 403 to extend, and then pushes the spring roll 5 to the qualified product conveyor belt 7. When the AI ​​processor determines that the current spring roll 5 is a defective product, the PLC controller first controls the rotating shaft of the first drive motor 205 to continue rotating, and continues to detect the spring roll 5. When the spring roll 5 determined to be a defective product moves to the feed port of the defective product conveyor belt 8, the PLC controller controls the piston rod of the second push plate 407 to extend, and then conveys the defective spring roll 5 to the defective product conveyor belt 8.

[0102] Parts not involved in the present invention are the same as the prior art or can be implemented by the prior art. Although the embodiments of the present invention have been shown and described, it is understood by those skilled in the art that various changes, modifications, substitutions and variations can be made to these embodiments without departing from the principles and spirit of the present invention, and the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A spring roll quality detection system based on machine vision and artificial intelligence, characterized in that: include: Mounting frame (1); An intermittent feeding mechanism (2), the intermittent feeding mechanism (2) being used to feed the spring rolls (5) on the feeding conveyor belt (6) to the image acquisition mechanism (3), the intermittent feeding mechanism (2) being arranged on the mounting frame (1); The image acquisition mechanism (3) is used to acquire appearance data of the spring rolls (5), and the image acquisition mechanism (3) is arranged on the mounting frame (1); a turning mechanism (9), the turning mechanism (9) being arranged on the intermittent feeding mechanism (2), the turning mechanism (9) being used to turn over the spring rolls (5) on the intermittent feeding mechanism (2), so that the image acquisition mechanism (3) can perform all-round detection on the surface of the spring rolls (5); A control box (10), wherein an AI processor and a PLC controller are arranged in the control box (10), the image acquisition mechanism (3) sends the collected appearance picture of the spring roll (5) to the AI ​​processor, the AI ​​processor analyzes the picture sent by the image acquisition mechanism (3) and obtains a determination result, and the AI ​​processor sends the obtained determination result to the PLC controller, and the control box (10) is arranged on one side of the mounting frame (1); A pushing mechanism (4), when the result obtained by the AI ​​processor is bad, the PLC controller controls the pushing mechanism (4) to push the bad spring roll (5) onto a bad product conveyor belt (8), and when the result obtained by the AI ​​processor is good, the PLC controller controls the pushing mechanism (4) to push the good spring roll (5) onto a good product conveyor belt (7), the pushing mechanism (4) being arranged on the mounting frame (1); The turning mechanism (9) and the intermittent feeding mechanism (2) are both controlled by the PLC controller.

2. The spring roll quality detection system based on machine vision and artificial intelligence according to claim 1, characterized in that: The intermittent feeding mechanism (2) comprises a feeding tray (201) and a support seat (202); a plurality of feeding grooves (2011) are provided on the feeding tray (201) at equal angles; the support seat (202) is fixedly mounted on the top of the mounting frame (1); a cross roller bearing (204) is arranged between the top of the support seat (202) and the bottom of the feeding tray (201); the top of the support seat (202) is fixedly connected to the outer ring of the cross roller bearing (204); the bottom of the feeding tray (201) is fixedly connected to the cross roller bearing ( The inner ring of the feeding tray (204) is fixedly connected, an inner toothed annular disk (203) is fixedly mounted on the bottom of the feeding tray (201), the inner toothed annular disk (203) is located on the inner side of the inner ring of the cross roller bearing (204), a first driving motor (205) is fixedly mounted on the top of the mounting frame (1), a first gear (206) is fixedly mounted on the rotating shaft of the first driving motor (205), the first gear (206) is meshed with the inner toothed annular disk (203), and the first driving motor (205) is controlled by the PLC controller.

3. The spring roll quality detection system based on machine vision and artificial intelligence according to claim 2, characterized in that: The turning mechanism (9) comprises a turning assembly, a toothed disc (906) and a second motor (907); The flip assembly is arranged in the feeding trough (2011), and the flip assembly includes a first flip roller (901) and a second flip roller (902), the first flip roller (901) and the second flip roller (902) are both rotatably connected to the inner lower end of the feeding trough (2011), the toothed disc (906) drives the first flip roller (901) and the second flip roller (902) to rotate synchronously through a transmission member, the toothed disc (906) is arranged on the inner side of the feeding disc (201), and the second motor (907) is fixedly connected to the feeding disc (201). The toothed disc (906) is connected to the top of the mounting frame (1), and is fixedly connected to the rotating shaft of the second motor (907). The rotating shaft of the second motor (907) is controlled to rotate, thereby driving the toothed disc (906) to rotate. The rotating toothed disc (906) drives the first turning roller (901) and the second turning roller (902) to rotate through a transmission member. The rotating first turning roller (901) and the second turning roller (902) drive the spring roll (5) to turn over, so that the image acquisition mechanism (3) can take all-round photos of the spring roll (5).

4. The spring roll quality detection system based on machine vision and artificial intelligence according to claim 3, characterized in that: The transmission member comprises a transmission rod (903), a second gear (904) and a fixed seat (905); the fixed seat (905) is fixedly connected to the inner top of the feeding tray (201); the transmission rod (903) is rotatably mounted on the fixed seat (905); a first synchronous wheel (9011) and a second synchronous wheel (9012) are fixedly connected to a rotating shaft at one end of the first flipping roller (901); a third synchronous wheel (9021) is fixedly connected to a rotating shaft at one end of the second flipping roller (902); One end of the transmission rod (903) is fixedly connected to a fourth synchronous wheel (9031); the fourth synchronous wheel (9031) is transmission-connected to the second synchronous wheel (9012) via a first synchronous belt (9013); the first synchronous wheel (9011) is transmission-connected to the third synchronous wheel (9021) via a second synchronous belt (9022); the second gear (904) is fixedly connected to the other end of the transmission rod (903); the second gear (904) is meshed with the toothed disc (906).

5. The spring roll quality detection system based on machine vision and artificial intelligence according to claim 4, characterized in that: The inner walls on both sides of the feeding trough (2011) are fixedly connected with guide inclined plates (2013), and the guide inclined plates (2013) are located above the first turning roller (901) and the second turning roller (902). The guide inclined plates (2013) can place the spring rolls (5) in the feeding trough (2011) between the first turning roller (901) and the second turning roller (902).

6. The spring roll quality detection system based on machine vision and artificial intelligence according to claim 4, characterized in that: The surface of the first turning roller (901) and the surface of the second turning roller (902) are both provided with anti-slip grooves.

7. The spring roll quality detection system based on machine vision and artificial intelligence according to claim 4, characterized in that: The image acquisition mechanism (3) comprises a first support frame (301), the first support frame (301) is fixedly mounted on the top of the mounting frame (1), a CCD camera (302) is fixedly mounted on the bottom of the upper end of the first support frame (301), and a signal output end of the CCD camera (302) is electrically connected to a signal input end of the AI ​​processor.

8. The spring roll quality detection system based on machine vision and artificial intelligence according to claim 7, characterized in that: A plurality of fill lights (303) are fixedly mounted on the bottom of the upper end of the first support frame (301).

9. The spring roll quality detection system based on machine vision and artificial intelligence according to claim 8, characterized in that: The material pushing mechanism (4) comprises a second support frame (401), and a qualified product pushing member and a defective product pushing member are arranged at the bottom of the upper end of the second support frame (401); The qualified product pushing member comprises a first mounting seat (402), a first pushing electric push rod (403) is fixedly mounted on the first mounting seat (402), and a first push plate (404) is fixedly connected to the piston rod end of the first pushing electric push rod (403); The defective product pushing member includes a second mounting seat (405), on which a second pushing electric push rod (406) is fixedly mounted, and the piston rod end of the second pushing electric push rod (406) is fixedly connected to a second push plate (407), and both the first pushing electric push rod (403) and the second pushing electric push rod (406) are controlled by the PLC controller.

10. The spring roll quality detection system based on machine vision and artificial intelligence according to claim 9, characterized in that: An annular groove (2012) is provided on the top of the feeding tray (201), and the annular groove (2012) is connected to the feeding groove (2011), and the lower end of the first push plate (404) and the lower end of the second push plate (407) are both slidably arranged inside the annular groove (2012).

Images