Product faceted screening method, sorting device, electronic equipment and medium

Abstract

The invention discloses a product faceted screening method, a sorting device, electronic equipment and a medium, and relates to the technical field of product sorting. The method comprises the steps that when a product is conveyed along a conveying track, an image sent by a photographing mechanism is acquired; the image is obtained by photographing the product on the transmission track by the photographing mechanism according to a preset frequency; performing instance segmentation on the image according to a pre-trained AI model to obtain a positioning identifier in the image; dividing the image according to the coordinates of the positioning identifier, and determining a judgment area; performing feature recognition on the products in the judgment area; when the facet features of the product cannot be recognized, the blowing module is started; and the products are blown to the material returning track through the air blowing module, so that the products return to the feeding mechanism through the material returning track. According to the method, the facet features of the product can be accurately identified through the AI model, and even the pose form change of the product can be stably detected.

Description

Technical Field

[0001] The present invention relates to the technical field of product sorting, and in particular to a product facet screening method, a sorting device, an electronic device and a medium. Background Art

[0002] Currently, the following methods are commonly used for faceted screening of samples:

[0003] 1. Sensor faceting: Utilizes sensors (such as optical fiber, photoelectric, color, eddy current, etc.) to identify product surface features. When the features of the product's surface to be tested match those of the discharge surface, the product passes through. If they do not match, the product is blown over or ejected. The disadvantage is that it can only identify products with single or distinct features. It is difficult to identify products with complex surfaces or less distinct features, and cannot guarantee feeding speed and accuracy.

[0004] 2. Traditional visual image processing algorithm faceting: This algorithm uses an industrial camera to capture images and then extract image features. Disadvantages: a. Because product feeding relies on the vibration of a vibrating plate, the product experiences random jitter as it moves on the plate. This can lead to dramatic changes in product lighting or large rotation amplitudes, making it difficult to stably identify image features. This leads to a high rate of false positives and negatives, impacting feeding speed and accuracy. b. Minor differences between product batches require experienced engineers to adjust image parameters when changing batches, impacting efficiency. Summary of the Invention

[0005] The present invention aims to address at least one of the technical problems existing in the prior art. To this end, the present invention proposes a product facet screening method, sorting device, electronic device, and medium. Using an AI model, the method accurately identifies the facet features of products, ensuring stable detection even when the product's posture and morphology change.

[0006] In a first aspect, a product facet screening method according to an embodiment of the present invention is applied to a host computer of a sorting device, wherein the sorting device includes a loading mechanism, a transmission track, a blowing module, a return track, and a photographing mechanism, wherein the loading mechanism is used to load products and transmit the products in sequence along the transmission track, the blowing module is arranged on one side of the transmission track, the return track is arranged on the other side of the transmission track, the return track is opposite to the blowing module, the return track is connected to the loading mechanism, the lens of the photographing mechanism is located above the transmission track, and a positioning mark is provided on the transmission track corresponding to the blowing module; the method includes:

[0007] When the product is transported along the transport track, an image sent by the photographing mechanism is acquired; the image is acquired by the photographing mechanism taking pictures of the product on the transport track at a preset frequency;

[0008] Perform instance segmentation on the image according to a pre-trained AI model to obtain the positioning marker in the image;

[0009] Dividing the image according to the coordinates of the positioning marker to determine a determination area;

[0010] Performing feature recognition on the product within the determination area;

[0011] When the facet features of the product cannot be identified, starting the blowing module;

[0012] The product is blown to the return track by the air blowing module, so that the product returns to the loading mechanism through the return track.

[0013] According to some embodiments of the present invention, the feeding mechanism includes a feeding funnel and a vibrating plate, the vibrating plate is connected to the feeding funnel, and the vibrating plate is connected to the entrance of the transmission track; when the product is transported along the transmission track, before the step of acquiring the image sent by the camera mechanism, the method further includes:

[0014] Put the product into the vibrating plate through the feeding funnel;

[0015] The products are arranged into the transmission track by using a frequency modulation vibration method through the vibration plate.

[0016] According to some embodiments of the present invention, the photographing mechanism includes:

[0017] Multi-axis stent;

[0018] A plurality of cameras are arranged on the multi-axis bracket, each of the cameras is provided with a lens, and the cameras are electrically connected to the host computer;

[0019] A plurality of light sources, each corresponding to each lens, and each light source is disposed at the corresponding lens;

[0020] The light source controller is electrically connected to the light source and the host computer respectively.

[0021] According to some embodiments of the present invention, the camera is provided with a first data interface and a second data interface, the camera is electrically connected to the host computer through the first data interface, the blowing module is provided with a solenoid valve, and the camera is electrically connected to the solenoid valve through the second data interface.

[0022] According to some embodiments of the present invention, the AI ​​model is trained through the following steps:

[0023] Collect sample pictures;

[0024] Classify and label the product features in the sample image;

[0025] Deep learning training is performed on the sample images after classification and annotation to obtain the AI ​​model.

[0026] According to some embodiments of the present invention, when the product is transported along the transport track, acquiring the image sent by the photographing mechanism includes:

[0027] Determining the image processing time of the host computer;

[0028] determining a photographing frequency of the photographing mechanism according to the image processing time;

[0029] When the product is transported along the transport track, the product is photographed by the photographing mechanism according to the photographing frequency to obtain an image of the product.

[0030] According to some embodiments of the present invention, blowing the product to the return track by the blowing module so that the product returns to the loading mechanism through the return track includes:

[0031] Obtaining the weight of the product and the transport speed of the transport track;

[0032] Adjusting the opening of the solenoid valve of the blowing module according to the weight of the product and the transmission speed of the transmission track;

[0033] The solenoid valve is opened to the opening degree, so that the blowing module blows the product to the return track, and the product returns to the loading mechanism through the return track.

[0034] In the second aspect, according to the sorting device of the embodiment of the present invention, the sorting device comprises a loading mechanism, a transmission track, a blowing module, a return track and a photographing mechanism, wherein the loading mechanism is used to load products and transmit the products in sequence along the transmission track, the blowing module is arranged on one side of the transmission track, the return track is arranged on the other side of the transmission track, the return track is opposite to the blowing module, the return track is connected to the loading mechanism, the lens of the photographing mechanism is located above the transmission track, and a positioning mark is provided on the transmission track at a position corresponding to the blowing module; the sorting device also includes a host computer, which is used to implement the product facet screening method as described in the embodiment of the first aspect.

[0035] In the third aspect, a storage medium according to an embodiment of the present invention includes at least one control processor and a memory for communicating with the at least one control processor; the memory stores instructions that can be executed by the at least one control processor, and the instructions are executed by the at least one control processor to enable the at least one control processor to execute the product faceted screening method as described in the embodiment of the first aspect.

[0036] In a fourth aspect, a storage medium according to an embodiment of the present invention stores computer-executable instructions, and the computer-executable instructions are used to enable a computer to execute the product faceted screening method described in the embodiment of the first aspect.

[0037] The product faceting screening method, sorting device, electronic device, and medium according to the embodiments of the present invention have at least the following beneficial effects: an image of a product being transported along a transport track is captured by a camera mechanism, and the image is instance-segmented using an AI model to obtain a positioning identifier. The image is then secondary segmented based on the positioning identifier to determine a judgment area. Finally, the product in the judgment area is feature-recognized, thereby determining the action of the blowing module based on the recognition results. The AI ​​model uses a deep learning algorithm to accurately identify product features and can stably detect changes in product posture and morphology, thereby improving feeding speed and faceting accuracy. When differences exist between product batches, only new samples need to be added to train the AI ​​model, without adjusting parameters, thereby reducing maintenance frequency and the number of parameter adjustments.

[0038] Additional aspects and advantages of the present invention will be set forth in part in the description which follows and, in part, will be obvious from the description which follows, or may be learned by practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] The above and / or additional aspects and advantages of the present invention will become apparent and readily understood from the following description of the embodiments with reference to the accompanying drawings, in which:

[0040] Figure 1 This is a flowchart of the steps of the product faceted screening method according to an embodiment of the present invention;

[0041] Figure 2 Schematic diagram of the structure of a sorting device according to an embodiment of the present invention;

[0042] Figure 3 Schematic diagram of the structure of the photographing mechanism according to an embodiment of the present invention;

[0043] Figure 4 Schematic diagram of electrical connections of a sorting device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0044] The embodiments of the present invention are described in detail below. Examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present application and are not to be construed as limiting the present application. The step numbers in the following embodiments are provided only for the convenience of explanation and do not limit the order of the steps. The order of execution of the steps in the embodiments can be adaptively adjusted according to the understanding of those skilled in the art.

[0045] In the description of the present invention, it should be understood that descriptions involving orientations, such as up, down, front, back, left, right, etc., indicating orientations or positional relationships, are based on the orientations or positional relationships shown in the accompanying drawings. They are only for the convenience of describing the present invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation. Therefore, they cannot be understood as limitations on the present invention.

[0046] The terms "first," "second," "third," and "fourth," etc., in the specification, claims, and accompanying drawings of the present invention are used to distinguish between different items, not to describe a specific order. Furthermore, the terms "including," "having," and any variations thereof, are intended to cover non-exclusive inclusions. For example, a process, method, system, product, or apparatus comprising a series of steps or elements is not limited to the listed steps or elements, but may optionally include steps or elements not listed, or may optionally include other steps or elements inherent to the process, method, product, or apparatus.

[0047] References to "embodiments" in this disclosure mean that a particular feature, structure, or characteristic described in connection with the embodiment may be included in at least one embodiment of the disclosure. The appearance of this phrase in various places in the specification does not necessarily refer to the same embodiment, nor does it constitute an independent or alternative embodiment that is mutually exclusive of other embodiments. It is understood, both explicitly and implicitly, by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0048] Currently, when loading some electronic components or other products in batches, it is necessary to ensure that the loading direction of the products is consistent to avoid some products from turning over or falling over. Therefore, when the products are discharged, they need to be screened by facet to ensure that the discharge direction of the products is consistent. Currently, the following methods are usually used for facet screening:

[0049] 1. Sensor faceting: Utilizes sensors (such as optical fiber, photoelectric, color, eddy current, etc.) to identify product surface features. When the features of the product's surface to be tested match those of the discharge surface, the product passes through. If they do not match, the product is blown over or ejected. The disadvantage is that it can only identify products with single or distinct features. It is difficult to identify products with complex surfaces or less distinct features, and cannot guarantee feeding speed and accuracy.

[0050] 2. Traditional visual image processing algorithm faceting: This algorithm uses an industrial camera to capture images and then extract image features. Disadvantages: a. Because product feeding relies on the vibration of a vibrating plate, the product experiences random jitter as it moves on the plate. This can lead to dramatic changes in product lighting or large rotation amplitudes, making it difficult to stably identify image features. This leads to a high rate of false positives and negatives, impacting feeding speed and accuracy. b. Minor differences between product batches require experienced engineers to adjust image parameters when changing batches, impacting efficiency.

[0051] To this end, embodiments of the present invention provide a product faceting screening method, sorting device, electronic device, and medium. Images of products transported along a conveyor track are captured by a camera mechanism, and the images are segmented by an AI model to obtain positioning identifiers. The images are then segmented again based on the positioning identifiers to determine a judgment area. Finally, features of the products in the judgment area are identified, and the action of the blowing module is determined based on the identification results. The AI ​​model uses a deep learning algorithm to accurately identify product features, allowing for stable detection even when product posture and morphology change, thereby improving feeding speed and faceting accuracy. When differences exist between product batches, only new samples need to be added to train the AI ​​model, eliminating the need to adjust parameters, thereby reducing maintenance frequency and the number of parameter adjustments.

[0052] The product faceted screening method, sorting device, electronic device and medium according to the embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0053] On the one hand, the embodiment of the present invention proposes a product facet screening method, which is applied to the host computer 600 of the sorting device, such as Figure 2 and Figure 3 As shown, the sorting device includes a feeding mechanism 100, a transmission track 200, a blowing module 300, a return track 400 and a photographing mechanism 500, wherein the feeding mechanism 100 is used to feed the products and transmit the products in sequence along the transmission track 200, the blowing module 300 is arranged on one side of the transmission track 200, and the return track 400 is arranged on the other side of the transmission track 200, the return track 400 is opposite to the blowing module 300, and the return track 400 is connected to the feeding mechanism 100, the lens 521 of the photographing mechanism 500 is located above the transmission track 200, and a positioning mark is provided on the transmission track 200 corresponding to the blowing module 300. Figure 1As shown, the product faceted screening method includes but is not limited to steps S100 to S600:

[0054] Step S100: When the product is transported along the transport track 200, an image sent by the photographing mechanism 500 is obtained; the image is obtained by the photographing mechanism 500 photographing the product on the transport track 200 at a preset frequency;

[0055] Step S200: Perform instance segmentation on the image based on the pre-trained AI model to obtain positioning markers in the image;

[0056] Step S300: dividing the image according to the coordinates of the positioning marker to determine the determination area;

[0057] Step S400: performing feature recognition on products within the determination area;

[0058] Step S500: When the facet features of the product cannot be identified, the blowing module 300 is activated;

[0059] Step S600: The product is blown to the return track 400 by the blowing module 300, so that the product returns to the loading mechanism through the return track 400.

[0060] Specifically, when the loading mechanism 100 loads a product and the product is transported along the conveyor track 200, the camera mechanism 500 photographs the product and obtains an image, which is then transmitted to the host computer 600. Using a trained AI model, the host computer 600 performs instance segmentation on the image, identifying the positioning markers on the conveyor track 200 in the image. Based on the coordinates of the positioning markers, the image is then divided and secondary positioning is performed to determine the judgment area. The AI ​​model then continues to perform feature recognition on the product within the judgment area. If the product's facet features are present within the judgment area, a determination of "OK" is made, indicating that the product has passed sorting and can be transferred to the next process. At this point, the air blowing module 300 does not need to be activated. If the product's facet features are not detected within the judgment area, a determination of "NG" is made, and an output signal is output to activate the air blowing module 300, which then discharges the product onto the return track 400. The product then returns to the loading mechanism via the return track 400, facilitating reloading by the loading mechanism.

[0061] like Figure 2 As shown, in some embodiments of the present application, the loading mechanism 100 includes a loading hopper 110 and a vibration plate 120. The vibration plate 120 is disposed in the loading hopper 110 and is connected to the entrance of the transmission track 200. The above-mentioned step S100: when the product is transported along the transmission track 200, before acquiring the image sent by the camera mechanism 500, the following two steps are also included:

[0062] The product is put into the vibrating plate 120 through the feeding funnel 110;

[0063] The products are arranged into the transmission track 200 by using a frequency modulation vibration method through the vibration plate 120.

[0064] Specifically, when loading products, batches of products are placed into the loading hopper 110 so that the products are located on the surface of the vibration plate 120. At this time, the vibration plate 120 uses frequency-modulated vibration waves to vibrate the products, so that the products are sequentially arranged and transported on the transmission track 200. In this example, the transmission track 200 is a linear vibration track, so that the products move along the transmission track 200 in an orderly manner.

[0065] like Figure 3 and Figure 4 As shown, in some embodiments of the present application, the photographing mechanism 500 includes: a multi-axis support 510, a plurality of cameras 520, a plurality of light sources 530 and a light source controller 540, wherein the camera 520 is arranged on the multi-axis support 510, each camera 520 is provided with a lens 521, and the camera 520 is electrically connected to the host computer 600; the light source 530 corresponds to the lens 521 one by one, and the light source 530 is arranged at the corresponding lens 521; the light source controller 540 is electrically connected to the light source 530 and the host computer 600 respectively. Among them, the multi-axis support 510 can adjust the posture, direction and position of the camera 520 in multiple directions, and the camera 520 is located above the blowing port of the blowing module 300, so that the lens 521 shoots the product from top to bottom, and the blowing port of the blowing module 300 is located in the center of the field of view of the camera 520. As shown Figure 4 As shown, the turning on and off of the light source 530, as well as the specific brightness, are controlled by the light source controller 540, and the light source controller 540 can receive signals from the motion control card 710 of the appearance sorting machine 700 (sorting device), so that the appearance sorting machine 700 can control the light source controller 540 through the motion control card 710.

[0066] like Figure 3As shown, in some embodiments of the present application, the camera 520 is provided with a first data interface 522 and a second data interface 523. The camera 520 is electrically connected to the host computer 600 via the first data interface 522. The air blowing module 530 is provided with a solenoid valve, and the camera 520 is electrically connected to the solenoid valve via the second data interface 523. Among them, the first data interface 522 can adopt a common interface such as USB3.0, GigE, 10GigE, CameraLink, CoaXpress, etc., and the first data interface 522 is used to realize data communication between the camera 520 and the host computer 600, so that the camera 520 can send the captured image to the host computer 600, and the host computer 600 can send a control signal to the camera 520 based on the result of image processing, so that the camera 520 controls the opening and closing of the solenoid valve of the air blowing module 530 via the second data interface 523. When the upper computer 600 determines through image recognition that the orientation of the product is normal, the solenoid valve is in the off state and the blowing module 530 does not blow air; when it determines that the orientation of the product is abnormal, the camera 520 controls the solenoid valve to be in the open state, and the blowing module 530 starts working to blow the product to the return track 400, so that the product returns to the vibration plate 120 through the return track 400, and under the action of the vibration plate 120, re-enters the transmission track 200.

[0067] Furthermore, in some embodiments of the present application, the AI ​​model in the host computer 600 is trained by the following steps:

[0068] Collect sample pictures;

[0069] Classify and label product features in sample images;

[0070] Perform deep learning training on the classified and labeled sample images to obtain an AI model.

[0071] In this embodiment, the sample is poured into the installed vibrating plate 120 of the appearance sorting machine. The vibrating plate 120 is then started, and the camera 520 takes real-time images. After the vibrating plate 120 runs smoothly for a period of time, the image acquisition is terminated. The cached sample images are classified and labeled according to image features. Deep learning training is performed using the labeled sample data to generate an AI model. After the AI ​​model is trained, during the screening process, images are captured in real time by the camera 520 and transmitted to the host computer 600. The host computer 600 receives the images from the camera 520, uses the trained AI model to perform image instance segmentation, and performs image judgment. Finally, the blowing action is determined based on the judgment results.

[0072] Furthermore, in some embodiments of the present application, the above-mentioned step S100: when the product is transported along the transport track 200, acquiring the image sent by the photographing mechanism 500, includes the following three sub-steps:

[0073] Step S110: determining the image processing time of the host computer 600;

[0074] Step S120: determining the photographing frequency of the photographing mechanism 500 according to the image processing time;

[0075] Step S130: When the product is transported along the transport track 200, the photographing mechanism 500 photographs the product according to the photographing frequency to obtain an image of the product.

[0076] In this example, the average image processing time (T) of the host computer 600 is 5-6ms. For this reason, the shooting frequency of the camera 520 is set to 200fps so that the shooting frequency of the camera 520 is adapted to the image processing time of the host computer 600, ensuring that the feeding speed is guaranteed while the image processing is accurate.

[0077] Furthermore, in some embodiments of the present application, the above-mentioned step S600: blowing the product to the return track 400 by the blowing module 300 so that the product returns to the loading mechanism 500 through the return track 400 includes the following three sub-steps:

[0078] Step S610: Obtain the weight of the product and the transmission speed of the transmission track 200;

[0079] Step S620: adjusting the opening of the solenoid valve of the blowing module 300 according to the weight of the product and the transmission speed of the transmission track 200;

[0080] Step S630 : Open the solenoid valve to an opening degree, so that the blowing module 300 blows the product to the return track 400 , so that the product returns to the loading mechanism 100 through the return track 400 .

[0081] In this example, the opening of the solenoid valve is adjusted according to the product weight and the feeding speed to achieve the best blowing and discharging effect without carrying the product, ensuring that the blowing module 300 can blow the product back to the return track 400.

[0082] Finally, you can open the sorting software of the host computer 600 to confirm the sorting effect. If there is a discharge error, review the cached image, add the misjudged image to the sample library, annotate it, and retrain the model. Repeat this step until the false positive rate meets the standard.

[0083] According to the product faceting screening method of the embodiment of the present application, a camera mechanism 500 captures an image of a product being transported along the transport track 200. The AI ​​model then performs instance segmentation on the image to obtain positioning markers. The image is then secondary segmented based on the positioning markers to determine a determination area. Finally, the product in the determination area is feature-recognized, and the action of the blowing module 300 is determined based on the recognition results. The AI ​​model utilizes a deep learning algorithm to accurately identify product features, allowing for stable detection even when the product's posture and morphology change, thereby improving feeding speed and faceting accuracy. When differences exist between product batches, only new samples need to be added to train the AI ​​model, eliminating the need for parameter adjustment, thereby reducing maintenance frequency and the number of parameter adjustments.

[0084] On the other hand, the embodiment of the present invention also provides a sorting device, such as Figures 2 to 4 As shown, the sorting device includes a loading mechanism 100, a transmission track 200, a blowing module 300, a return track 400 and a photographing mechanism 500, wherein the loading mechanism 100 is used to load the products and transmit the products in sequence along the transmission track 200, the blowing module 300 is arranged on one side of the transmission track 200, and the return track 400 is arranged on the other side of the transmission track 200, the return track 400 is opposite to the blowing module 300, and the return track 400 is connected to the loading mechanism 100, the lens of the photographing mechanism 500 is located above the transmission track 200, and a positioning mark is provided on the position of the transmission track 200 corresponding to the blowing module 300; the sorting device also includes a host computer 600, which is used to implement the product facet screening method described in the above-mentioned embodiment.

[0085] It should be noted that the contents of the above method embodiments are all applicable to the present device embodiments. The functions specifically implemented by this embodiment are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0086] On the other hand, an embodiment of the present invention further provides a storage medium, which is a computer-readable storage medium and stores a computer program. When the computer program is executed by a processor, the above-mentioned product faceted screening method is implemented.

[0087] The memory, as a non-transient computer-readable storage medium, can be used to store non-transient software programs and non-transient computer executable programs. In addition, the memory may include a high-speed random access memory, and may also include a non-transient memory, such as at least one disk storage device, a flash memory device, or other non-transient solid-state storage device. In some embodiments, the memory optionally includes a memory remotely located relative to the processor, and these remote memories can be connected to the processor via a network. Examples of the above-mentioned networks include but are not limited to the Internet, an intranet, a local area network, a mobile communication network and a combination thereof. The device embodiments described above are merely schematic, wherein the units described as separate components may or may not be physically separated, and are located in one place, or may be distributed to multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the present embodiment.

[0088] Although specific embodiments are described herein, those skilled in the art will recognize that many other modifications or alternative embodiments are also within the scope of this disclosure. For example, any of the functions and / or processing capabilities described in conjunction with a particular device or component may be performed by any other device or component. In addition, although various exemplary implementations and architectures have been described in accordance with embodiments of the present disclosure, those skilled in the art will recognize that many other modifications to the exemplary implementations and architectures described herein are also within the scope of this disclosure.

[0089] Some aspects of the present disclosure have been described above with reference to the block diagrams and flow charts of the systems, methods, systems and / or computer program products according to the exemplary embodiments. It should be understood that the combination of one or more blocks in the block diagram and the flow chart and the blocks in the block diagram and the flow chart can be realized by executing computer executable program instructions respectively. Equally, according to some embodiments, some blocks in the block diagram and the flow chart may not need to be executed in the order shown, or may not need to be executed in full. In addition, additional components and / or operations beyond those components and / or operations shown in the blocks in the block diagram and the flow chart may be present in certain embodiments.

[0090] Therefore, the blocks in the block diagrams and flow charts support combinations of means for performing the specified functions, combinations of elements or steps for performing the specified functions, and program instruction means for performing the specified functions. It should also be understood that each block in the block diagrams and flow charts, and combinations of blocks in the block diagrams and flow charts, can be implemented by a dedicated hardware computer system that performs the specific functions, elements, or steps, or a combination of dedicated hardware and computer instructions.

[0091] The program modules, applications, etc. described herein may include one or more software components, including, for example, software objects, methods, data structures, etc. Each such software component may include computer-executable instructions that, in response to execution, cause at least a portion of the functionality described herein (e.g., one or more operations of the illustrative methods described herein) to be performed.

[0092] Software component can be encoded with any one in various programming languages.A kind of exemplary programming language can be low-level programming language, such as the assembly language associated with specific hardware architecture and / or operating system platform.Comprise that the software component of assembly language instruction may need to be converted to executable machine code by assembler before being executed by hardware architecture and / or platform.Another exemplary programming language can be a more advanced programming language, and it can be transplanted across multiple architectures.Comprise that the software component of more advanced programming language may need to be converted to intermediate representation by interpreter or compiler before execution.Other examples of programming language include but are not limited to macro language, shell or command language, job control language, script language, database query or search language or report writing language.In one or more exemplary embodiments, the software component that comprises the instruction of one in the above-mentioned programming language example can be directly executed by operating system or other software component, without first being converted into another form.

[0093] Software components can be stored as files or other data storage structures. Software components of similar types or related functions can be stored together, such as in a specific directory, folder, or library. Software components can be static (e.g., preset or fixed) or dynamic (e.g., created or modified at execution time).

[0094] The embodiments of the present invention are described in detail above with reference to the accompanying drawings. However, the present invention is not limited to the above embodiments. Various changes can be made within the knowledge of ordinary technicians in the relevant technical field without departing from the scope of the present invention.

Claims

1. A product faceted screening method, characterized in that: A host computer is applied to a sorting device, the sorting device includes a loading mechanism, a transmission track, a blowing module, a return track and a photographing mechanism, wherein the loading mechanism is used to load products and transmit the products in sequence along the transmission track, the blowing module is arranged on one side of the transmission track, the return track is arranged on the other side of the transmission track, the return track is opposite to the blowing module, the return track is connected to the loading mechanism, the lens of the photographing mechanism is located above the transmission track, and a positioning mark is provided on the transmission track corresponding to the blowing module; the method includes: When the product is transported along the transport track, an image sent by the photographing mechanism is acquired; the image is acquired by the photographing mechanism taking pictures of the product on the transport track at a preset frequency; Perform instance segmentation on the image according to a pre-trained AI model to obtain the positioning marker in the image; Dividing the image according to the coordinates of the positioning marker to determine a determination area; Performing feature recognition on the product within the determination area; When the facet features of the product cannot be identified, starting the blowing module; The product is blown to the return track by the air blowing module, so that the product returns to the loading mechanism through the return track.

2. The product faceted screening method according to claim 1, characterized in that: The feeding mechanism includes a feeding funnel and a vibration plate, the vibration plate is connected to the feeding funnel, and the vibration plate is connected to the entrance of the transmission track; when the product is transported along the transmission track, before the step of acquiring the image sent by the camera mechanism, the step further includes: Put the product into the vibrating plate through the feeding funnel; The products are arranged into the transmission track by using a frequency modulation vibration method through the vibration plate.

3. The product faceted screening method according to claim 1, characterized in that: The photographing mechanism comprises: Multi-axis stent; A plurality of cameras are arranged on the multi-axis bracket, each of the cameras is provided with a lens, and the cameras are electrically connected to the host computer; A plurality of light sources, each corresponding to each lens, and each light source is disposed at the corresponding lens; The light source controller is electrically connected to the light source and the host computer respectively.

4. The product faceted screening method according to claim 1, characterized in that: The camera is provided with a first data interface and a second data interface, and the camera is electrically connected to the host computer via the first data interface. The blowing module is provided with a solenoid valve, and the camera is electrically connected to the solenoid valve via the second data interface.

5. The product faceted screening method according to claim 1, characterized in that: The AI ​​model is trained through the following steps: Collect sample pictures; Classify and label the product features in the sample image; Deep learning training is performed on the sample images after classification and annotation to obtain the AI ​​model.

6. The product faceted screening method according to claim 1, characterized in that: When the product is transported along the transport track, acquiring the image sent by the photographing mechanism includes: Determining the image processing time of the host computer; determining a photographing frequency of the photographing mechanism according to the image processing time; When the product is transported along the transport track, the product is photographed by the photographing mechanism according to the photographing frequency to obtain an image of the product.

7. The product faceted screening method according to claim 1, characterized in that: The method of blowing the product to the return track by the air blowing module so that the product returns to the feeding mechanism through the return track includes: Obtaining the weight of the product and the transport speed of the transport track; Adjusting the opening of the solenoid valve of the blowing module according to the weight of the product and the transmission speed of the transmission track; The solenoid valve is opened to the opening degree, so that the blowing module blows the product to the return track, and the product returns to the loading mechanism through the return track.

8. A sorting device, characterized in that: The sorting device includes a loading mechanism, a transmission track, a blowing module, a return track and a photographing mechanism, wherein the loading mechanism is used to load products and transmit the products in sequence along the transmission track, the blowing module is arranged on one side of the transmission track, and the return track is arranged on the other side of the transmission track, the return track is opposite to the blowing module, and the return track is connected to the loading mechanism, the lens of the photographing mechanism is located above the transmission track, and a positioning mark is provided on the transmission track at a position corresponding to the blowing module; the sorting device also includes a host computer, and the host computer is used to implement the product facet screening method as described in any one of claims 1 to 7.

9. An electronic device, characterized in that: It includes at least one control processor and a memory for communicating with the at least one control processor; the memory stores instructions that can be executed by the at least one control processor, and the instructions are executed by the at least one control processor to enable the at least one control processor to execute the product faceted screening method as described in any one of claims 1 to 7.

10. A storage medium, characterized in that: The storage medium stores computer-executable instructions, and the computer-executable instructions are used to enable a computer to execute the product faceted screening method according to any one of claims 1 to 7.

Images