Sliding rail type counting system, method and device and counting equipment

Abstract

The invention discloses a sliding rail type counting system, method and device and counting equipment. The sliding rail type counting system comprises a sliding rail, a sliding block capable of moving along the sliding rail, image acquisition equipment fixed to the sliding block, control equipment and the counting equipment. The control equipment controls the sliding block to move to the calibration point along the sliding rail and controls the image acquisition equipment to acquire an image containing the goods to be counted; the image acquisition device sends an acquired image to the counting device; and the counting device inputs the image into a pre-trained cargo counting model, and determines the number of cargos in the image based on the cargo counting model. According to the automatic counting scheme, on one hand, the cargo counting efficiency is improved, the labor cost is reduced, and on the other hand, the problems that due to the fatigue of workers, counting errors are likely to be generated, the counting accuracy is poor, and large economic losses are caused to enterprises are solved.

Description

Technical Field

[0001] The present application relates to the field of intelligent warehousing and logistics technology, and in particular to a slide rail counting system, method, device and counting equipment. Background Art

[0002] The quantity of goods must be calculated in every link of production, transportation and sales. It is especially important to deal with the counting differences raised by customers in the sales link. At present, the number of goods is still counted manually. The problems of manual counting are that, on the one hand, it is time-consuming and labor-intensive, and the labor cost is high; on the other hand, counting errors are easily generated due to the fatigue of the workers themselves, and the counting accuracy is poor, causing great economic losses to the company. Summary of the invention

[0003] The present application provides a slide rail counting system, method, device and counting equipment to solve the problems of the counting methods in the prior art being time-consuming and labor-intensive, having high labor costs and poor counting accuracy.

[0004] In a first aspect, the present application provides a slide rail type counting system, the system comprising: a slide rail, a slider movable along the slide rail, an image acquisition device fixed on the slider, a control device and a counting device;

[0005] The control device is connected to the slider and the image acquisition device respectively, and the counting device is connected to the image acquisition device;

[0006] The control device is used to control the slider to move along the slide rail to a calibration point, and control the image acquisition device to acquire an image containing the goods to be counted;

[0007] The image acquisition device is used to send the image to the counting device;

[0008] The counting device is used to input the image into a pre-trained goods counting model, and determine the quantity of goods in the image based on the goods counting model.

[0009] In a second aspect, the present application provides a slide rail counting method, the method comprising:

[0010] Receiving an image sent by an image acquisition device, wherein the image is an image of goods to be counted, which is acquired by controlling the slide block to move to a calibration point along a slide rail and controlling the image acquisition device to acquire the image;

[0011] The image is input into a pre-trained goods counting model, and the quantity of goods in the image is determined based on the goods counting model.

[0012] In a third aspect, the present application provides a slide rail counting device, the device comprising:

[0013] A receiving module, used for receiving an image sent by an image acquisition device, wherein the image is an image of goods to be counted, which is acquired by controlling the slide block to move to a calibration point along the slide rail and controlling the image acquisition device to acquire the image;

[0014] The determination module is used to input the image into a pre-trained goods counting model, and determine the quantity of goods in the image based on the goods counting model.

[0015] In a fourth aspect, the present application provides a counting device, including a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory communicate with each other via the communication bus;

[0016] Memory, used to store computer programs;

[0017] The processor is used to implement the method steps when executing the program stored in the memory.

[0018] In a fifth aspect, the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method steps described are implemented.

[0019] The present application provides a slide rail counting system, method, apparatus and counting device, the system comprising: a slide rail, a slider movable along the slide rail, an image acquisition device fixed on the slider, a control device and a counting device; the control device is respectively connected to the slider and the image acquisition device, and the counting device is connected to the image acquisition device; the control device is used to control the slider to move to a calibration point along the slide rail, and control the image acquisition device to acquire an image containing goods to be counted; the image acquisition device is used to send the image to the counting device; the counting device is used to input the image into a pre-trained goods counting model, and determine the quantity of goods in the image based on the goods counting model.

[0020] The above technical solution has the following advantages or beneficial effects:

[0021] The present application provides a slide rail counting system, including a slide rail, a slider movable along the slide rail, an image acquisition device fixed on the slider, a control device and a counting device. The control device controls the slider to move along the slide rail to a calibration point, and controls the image acquisition device to acquire an image containing goods to be counted; the image acquisition device sends the acquired image to the counting device; the counting device inputs the image into a pre-trained goods counting model, and determines the number of goods in the image based on the goods counting model. The present application provides an automated counting solution, which, on the one hand, improves the efficiency of goods counting and reduces labor costs, and on the other hand, avoids the problem of counting errors that are prone to occur due to workers' own fatigue, poor counting accuracy, and large economic losses to the enterprise. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0023] Figure 1 Structural diagram of the slide rail counting system provided for this application;

[0024] Figure 2 Schematic diagram of the slide rail counting process provided for this application;

[0025] Figure 3 A schematic diagram of the structure of the slide rail counting device provided in this application;

[0026] Figure 4 Schematic diagram of the counting device structure provided for this application. DETAILED DESCRIPTION

[0027] In order to make the purpose and implementation method of the present application clearer, the exemplary implementation method of the present application will be clearly and completely described below in conjunction with the drawings in the exemplary embodiments of the present application. Obviously, the described exemplary embodiments are only part of the embodiments of the present application, rather than all the embodiments.

[0028] It should be noted that the brief description of terms in this application is only for the convenience of understanding the embodiments described below, and is not intended to limit the embodiments of this application. Unless otherwise specified, these terms should be understood according to their common and usual meanings.

[0029] The terms "first", "second", "third", etc. in the specification and claims of this application and the above drawings are used to distinguish similar or similar objects or entities, and do not necessarily mean to limit a specific order or sequence, unless otherwise noted. It should be understood that the terms used in this way can be interchangeable under appropriate circumstances.

[0030] The terms "comprises," "comprising," and "having," and any variations thereof, are intended to cover but not exclude inclusion, for example, a product or device comprising a list of components is not necessarily limited to all the components expressly listed but may include other components not expressly listed or inherent to such product or device.

[0031] The term "module" refers to any known or later developed hardware, software, firmware, artificial intelligence, fuzzy logic, or combination of hardware and / or software code that is capable of performing the functions associated with that element.

[0032] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit it. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present application.

[0033] For the convenience of explanation, the above description has been made in conjunction with specific embodiments. However, the above exemplary discussion is not intended to be exhaustive or limit the embodiments to the specific forms disclosed above. Based on the above teachings, various modifications and variations can be obtained. The selection and description of the above embodiments are to better explain the principles and practical applications, so that those skilled in the art can better use the embodiments and various different variations of the embodiments suitable for specific use considerations.

[0034] Figure 1 This is a structural diagram of a slide rail type counting system provided in the present application, wherein the system comprises: a slide rail 11, a slider 12 movable along the slide rail, an image acquisition device 13 fixed on the slider, a control device 14 and a counting device 15;

[0035] The control device 14 is connected to the slider 12 and the image acquisition device 13 respectively, and the counting device 15 is connected to the image acquisition device 13;

[0036] The control device 14 is used to control the slider 12 to move along the slide rail 11 to a calibration point, and control the image acquisition device 13 to acquire an image containing goods to be counted;

[0037] The image acquisition device 13 is used to send the image to the counting device 15;

[0038] The counting device 15 is used to input the image into a pre-trained goods counting model, and determine the quantity of goods in the image based on the goods counting model.

[0039] The control device can be a PC, a tablet computer, a server, etc., and the counting device can be a PC, a tablet computer, a server, etc. The control device and the counting device can be the same device or different devices.

[0040] The control device is connected to the slider and the image acquisition device by wire or wireless, and the counting device is connected to the image acquisition device by wire or wireless. The image acquisition device can be a camera, a video camera, etc., and the image acquisition device shoots toward the goods to be counted.

[0041] The slide rail is pre-determined with calibration points, which are determined according to the stacking height of the goods to be counted. If the stacking height of the goods to be counted is high, the position of the calibration point can be relatively high; if the stacking height of the goods to be counted is low, the position of the calibration point can be relatively low.

[0042] The control device controls the slider to move along the slide rail to the calibration point. After reaching the calibration point, the slider is controlled to be fixed, and then the image acquisition device is controlled to acquire an image containing the goods to be counted. After the image acquisition device acquires the image containing the goods to be counted, the image is sent to the counting device. A pre-trained goods counting model is deployed in the counting device, and the image is input into the pre-trained goods counting model, and the number of goods in the image is determined based on the goods counting model.

[0043] It should be noted that the goods to be counted in the present application can be wooden boards or other goods. Taking template counting as an example, the process of determining the number of goods in an image based on a goods counting model is described. Through the model training stage: the collected images are labeled accordingly, image preprocessing is performed, and the complete long wooden board image is cut into small wooden boards of random sizes according to specifications; the YOLOv7 target detection network is used to train the image and the corresponding label to obtain the corresponding network model; the wooden board detection and counting stage: the image to be detected is preprocessed, and the complete long wooden board image is cut into small wooden boards of equal sizes according to specifications, and input into the network model obtained in the model training stage, and network reasoning is performed to obtain the predicted target frame; the target frame obtained by filtering the NMS algorithm, the SORT target tracking algorithm is used to achieve corresponding matching of the target frames of the small wooden boards on the same layer, and then the target frames on the same layer are filtered and connected, and the number of layers of the target frame is calculated to obtain the number of wooden boards.

[0044] The present application provides a slide rail counting system, including a slide rail, a slider movable along the slide rail, an image acquisition device fixed on the slider, a control device and a counting device. The control device controls the slider to move along the slide rail to a calibration point, and controls the image acquisition device to acquire an image containing goods to be counted; the image acquisition device sends the acquired image to the counting device; the counting device inputs the image into a pre-trained goods counting model, and determines the number of goods in the image based on the goods counting model. The present application provides an automated counting solution, which, on the one hand, improves the efficiency of goods counting and reduces labor costs, and on the other hand, avoids the problem of counting errors that are prone to occur due to workers' own fatigue, poor counting accuracy, and large economic losses to the enterprise.

[0045] In the present application, the control device is used to obtain the stacking height of the goods to be counted, and determine the calibration point on the slide rail according to the stacking height, the field of view of the image acquisition device and the horizontal distance between the image acquisition device and the goods to be counted.

[0046] It should be noted that the calibration point determined on the slide rail can be one or more. If, based on the stacking height, the field of view of the image acquisition device, and the horizontal distance between the image acquisition device and the goods to be counted, the image captured by the image acquisition device at one calibration point can include the completely stacked goods, then only one calibration point can be determined. When the image captured by the image acquisition device at one calibration point cannot include the completely stacked goods, multiple calibration points need to be determined, and the images captured by the image acquisition device at multiple calibration points can include the completely stacked goods after image stitching and fusion.

[0047] The control device is used to control the slider to move along the slide rail to a plurality of calibration points, and to control the image acquisition device to acquire images containing goods to be counted at the plurality of calibration points respectively;

[0048] The image acquisition device is used to send the acquired multiple images to the counting device respectively;

[0049] The counting device is used to determine the quantity of goods based on the goods counting model and the multiple images.

[0050] In the present application, for the case of multiple calibration points, the control device controls the slider to move along the slide rail to multiple calibration points in a certain order, and controls the image acquisition device to capture images at each calibration point. Optionally, in the order from the top to the bottom of the slide rail, when the slider is controlled to move along the slide rail to the first calibration point, the slider is controlled to be fixed and the image acquisition device is controlled to capture images. Then the control device controls the slider to move along the slide rail to the second calibration point, and then controls the slider to be fixed, and controls the image acquisition device to capture images. Until the image acquisition device captures images at each preset calibration point. The image acquisition device can send the captured image to the counting device after each image is captured, or it can send multiple images to the counting device together after capturing images of all calibration points. The counting device determines the quantity of goods based on the goods counting model and multiple images.

[0051] Optionally, the counting device is used to input the multiple images into the cargo counting model respectively, and determine the sub-quantities of cargo in the multiple images based on the cargo counting model respectively; and determine the quantity of the cargo according to the sub-quantities of cargo in the multiple images.

[0052] In some calibration cases, if there are multiple calibration points and there is no overlap of goods in the images captured by the various calibration points, the multiple images can be respectively input into the goods counting model, and the sub-quantities of goods in the multiple images can be respectively determined based on the goods counting model, and then the sum of the sub-quantities of goods in the multiple images can be used as the determined quantity of goods.

[0053] Taking into account that in most cases there is overlap of goods in the images collected by each calibration point, in order to make the goods counting more accurate, the counting device is used to stitch and fuse the multiple images to obtain a fused image, input the fused image into the goods counting model, and determine the number of goods in the fused image based on the goods counting model.

[0054] Optionally, multiple images may be stitched and fused by an image stitching algorithm, including but not limited to an APAP method, an SPHP method, and a PT method. Multiple images are stitched and fused to obtain a fused image, which is then input into a goods counting model to determine the quantity of goods in the fused image based on the goods counting model.

[0055] In order to make the image stitching and fusion more accurate, the image acquisition device is used to respectively acquire images containing the goods to be counted at the multiple calibration points, and determine the corresponding relationship between the multiple images and the calibration points; and send the multiple images and the corresponding relationship between the multiple images and the calibration points to the counting device;

[0056] The counting device is used to sort the multiple images according to the correspondence between the multiple images and the calibration points and in the order of the calibration points, and to perform splicing and fusion processing on the multiple images according to the sorting result to obtain a fused image.

[0057] In the present application, the image acquisition device respectively acquires images containing the goods to be counted at multiple calibration points, and determines the correspondence between the multiple images and the calibration points. That is to say, the image acquisition device records the correspondence between the acquired images and the calibration points. For example, three calibration points are set, and the slide rails are calibration point 1, calibration point 2, and calibration point 3 from top to bottom. The image acquired at calibration point 1 records the correspondence between the image and calibration point 1, the image acquired at calibration point 2 records the correspondence between the image and calibration point 2, and the image acquired at calibration point 3 records the correspondence between the image and calibration point 3. The multiple images and the correspondence between the multiple images and the calibration points are sent to the counting device.

[0058] The counting device sorts the multiple images according to the correspondence between the multiple images and the calibration points and in the order of the calibration points, and splices and fuses the multiple images according to the sorting result to obtain a fused image. That is to say, the adjacent images are spliced ​​according to the adjacent order of the calibration points. Taking the slide rail as an example, which is calibration point 1, calibration point 2 and calibration point 3 from top to bottom, the images collected by calibration point 1 and calibration point 2 can be spliced ​​and fused first, and then the fusion result can be spliced ​​and fused with the image collected by calibration point 3 to obtain the final fused image. Alternatively, the images collected by calibration point 2 and calibration point 3 can be spliced ​​and fused first, and then the fusion result can be spliced ​​and fused with the image collected by calibration point 1 to obtain the final fused image. Compared with the random splicing and fusion method, image fusion in the order of calibration points can reduce the impact of image distortion, improve the accuracy of image splicing and fusion, and thus improve the accuracy of cargo counting.

[0059] The counting device is used to determine the images that are adjacent in order according to the ordering results of the multiple images, respectively identify the same reference objects in the images that are adjacent in order, and perform splicing and fusion processing on the images that are adjacent in order according to the same reference objects.

[0060] In the present application, when performing image stitching and fusion, after determining the images with adjacent order according to the sorting results of multiple images, the same reference objects in the images with adjacent order are identified. Among them, a pre-trained target recognition model can be deployed in the counting device, and the same reference objects in the images with adjacent order are identified based on the target recognition model. The reference object can be a shelf that is fixed in the scene, etc. In order to facilitate image stitching, fixed objects of regular shapes can be used as reference objects. According to the position information of the same reference objects in the images with adjacent order, the images with adjacent order are stitched and fused so that the same reference objects in the two images overlap to obtain a fused image.

[0061] The counting process is explained below using wooden board counting as an example.

[0062] Model training stage: the collected images are labeled accordingly, and image preprocessing is performed, and the complete long wooden board images are cut into small wooden boards of random sizes according to specifications; the YOLOv7 target detection network is used to train the images and corresponding labels to obtain the corresponding network model; wooden board detection and counting stage: the images to be detected are preprocessed, and the complete long wooden board images are cut into small wooden boards of equal sizes according to specifications, and input into the network model obtained in the model training stage, and network inference is performed to obtain the predicted target frame; the target frame obtained by filtering the NMS algorithm is matched with the target frame of the small wooden boards on the same layer by using the SORT target tracking algorithm, and then the target frames on the same layer are filtered and connected, and the number of layers of the target frame is calculated to obtain the number of wooden boards.

[0063] The steps of the model training stage include: image preprocessing: collect stacked wooden board images, place the camera in front of the stacked wooden boards to take pictures, obtain clear and complete real-life pictures of the stacked wooden boards, and mark each wooden board in the acquired image; evenly cut the collected wooden board images into small wooden strip images of a set width, and then perform data enhancement processing; YOLOv7 detection: after image preprocessing, input the cut small wooden strip images into the YOLOv7 network to calculate the loss of the predicted box and the true box, update the model weights, and train multiple times to obtain an effective network model; then through network reasoning, obtain a series of confidence, category and candidate box predictions; then use the non-maximum suppression algorithm as the post-processing algorithm of the YOLOv7 detection task to select the optimal bounding box from the candidate box set and suppress other candidate boxes.

[0064] In the YOLOv7 detection step, the non-maximum suppression algorithm is used to take the category confidence as the evaluation index of the optimal bounding box, the candidate boxes are arranged in descending order according to the category confidence, and the candidate box with the highest category confidence score is selected as the optimal bounding box. The calculation formula in the non-maximum suppression algorithm is as follows:

[0065]

[0066] Among them, si represents the category confidence, IoU is the intersection over union ratio, M represents the optimal bounding box, bi represents the candidate box, and Nt represents the manually set threshold.

[0067] The steps of the wooden board detection and counting stage include: SORT target tracking algorithm: the intersection-and-union ratio between each detection and all predicted bounding boxes of the existing target is used as a metric for the target relationship between the previous and next frames. After the YOLOv7 detection algorithm is used to obtain the detection frame coordinates and credibility corresponding to the small wooden board target in each frame image, the tracking operation of the small wooden board target on the same layer is performed, and the Kalman filter is used to predict the target position of the next frame image. The detection result of YOLOv7 is input into the Kalman filter to obtain the estimated value of the frame; the improved Hungarian algorithm is used for data association, and the intersection-and-union ratio of the detection frame and the prediction frame is used to determine the tracking of the cut small wooden boards on the same layer, and linear allocation is performed to associate the ID between frames; the obtained wooden board layer ID is counted to obtain the number of wooden boards in the stacked wooden boards.

[0068] In the steps of the SORT target tracking algorithm, the estimated value of the current frame image is calculated through a recursive estimation method after the estimated value of the previous frame image and the observed value of the current frame image are known. The Kalman filter algorithm is usually implemented through the state equation and the observation equation as follows: xk=Axk-1+Buk+wkZk=Hxk+vk, wherein xk is the control information of the filter system at time k; xk-1 is the control information of the filter system at time k-1; uk is the state control vector of the filter system; A is the state matrix of the filter system; B is the gain coefficient of the filter system; wk is the process noise of the filter system; Zk is the value measured by the filter system at time k; vk is the noise measured by the filter system, and H is the system constant of the filter system.

[0069] Figure 2 The schematic diagram of the slide rail counting process provided for this application includes the following steps:

[0070] S101: receiving an image sent by an image acquisition device, wherein the image is an image of goods to be counted, which is acquired by controlling a slider to move along a slide rail to a calibration point and controlling the image acquisition device to acquire the image;

[0071] S102: Input the image into a pre-trained goods counting model, and determine the quantity of goods in the image based on the goods counting model.

[0072] The slide rail counting method provided in the present application is applied to counting equipment.

[0073] The method comprises:

[0074] The multiple images are respectively input into the goods counting model, and the sub-quantities of goods in the multiple images are respectively determined based on the goods counting model; and the quantity of the goods is determined according to the sub-quantities of goods in the multiple images.

[0075] The method comprises:

[0076] The multiple images are spliced ​​and fused to obtain a fused image, the fused image is input into the goods counting model, and the quantity of goods in the fused image is determined based on the goods counting model.

[0077] The method comprises:

[0078] According to the correspondence between the multiple images and the calibration points, the multiple images are sorted in the order of the calibration points, and the multiple images are spliced ​​and fused according to the sorting result to obtain a fused image.

[0079] The method comprises:

[0080] According to the sorting results of the multiple images, images that are adjacent in sorting are determined, identical reference objects in the images that are adjacent in sorting are respectively identified, and stitching and fusing processing is performed on the images that are adjacent in sorting according to the identical reference objects.

[0081] The purpose of this application is to solve the problems of low efficiency and low counting accuracy of manual counting methods. The technical solution adopted is a slide rail counting system, including a linear slide rail, a slider, an image acquisition device, a control device and a counting device. The function of the linear slide rail is to control the movement of the image acquisition device installed on the slider; the image acquisition device acquires images of the target goods according to the movement of the slider on the linear slide rail; the control device controls the movement of the slider on the linear slide rail according to multiple calibration points set according to the actual height of the goods; the counting device recognizes and analyzes multiple images through a visual detection algorithm, accurately identifies the number of plates in the image, and merges the detected number of plates through an algorithm to achieve rapid counting of the number of goods.

[0082] In the workflow of the slide rail counting system, the plate goods to be inspected are first placed at the position to be inspected, and multiple calibration points are set for the goods according to their actual height. The image acquisition device loaded on the slider on the slide rail device is controlled by the control device to move, and images are collected from top to bottom according to the multiple calibration points. After the image acquisition work is completed, the slider on the slide rail device is reset to the initial position, and the number of plates in the image is detected by the visual algorithm in the counting device. Finally, the number of multiple images is combined through the algorithm to output the number of goods.

[0083] The slide rail counting solution provided by the present application improves the efficiency of logistics and warehousing: the slide rail counting device can quickly and accurately identify the number of plates, and can quickly complete the task of plate counting in a short time, thereby improving the efficiency of logistics and warehousing. Reduce labor costs: The traditional manual counting method requires a lot of manpower and time, and is prone to problems such as fatigue and mistakes, resulting in low efficiency. The slide rail counting device can replace manual operation, reduce labor costs, and improve work efficiency. Easy to install: The slide rail counting device has a simple structure, is easy to install, and its subsequent maintenance is also relatively simple. Low manufacturing cost: The slide rail counting device is different from other expensive mechanical equipment, has a lower cost, and can save manufacturing costs.

[0084] The specific installation implementation steps are as follows:

[0085] Install the slide rail device: First, you need to install the slide rail device and install the slide rail counting device to the designated position to facilitate counting work.

[0086] Install the camera (image acquisition device): Install the camera on the slider of the slide rail device, and configure the focal length, exposure and other shooting parameters to ensure the image acquisition effect.

[0087] Set calibration points: Set multiple calibration points according to the actual height of the goods. The camera can collect images of the goods one by one through the calibration points.

[0088] Deploy visual algorithm services to ensure real-time processing and analysis of visual data.

[0089] Debug the slide rail device: Check various parameters of the slide rail according to actual conditions to ensure that the slide rail device can capture images more accurately.

[0090] After the installation and debugging of the slide rail counting device is completed, it is ready for counting. When the goods to be processed move to the designated location, the image acquisition system controls the slide rail device to move the camera to the designated shooting point, and collects images from top to bottom according to the multiple calibration points set. After the goods image is collected, the slider on the slide rail device moves to the initial position, waiting for the visual algorithm system to detect and process the image. After the detection and processing is completed, the system outputs the number of goods. At this point, the workflow of the slide rail counting device ends, waiting for the next goods to be processed to repeat the above operation.

[0091] Figure 3 The schematic diagram of the structure of the slide rail counting device provided for this application includes:

[0092] The receiving module 31 is used to receive an image sent by an image acquisition device, wherein the image is an image of goods to be counted that is acquired by controlling the slider to move along the slide rail to a calibration point and controlling the image acquisition device to acquire the image;

[0093] The determination module 32 is used to input the image into a pre-trained goods counting model, and determine the quantity of goods in the image based on the goods counting model.

[0094] The present application also provides a counting device, such as Figure 4 As shown, it includes: a processor 301, a communication interface 302, a memory 303 and a communication bus 304, wherein the processor 301, the communication interface 302, and the memory 303 communicate with each other through the communication bus 304;

[0095] The memory 303 stores a computer program, and when the program is executed by the processor 301, the processor 301 executes any one of the above method steps.

[0096] The communication bus mentioned in the above counting device can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. The communication bus can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, only one thick line is used in the figure, but it does not mean that there is only one bus or one type of bus.

[0097] The communication interface 302 is used for communication between the counting device and other devices.

[0098] The memory may include a random access memory (RAM) or a non-volatile memory (NVM), such as at least one disk memory. Optionally, the memory may also be at least one storage device located away from the aforementioned processor.

[0099] The above-mentioned processor can be a general-purpose processor, including a central processing unit, a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application-specific integrated circuit, a field programmable gate array or other programmable logic device, a discrete gate or transistor logic device, a discrete hardware component, etc.

[0100] The present application also provides a computer storage readable storage medium, wherein the computer readable storage medium stores a computer program executable by a counting device, and when the program runs on the counting device, the counting device implements any of the above method steps when executing.

[0101] Although the preferred embodiments of the present application have been described, those skilled in the art may make other changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications falling within the scope of the present application.

[0102] Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the claims of the present application and their equivalents, the present application is also intended to include these modifications and variations.

Claims

1. A slide rail counting system, characterized in that: The system comprises: a slide rail, a slider movable along the slide rail, an image acquisition device fixed on the slider, a control device and a counting device; The control device is connected to the slider and the image acquisition device respectively, and the counting device is connected to the image acquisition device; The control device is used to control the slider to move along the slide rail to a calibration point, and control the image acquisition device to acquire an image containing the goods to be counted; The image acquisition device is used to send the image to the counting device; The counting device is used to input the image into a pre-trained goods counting model and determine the quantity of goods in the image based on the goods counting model.

2. The system according to claim 1, wherein The control device is used to obtain the stacking height of the goods to be counted, and determine the calibration point on the slide rail according to the stacking height, the field of view of the image acquisition device and the horizontal distance between the image acquisition device and the goods to be counted.

3. The system according to claim 1 or 2, characterized in that The control device is used to control the slider to move along the slide rail to a plurality of calibration points, and to control the image acquisition device to acquire images containing the goods to be counted at the plurality of calibration points respectively; The image acquisition device is used to send the acquired multiple images to the counting device respectively; The counting device is configured to determine the quantity of goods based on the goods counting model and the plurality of images.

4. The system according to claim 3, wherein: The counting device is configured to input the plurality of images into the cargo counting model respectively, determine sub-quantities of cargo in the plurality of images based on the cargo counting model respectively, and determine the quantity of cargo according to the sub-quantities of cargo in the plurality of images.

5. The system according to claim 3, wherein: The counting device is configured to perform splicing and fusing processing on the multiple images to obtain a fused image, input the fused image into the goods counting model, and determine the quantity of goods in the fused image based on the goods counting model.

6. The system according to claim 5, wherein: The image acquisition device is configured to respectively acquire images containing the goods to be counted at the plurality of calibration points, and determine corresponding relationships between the plurality of images and the calibration points; and transmit the plurality of images and the corresponding relationships between the plurality of images and the calibration points to the counting device; The counting device is used to sort the multiple images in the order of the calibration points according to the correspondence between the multiple images and the calibration points, and to perform splicing and fusion processing on the multiple images according to the sorting result to obtain a fused image.

7. The system according to claim 6, wherein: The counting device is used to determine adjacent images according to the sorting results of the multiple images, respectively identify the same reference objects in the adjacent images, and perform splicing and fusion processing on the adjacent images according to the same reference objects.

8. A slide rail counting method, characterized in that: The method comprises: Receiving an image sent by an image acquisition device, wherein the image is an image of goods to be counted, which is acquired by controlling a slider to move along a slide rail to a calibration point and controlling the image acquisition device to acquire the image; The image is input into a pre-trained goods counting model, and the number of goods in the image is determined based on the goods counting model.

9. A slide rail counting device, characterized in that: The device comprises: a receiving module, configured to receive an image sent by an image acquisition device, wherein the image is an image of goods to be counted, which is acquired by controlling a slider to move along a slide rail to a calibration point and controlling the image acquisition device to acquire the image; The determination module is configured to input the image into a pre-trained goods counting model and determine the quantity of goods in the image based on the goods counting model.

10. A counting device, characterized in that: It includes a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory communicate with each other via the communication bus; Memory for storing computer programs; The processor is configured to implement the method steps described in claim 8 when executing the program stored in the memory.

Images