Article placement apparatus control method, device, equipment, medium and program product

Abstract

Embodiments of the present disclosure disclose an article placement device control method, device, equipment, medium and program product. A specific embodiment of the method comprises: inputting an article display image into a general object detection model to obtain a set of image coordinate information; determining a set of article images; inputting each article image into a first article recognition model; inputting each article image into a second article recognition model; for each first article recognition information, in response to determining that a probability value is greater than a first preset threshold, determining an included article number as an article recognition result; in response to determining that the probability value is less than or equal to the first preset threshold: selecting target second article recognition information; in response to an included similarity being greater than a second preset threshold, determining the included article number as the article recognition result; in response to determining that the similarity is less than or equal to the second preset threshold, controlling an article placement device to perform an article restocking operation. The embodiment improves the accuracy of article recognition and article placement.

Description

Technical Field

[0001] The embodiments disclosed herein relate to the field of computer technology, specifically to methods, apparatus, devices, media, and program products for controlling the placement of items. Background Technology

[0002] With the development of computer technology and the establishment of intelligent warehousing systems, automatic identification of items can be achieved. Currently, the common method for placing items is to identify them using a unified image recognition model and then arrange them according to the recognition results.

[0003] However, the inventors discovered that when placing items in the above manner, the following technical problems often arise:

[0004] First, the lack of multiple recognition models for object identification resulted in a low accuracy rate for object identification, which in turn led to a high error rate in the placement of objects.

[0005] Second, the lack of an image recognition model for identifying similar items resulted in a low accuracy rate for identifying similar items, which in turn led to a high error rate in the placement of items.

[0006] Third, the lack of an image recognition model specifically designed for new product identification resulted in a low accuracy rate for new product identification, which in turn led to a high error rate in the placement of items.

[0007] The information disclosed in this background section is only intended to enhance the understanding of the background of the inventive concept, and therefore may contain information that does not constitute prior art known to those skilled in the art. Summary of the Invention

[0008] The summary portion of this disclosure is intended to provide a brief overview of the concepts, which will be described in detail in the detailed description portion. This summary portion is not intended to identify key or essential features of the claimed technical solutions, nor is it intended to limit the scope of the claimed technical solutions.

[0009] Some embodiments of this disclosure provide methods, apparatuses, electronic devices, computer-readable media, and program products for controlling items placement equipment to solve one or more of the technical problems mentioned in the background section above.

[0010] In a first aspect, some embodiments of this disclosure provide a method for controlling an item placement device. The method includes: acquiring an item display image, wherein the item display image shows various items; inputting the item display image into a pre-trained general object detection model to obtain image coordinate information corresponding to each item as an image coordinate information set; determining an item image set based on the image coordinate information set; inputting each item image in the item image set into a pre-trained first item recognition model to obtain a first item recognition information set, wherein the first item recognition information in the first item recognition information set corresponds to an item image in the item image set, and the first item recognition information in the first item recognition information set includes an item number and a probability value corresponding to the item number; inputting each item image in the item image set into a pre-trained second item recognition model to obtain a second item recognition information set, wherein the second item recognition information in the second item recognition information set corresponds to an item image in the item image set, and the second item recognition information in the second item recognition information set includes an item number and a similarity value corresponding to at least one target item image. At least one target item image corresponds to the item number; for each first item identification information in the first item identification information set, in response to determining that the probability value included in the first item identification information is greater than a first preset threshold, the item number included in the first item identification information is determined as the item identification result; for each first item identification information in the first item identification information set, in response to determining that the probability value included in the first item identification information is less than or equal to the first preset threshold, the following steps are performed: the item image corresponding to the first item identification information is determined as the target item image; second item identification information corresponding to the target item image is selected from the second item identification information set as the target second item identification information; in response to determining that the similarity included in the target second item identification information is greater than a second preset threshold, the item number included in the target second item identification information is determined as the item identification result; in response to determining that the similarity included in the target second item identification information is less than or equal to the second preset threshold, the associated item placement device is controlled to perform an item addition operation corresponding to the target item, wherein the target item is the item corresponding to the item number included in the target second item identification information.

[0011] Secondly, some embodiments of this disclosure provide a control device for an item placement device. The device includes: an acquisition unit configured to acquire an item display image, wherein the item display image shows various items; a first input unit configured to input the item display image into a pre-trained general object detection model to obtain image coordinate information corresponding to each item as an image coordinate information set; a first determination unit configured to determine an item image set based on the image coordinate information set; a second input unit configured to input each item image in the item image set into a pre-trained first item recognition model to obtain a first item recognition information set, wherein the first item recognition information in the first item recognition information set corresponds to an item image in the item image set, and the first item recognition information in the first item recognition information set includes an item number and a probability value corresponding to the item number; and a third input unit configured to input each item image in the item image set into a pre-trained second item recognition model to obtain a second item recognition information set, wherein the second item recognition information in the second item recognition information set corresponds to an item image in the item image set, and the second item recognition information in the second item recognition information set includes an item number and a probability value corresponding to the item number. The similarity of at least one target item image, wherein the at least one target item image corresponds to the item number; a second determining unit, configured to, for each first item identification information in the first item identification information set, in response to determining that the probability value included in the first item identification information is greater than a first preset threshold, determine the item number included in the first item identification information as an item identification result; an execution unit, configured to, for each first item identification information in the first item identification information set, in response to determining that the probability value included in the first item identification information is less than or equal to the first preset threshold, execute the following steps: determine the item image corresponding to the first item identification information as a target item image; select second item identification information corresponding to the target item image from the second item identification information set as target second item identification information; in response to determining that the similarity included in the target second item identification information is greater than a second preset threshold, determine the item number included in the target second item identification information as an item identification result; in response to determining that the similarity included in the target second item identification information is less than or equal to the second preset threshold, control an associated item placement device to perform an item addition operation corresponding to the target item, wherein the target item is the item corresponding to the item number included in the target second item identification information.

[0012] Thirdly, some embodiments of this disclosure provide an electronic device, including: one or more processors; and a storage device having one or more programs stored thereon, wherein when the one or more programs are executed by the one or more processors, the one or more processors implement the method described in any implementation of the first aspect above.

[0013] Fourthly, some embodiments of this disclosure provide a computer-readable medium having a computer program stored thereon, wherein the program, when executed by a processor, implements the method described in any of the implementations of the first aspect above.

[0014] Fifthly, some embodiments of this disclosure provide a computer program product, including a computer program that, when executed by a processor, implements the method described in any of the implementations of the first aspect above.

[0015] The above-described embodiments of this disclosure have the following beneficial effects: the item placement device control method of some embodiments of this disclosure improves the accuracy of item recognition and reduces the error rate of item placement. Specifically, the reason for the low accuracy of item recognition and the high error rate of item placement is that multiple recognition models are not used to recognize the items, resulting in a low accuracy of item recognition and thus a high error rate of item placement. Based on this, the item placement device control method of some embodiments of this disclosure first acquires an item display image. The item display image shows each item. Thus, an item display image including each item can be obtained. Then, the item display image is input into a pre-trained general object detection model to obtain the image coordinate information corresponding to each item as an image coordinate information set. Thus, coordinate information representing the image position can be obtained, which can be used to determine the item image. Afterwards, an item image set is determined based on the image coordinate information set. Thus, item images of each item can be obtained. Next, each item image in the item image set is input into a pre-trained first item recognition model to obtain a first item recognition information set. The first item identification information in the aforementioned first item identification information set corresponds to the item images in the aforementioned item image set. The first item identification information in the aforementioned first item identification information set includes an item number and a probability value corresponding to that item number. Thus, the identified item number and the probability value used to determine whether the item to be identified is an existing item can be obtained. Then, each item image in the aforementioned item image set is input into a pre-trained second item identification model to obtain a second item identification information set. The second item identification information in the aforementioned second item identification information set corresponds to the item images in the aforementioned item image set. The second item identification information in the aforementioned second item identification information set includes an item number and a similarity score corresponding to at least one target item image. The at least one target item image corresponds to the item number. Thus, the identified item number and the similarity score used to determine whether the item to be identified is a new item can be obtained. Subsequently, for each piece of first item identification information in the aforementioned first item identification information set, in response to determining that the probability value included in the first item identification information is greater than a first preset threshold, the item number included in the first item identification information is determined as the item identification result. Therefore, when the probability value included in the first item identification information meets the determination condition of the probability value, the item number included in the first item identification information can be directly determined as the item identification result, so that the item corresponding to the item identification result can represent the identified item.Next, for each piece of first item identification information in the aforementioned first item identification information set, in response to determining that the probability value included in the aforementioned first item identification information is less than or equal to the aforementioned first preset threshold, the following steps are performed: First, the item image corresponding to the aforementioned first item identification information is determined as the target item image. Second, the second item identification information corresponding to the aforementioned target item image is selected from the aforementioned second item identification information set as the target second item identification information. Third, in response to determining that the similarity included in the aforementioned target second item identification information is greater than the second preset threshold, the item number included in the aforementioned target second item identification information is determined as the item identification result. Thus, when the probability value included in the first item identification information does not meet the probability value determination condition, further judgment can be made through the similarity in the second item identification information. Thus, when the similarity meets the similarity determination condition, the item number included in the target second item identification information is determined as the item identification result. Consequently, the item corresponding to the item identification result can represent the identified item. Finally, in response to determining that the similarity included in the aforementioned target second item identification information is less than or equal to the aforementioned second preset threshold, the associated item placement device is controlled to perform the item addition operation corresponding to the target item. The target item mentioned above refers to the item corresponding to the item number included in the second item identification information. Therefore, when the similarity does not meet the similarity judgment condition, it can be determined that the target item does not exist, i.e., the target item is a new product. Thus, the new product can be added. Also, because the item identification information identified by the first item identification model can identify items that meet the probability value judgment condition, when the probability value does not meet the judgment condition, items that meet the similarity judgment condition can be further identified based on the item identification information identified by the second item identification model. When the similarity does not meet the judgment condition, the item can be determined to be a new product. This improves the accuracy of item identification, thereby reducing the error rate of item placement. Attached Figure Description

[0016] The above and other features, advantages, and aspects of the embodiments of this disclosure will become more apparent from the accompanying drawings and the following detailed description. Throughout the drawings, the same or similar reference numerals denote the same or similar elements. It should be understood that the drawings are schematic, and elements are not necessarily drawn to scale.

[0017] Figure 1 This is a flowchart of some embodiments of the item placement device control method according to the present disclosure;

[0018] Figure 2 These are schematic diagrams of some embodiments of the item placement equipment control device according to the present disclosure;

[0019] Figure 3This is a schematic diagram of the structure of an electronic device suitable for implementing some embodiments of the present disclosure. Detailed Implementation

[0020] Embodiments of this disclosure will now be described in more detail with reference to the accompanying drawings. While some embodiments of this disclosure are shown in the drawings, it should be understood that this disclosure can be implemented in various forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided to provide a more thorough and complete understanding of this disclosure. It should be understood that the accompanying drawings and embodiments of this disclosure are for illustrative purposes only and are not intended to limit the scope of protection of this disclosure.

[0021] It should also be noted that, for ease of description, only the parts relevant to the invention are shown in the accompanying drawings. Unless otherwise specified, the embodiments and features described in this disclosure can be combined with each other.

[0022] It should be noted that the concepts of "first" and "second" mentioned in this disclosure are used only to distinguish different devices, modules or units, and are not used to limit the order of functions performed by these devices, modules or units or their interdependencies.

[0023] It should be noted that the terms "a" and "a plurality of" used in this disclosure are illustrative rather than restrictive, and those skilled in the art should understand that, unless otherwise expressly indicated in the context, they should be understood as "one or more".

[0024] The names of messages or information exchanged between multiple devices in the embodiments of this disclosure are for illustrative purposes only and are not intended to limit the scope of such messages or information.

[0025] This disclosure will now be described in detail with reference to the accompanying drawings and embodiments.

[0026] Figure 1 A flow 100 of some embodiments of a method for controlling an item placement device according to the present disclosure is shown. The method for controlling an item placement device includes the following steps:

[0027] Step 101: Obtain the image of the item display.

[0028] In some embodiments, the executing entity (e.g., a computing device) of the item placement device control method can acquire item display images from an image acquisition device via a wired or wireless connection. The item display images show various items. These images can be images captured by the image acquisition device. The image acquisition device can be a device with a shooting function. For example, the image acquisition device can include, but is not limited to, cameras, computers, and mobile phones. It should be noted that the wireless connection methods can include, but are not limited to, 3G / 4G connections, WiFi connections, Bluetooth connections, WiMAX connections, Zigbee connections, UWB (ultra-wideband) connections, and other currently known or future known wireless connection methods.

[0029] Step 102: Input the images of the displayed items into a pre-trained general object detection model to obtain the image coordinate information of each item as a set of image coordinate information.

[0030] In some embodiments, the execution entity can input the displayed item images into a pre-trained general object detection model to obtain image coordinate information corresponding to each item. The general object detection model can be a neural network model that takes the displayed item images as input and outputs the corresponding image coordinate information. For example, the neural network model can be the YOLO network (you only look once). The image coordinate information can be information representing the coordinates of the item images. Specifically, the image coordinate information can be (x, y, w, h). x can be the horizontal coordinate of the image center. y can be the vertical coordinate of the image center. w can be the image length. h can be the image width. As an example, the image coordinate information can be (0.436262, 0.474010, 0.383663, 0.178218).

[0031] Step 103: Determine the set of item images based on the set of image coordinate information.

[0032] In some embodiments, the execution entity may determine the set of item images based on the set of image coordinate information.

[0033] In some optional implementations of certain embodiments, for each image coordinate information in the aforementioned image coordinate information set, the execution entity may perform the following steps:

[0034] The first step is to determine the image position corresponding to the aforementioned image coordinate information, including the image center coordinates. The image center coordinates can be the location of the center point of the image. These image center coordinates characterize the position of the object image corresponding to the aforementioned image coordinate information. For example, the image center coordinates could be (0.436262, 0.474010). In practice, the executing entity can determine the image position corresponding to the aforementioned image coordinate information using the image center coordinates.

[0035] The second step is to determine the image size corresponding to the aforementioned image coordinate information, based on the image length and image width. The image size represents the size of the object image corresponding to the aforementioned image coordinate information. In practice, the executing entity can combine the image length and image width to obtain the image size corresponding to the aforementioned image coordinate information. For example, the image size could be (0.383663, 0.178218).

[0036] The third step is to determine the item image corresponding to the aforementioned image coordinate information based on the image position and size. The item image can be the image corresponding to the aforementioned image position and size. In practice, the executing entity can use the image center coordinates as the center and crop a rectangular image region of the aforementioned image size as the item image.

[0037] Step 104: Input each item image in the item image set into the pre-trained first item recognition model to obtain the first item recognition information set.

[0038] In some embodiments, the execution entity may input each item image in the item image set into a pre-trained first item recognition model to obtain a first item recognition information set. The first item recognition information in the first item recognition information set corresponds to the item images in the item image set. The first item recognition information in the first item recognition information set includes an item number and a probability value corresponding to that item number. The probability value can be the probability that the item corresponding to the item image to be recognized belongs to the item corresponding to the item number. The first item recognition model can be a neural network model that takes item images as input and outputs the first item recognition information corresponding to those item images. For example, the neural network model can be a TransFG model (Transformer, self-attention model).

[0039] Optionally, the first item recognition model described above can be trained in the following way:

[0040] The first step is to obtain a first sample set. The first sample in the first sample set includes first sample item images and corresponding first sample item recognition information. The first sample item recognition information can be a sample label corresponding to the first sample item image. The first sample item recognition information can include an item number and a probability value corresponding to that item number. It should be noted that the execution entity for training the first item recognition model can be the aforementioned execution entity or other computing devices.

[0041] The second step involves performing the following first training step based on the first sample set:

[0042] The first training step involves inputting the first sample item images of at least one first sample from the first sample set into an initial first item recognition model to obtain the first item recognition information corresponding to each of the at least one first sample. The initial first item recognition model is an initial item recognition model that can obtain first item recognition information based on item images. This initial item recognition model can be the item recognition model to be trained.

[0043] The second training step involves comparing the first item identification information corresponding to each of the at least one first sample with the corresponding first sample item identification information. This comparison can be a comparison of the accuracy of the first item identification information corresponding to each of the at least one first sample with the accuracy of the corresponding first sample item identification information.

[0044] The third training step involves determining whether the initial first item recognition model has reached a preset optimization objective based on the comparison results. Here, the optimization objective can be that the loss function value of the initial first item recognition model to be trained is less than a first preset threshold. The loss function can be the cross-entropy loss function. The first preset threshold can be 0.1.

[0045] The fourth training step is to determine the initial first item recognition model as the first item recognition model after training, in response to the determination that the initial first item recognition model has achieved the above optimization objective.

[0046] In some optional implementations of certain embodiments, the step of training the first item recognition model described above may further include:

[0047] Third, in response to the determination that the initial first item recognition model has not achieved the above optimization objective, the network parameters of the initial first item recognition model are adjusted, and an unused first sample is used to form a first sample set. The adjusted initial first item recognition model is then used as the initial first item recognition model, and the above first training step is executed again. As an example, the back propagation algorithm (BP algorithm) and gradient descent method (such as mini-batch gradient descent algorithm) can be used to adjust the network parameters of the above first item recognition model.

[0048] The first to third steps and related content described above constitute an inventive point of this disclosure, solving the second technical problem mentioned in the background art: "The failure to use an image recognition model for similar object recognition leads to a low accuracy rate in identifying similar objects, which in turn leads to a high error rate in object placement." The factors leading to low accuracy in identifying similar objects and a high error rate in object placement are often as follows: the failure to use an image recognition model for similar object recognition leads to a low accuracy rate in identifying similar objects, which in turn leads to a high error rate in object placement. Solving these factors can reduce the error rate in object placement. To achieve this effect, firstly, a first sample set is obtained. The first sample in the first sample set includes a first sample object image and first sample object recognition information corresponding to the first sample object image. This obtains data that can be used to train the initial first object recognition model. Secondly, based on the first sample set, the following first training step is performed: the first sample object images of at least one first sample in the first sample set are input into the initial first object recognition model to obtain the first object recognition information corresponding to each of the at least one first sample. The first item identification information corresponding to each of the at least one first sample is compared with the corresponding first sample item identification information. Based on the comparison result, it is determined whether the initial first item identification model has reached the preset optimization objective. In response to determining that the initial first item identification model has reached the optimization objective, the initial first item identification model is determined as the trained first item identification model. Thus, model training is completed, and the comparison result determines whether the initial first item identification model has reached the preset optimization objective, and the initial first item identification model that has reached the optimization objective is determined as the first item identification model. This improves the accuracy of similar item identification. Finally, in response to determining that the initial first item identification model has not reached the optimization objective, the network parameters of the initial first item identification model are adjusted, and an unused first sample is used to form a first sample set. The adjusted initial first item identification model is used as the initial first item identification model, and the first training step is executed again. Therefore, because the initial first item identification model has not reached the optimization objective, the network parameters of the initial first item identification model can be continuously optimized, and the model can be trained by continuously using unused first samples to form a first sample set, which can be used to generate a better similar item identification model. Because it's based on the first sample set, the initial first item recognition model can be continuously trained, and its network parameters can be continuously optimized. This yields network parameters that allow the first item recognition model to achieve high accuracy in recognizing similar items, thereby improving the accuracy of similar item recognition and reducing the error rate in item placement.

[0049] Step 105: Input each item image in the item image set into the pre-trained second item recognition model to obtain the second item recognition information set.

[0050] In some embodiments, the execution entity may input each item image in the item image set into a pre-trained second item recognition model to obtain a second item recognition information set. The second item recognition information in the second item recognition information set corresponds to the item images in the item image set. The second item recognition information in the second item recognition information set includes an item number and a similarity score corresponding to at least one target item image. The at least one target item image corresponds to the item number. The item number corresponds to the target item image with the highest similarity score among the at least one target item images. Here, the target item image may be an image of an item corresponding to a pre-selected item number. The item number may be an encoded symbol that distinguishes the target item from any other item. For example, the item number may be SKU:10111948B. The second item recognition model may be a neural network model that takes item images as input and outputs the second item recognition information corresponding to the item images. For example, the neural network model may be a CNN (Convolutional Neural Network).

[0051] Optionally, the second item recognition model described above is trained in the following way:

[0052] The first step is to obtain a second sample set. This second sample set includes second sample item images and corresponding second sample item identification information. The second sample item identification information can be sample labels corresponding to the second sample item images. This information may include item numbers and similarity scores to at least one target item image. It should be noted that the entity training the second item recognition model can be the aforementioned execution entity or other computing devices.

[0053] The second step involves performing the following second training step based on the second sample set:

[0054] The first sub-step involves inputting the second sample item images of at least one second sample from the second sample set into an initial second item recognition model to obtain the second item recognition information corresponding to each of the at least one second sample. The initial second item recognition model is an initial item recognition model that can obtain second item recognition information based on item images. This initial item recognition model can be an item recognition model to be trained.

[0055] The second sub-step involves comparing the second item identification information corresponding to each of the at least one second sample with the corresponding second sample item identification information. This comparison can be a comparison of the accuracy of the second item identification information corresponding to each of the at least one second sample with the accuracy of the corresponding second sample item identification information.

[0056] The third sub-step involves determining, based on the comparison results, whether the initial second item recognition model has achieved the preset optimization objective. Here, the optimization objective can be that the loss function value of the initial second item recognition model to be trained is less than a second preset threshold. The loss function can be a contrastive loss function. The second preset threshold can be 0.08.

[0057] The fourth sub-step is to determine the initial second item recognition model as the trained second item recognition model in response to the determination that the initial second item recognition model has achieved the above optimization objective.

[0058] In some optional implementations of certain embodiments, the step of training the second item recognition model described above may further include:

[0059] The third step, in response to the determination that the initial second item recognition model did not achieve the above optimization objective, involves adjusting the network parameters of the initial second item recognition model, forming a second sample set using unused second samples, and using the adjusted initial second item recognition model as the initial second item recognition model, then executing the above second training step again. The back propagation algorithm (BP algorithm) and gradient descent methods (such as mini-batch gradient descent) can be used to adjust the network parameters of the second item recognition model.

[0060] The first to third steps and related content described above constitute an inventive point of this disclosure, solving the third technical problem mentioned in the background art: "The failure to use an image recognition model specifically for new product recognition leads to a low accuracy rate in new product recognition, which in turn results in a high error rate in item placement." The factors leading to low accuracy in new product recognition and a high error rate in item placement are often as follows: the failure to use an image recognition model specifically for new product recognition leads to a low accuracy rate in new product recognition, which in turn results in a high error rate in item placement. Solving these factors can reduce the error rate in item placement. To achieve this effect, firstly, a second sample set is obtained. The second sample in the second sample set includes second sample item images and corresponding second sample item recognition information. This obtains data that can be used to train the initial second item recognition model. Secondly, based on the second sample set, the following second training step is performed: the second sample item images of at least one second sample in the second sample set are input into the initial second item recognition model to obtain the second item recognition information corresponding to each of the at least one second sample. The second item identification information corresponding to each of the at least one second sample is compared with the corresponding second sample item identification information. Based on the comparison result, it is determined whether the initial second item identification model has reached the preset optimization objective. In response to determining that the initial second item identification model has reached the optimization objective, the initial second item identification model is identified as the trained second item identification model. Thus, model training is completed. The comparison result determines whether the initial second item identification model has reached the preset optimization objective, and the initial second item identification model that has reached the optimization objective is identified as the second item identification model, improving the accuracy of new item identification. Finally, in response to determining that the initial second item identification model has not reached the optimization objective, the network parameters of the initial second item identification model are adjusted, and a second sample set is formed using unused second samples. The adjusted initial second item identification model is used as the initial second item identification model, and the second training step is executed again. Therefore, because the initial second item identification model has not reached the optimization objective, the network parameters of the initial second item identification model can be continuously optimized, and the model can be trained by continuously using unused second samples to form a second sample set, which can be used to generate a better new item identification model. Because it is based on the second sample set, the initial second item recognition model can be continuously trained and the network parameters of the initial second item recognition model can be continuously optimized. This results in network parameters that enable the second item recognition model to achieve a high accuracy in item recognition, thereby improving the accuracy of new item recognition and reducing the error rate of item placement.

[0061] Step 106: For each piece of first item identification information in the first item identification information set, in response to determining that the probability value included in the first item identification information is greater than a first preset threshold, the item number included in the first item identification information is determined as the item identification result.

[0062] In some embodiments, for each piece of first item identification information in the aforementioned first item identification information set, in response to determining that the probability value included in the first item identification information is greater than a first preset threshold, the executing entity may determine the item number included in the first item identification information as the item identification result. The first preset threshold may be a pre-set probability value. For example, the first preset threshold may be 0.7.

[0063] In some optional implementations of certain embodiments, the aforementioned execution entity may further perform the following steps:

[0064] The first step is to determine the item number included in the first item identification information as the first item number.

[0065] The second step is to determine the inventory quantity of the first item corresponding to the first item number. In practice, the executing entity can obtain the inventory quantity of the item with the first item number from the warehouse system terminal that stores inventory data as the first item inventory quantity. This inventory quantity can be the remaining inventory quantity of the item. The warehouse system terminal can be a terminal that stores the first item inventory quantity.

[0066] Third, in response to determining that the inventory quantity of the first item is greater than the third preset threshold, the associated sound playback device is controlled to play a first inventory notification message. This first inventory notification message can be information reminding the executing entity of the inventory quantity of the first item. For example, the first inventory notification message could be "The remaining inventory quantity of item 001 is 150, the inventory is sufficient, and no restocking is needed at this time." The third preset threshold can be 100. "150" can be the inventory quantity of the first item. "001" can be the item number. The sound playback device can be a device used to play sound. For example, the sound playback device can include, but is not limited to, an amplifier, speakers, a multimedia control console, and a digital mixing console.

[0067] Fourth, in response to determining that the inventory quantity of the first item is less than or equal to the third preset threshold, the sound playback device is controlled to play a second inventory notification message. This second inventory notification message can be a reminder to the executing entity to pay attention to the inventory quantity of the first item and to replenish it in a timely manner. For example, the second inventory notification message could be: "The remaining inventory quantity of item 002 is 30. Inventory is insufficient; please replenish promptly." Here, "30" represents the inventory quantity of the first item, and "002" represents the item number.

[0068] Step 107: For each piece of first item identification information in the first item identification information set, in response to determining that the probability value included in the first item identification information is less than or equal to a first preset threshold, the following steps are performed:

[0069] Step 1071: Determine the item image corresponding to the first item identification information as the target item image.

[0070] Step 1072: Select the second item identification information of the corresponding target item image from the second item identification information set as the target second item identification information.

[0071] Step 1073: In response to determining that the similarity of the target second item identification information is greater than the second preset threshold, the item number included in the target second item identification information is determined as the item identification result.

[0072] In some embodiments, in response to determining that the similarity included in the target second item identification information is greater than a second preset threshold, the executing entity may determine the item number included in the target second item identification information as the item identification result. The second preset threshold may be a pre-set similarity value. For example, the second preset threshold may be 0.75.

[0073] In some optional implementations of certain embodiments, the aforementioned execution entity may further perform the following steps:

[0074] The first step is to determine the item number included in the above-mentioned target second item identification information as the second item number.

[0075] The second step is to determine the inventory quantity of the second item corresponding to the aforementioned second item number. In practice, the executing entity can obtain the inventory quantity of the item with the aforementioned second item number from the warehouse system terminal that stores inventory data as the inventory quantity of the second item. This inventory quantity can be the remaining inventory quantity. The warehouse system terminal can be a terminal that stores the aforementioned inventory quantity of the second item.

[0076] Third, in response to determining that the inventory quantity of the second item is greater than a fourth preset threshold, the associated sound playback device is controlled to play a third inventory prompt message. The fourth preset threshold can be 60. The third inventory prompt message can be information reminding the executing entity to pay attention to the item type and inventory quantity. The item type can be an item type representing the length of time the target item has been in storage. For example, the item type can include, but is not limited to, new products and similar products. The executing entity can determine the item type based on the relationship between the probability value and the first preset threshold. Specifically, the executing entity can determine items with a probability value greater than the first preset threshold as similar products, and items with a probability value less than or equal to the first preset threshold as new products. For example, the third inventory prompt message could be "Item 003 may be a new product, remaining inventory quantity is 90, please pay attention to sales." "003" can be the item number. "90" can be the inventory quantity of the second item.

[0077] Fourth, in response to determining that the inventory quantity of the second item is less than or equal to the fourth preset threshold, the sound playback device is controlled to play a fourth inventory notification message. This fourth inventory notification message can be a reminder to the executing entity regarding the item type and inventory adequacy. For example, the fourth inventory notification message could be, "Item 004 may be new; remaining inventory quantity is 30. Please replenish stock promptly." Here, "004" can be the item number, and "30" can be the inventory quantity of the second item.

[0078] Step 1074: In response to determining that the similarity of the target second item identification information is less than or equal to a second preset threshold, control the associated item placement device to perform the item addition operation for the corresponding target item.

[0079] In some embodiments, in response to determining that the similarity included in the target second item identification information is less than or equal to the second preset threshold, the executing entity can control the associated item placement device to perform an item addition operation for the corresponding target item. The target item is the item corresponding to the item number included in the target second item identification information.

[0080] In practice, in response to determining that the similarity of the second target item identification information is less than or equal to the second preset threshold, the executing entity can control the associated item placement device to perform the corresponding target item new item operation through the following steps:

[0081] The first step is to obtain the first item information corresponding to the aforementioned target item. This first item information can be information specific to the target item. This first item information may include, but is not limited to, the country code, producer code, item code, shelf number code, and inspection code. In practice, the executing entity can obtain the first item information corresponding to the target item from a warehouse system terminal that stores inventory data. This warehouse system terminal can be a terminal that stores the first item information of the target item.

[0082] The second step involves generating item barcode information based on the aforementioned first item information. This item barcode information can represent information indicating whether the target item can be stored or removed from the warehouse. For example, the item barcode information could be 6901234 56789 564 2. Here, 690 can represent the country code, 1234 can represent the producer code, 56789 can represent the item code, 564 can represent the shelf number code, and 2 can represent the check digit. In practice, the executing entity can use barcode generator software to generate the item barcode information corresponding to the aforementioned first item information. This barcode generator software can be provided by any software vendor.

[0083] The third step is to determine the inventory quantity of the target item based on its item number. In practice, the executing entity can obtain the inventory quantity corresponding to the item number from the warehouse system terminal that stores inventory data.

[0084] The fourth step involves controlling the item placement device to perform a new item replenishment operation corresponding to the target item, based on the item barcode information and the inventory quantity. In practice, firstly, the executing entity can determine a candidate value by multiplying the inventory quantity by a preset value. The preset value can be any value greater than 0 and less than 1. Then, the rounded value of the candidate value can be determined as the quantity of the target item to be replenished. Here, the rounded value can be either rounded up or rounded down from the candidate value. Secondly, the executing entity can determine the location indicated by the shelf number code included in the item barcode information as the placement location of the target item. Finally, the executing entity can control the item placement device to place the target quantity of the target item into the determined placement location. Here, the item placement device can be a device used for placing items. For example, the item placement device can include, but is not limited to, intelligent robotic arms and intelligent robots.

[0085] Optionally, for each target item corresponding to the above-mentioned item update operation, the execution entity may also perform the following steps:

[0086] The first step is to scan the barcode of the target item to obtain the item barcode information. In practice, the executing entity can use Radio Frequency Identification (RFID) technology to scan the barcode of the target item to obtain the item barcode information.

[0087] The second step involves scanning the shelf barcode corresponding to the target item on the shelf where it is placed to obtain the shelf barcode information. Here, the shelf barcode corresponding to the target item can be the barcode located at the position where the target item is placed on the shelf. In practice, the executing entity can use radio frequency identification (RFID) technology to scan the shelf barcode corresponding to the target item on the shelf where it is placed to obtain the shelf barcode information.

[0088] The third step is to control the sound playback device to play a notification message indicating that the items have been placed, in response to the determination that the barcode information of the above-mentioned items matches the barcode information of the above-mentioned shelves.

[0089] In practice, in response to determining that the barcode information of the aforementioned item matches the barcode information of the aforementioned shelf, the executing entity can control the sound playback device to play a notification message indicating that the item has been placed. This notification message can be a message indicating that the item has been placed. For example, the notification message could be "Item 006 has been placed." Here, the matching can refer to the matching of the content of the barcode information of the aforementioned item with the content of the barcode information of the aforementioned shelf. Specifically, the matching can mean that the barcode information of the aforementioned item is identical to the country code, producer code, item code, and shelf number code included in the barcode information of the aforementioned shelf.

[0090] The above-described embodiments of this disclosure have the following beneficial effects: the item placement device control method of some embodiments of this disclosure improves the accuracy of item recognition and reduces the error rate of item placement. Specifically, the reason for the low accuracy of item recognition and the high error rate of item placement is that multiple recognition models are not used to recognize the items, resulting in a low accuracy of item recognition and thus a high error rate of item placement. Based on this, the item placement device control method of some embodiments of this disclosure first acquires an item display image. The item display image shows each item. Thus, an item display image including each item can be obtained. Then, the item display image is input into a pre-trained general object detection model to obtain the image coordinate information corresponding to each item as an image coordinate information set. Thus, coordinate information representing the image position can be obtained, which can be used to determine the item image. Afterwards, an item image set is determined based on the image coordinate information set. Thus, item images of each item can be obtained. Next, each item image in the item image set is input into a pre-trained first item recognition model to obtain a first item recognition information set. The first item identification information in the aforementioned first item identification information set corresponds to the item images in the aforementioned item image set. The first item identification information in the aforementioned first item identification information set includes an item number and a probability value corresponding to that item number. Thus, the identified item number and the probability value used to determine whether the item to be identified is an existing item can be obtained. Then, each item image in the aforementioned item image set is input into a pre-trained second item identification model to obtain a second item identification information set. The second item identification information in the aforementioned second item identification information set corresponds to the item images in the aforementioned item image set. The second item identification information in the aforementioned second item identification information set includes an item number and a similarity score corresponding to at least one target item image. The at least one target item image corresponds to the item number. Thus, the identified item number and the similarity score used to determine whether the item to be identified is a new item can be obtained. Subsequently, for each piece of first item identification information in the aforementioned first item identification information set, in response to determining that the probability value included in the first item identification information is greater than a first preset threshold, the item number included in the first item identification information is determined as the item identification result. Therefore, when the probability value included in the first item identification information meets the determination condition of the probability value, the item number included in the first item identification information can be directly determined as the item identification result, so that the item corresponding to the item identification result can represent the identified item.Next, for each piece of first item identification information in the aforementioned first item identification information set, in response to determining that the probability value included in the aforementioned first item identification information is less than or equal to the aforementioned first preset threshold, the following steps are performed: First, the item image corresponding to the aforementioned first item identification information is determined as the target item image. Second, the second item identification information corresponding to the aforementioned target item image is selected from the aforementioned second item identification information set as the target second item identification information. Third, in response to determining that the similarity included in the aforementioned target second item identification information is greater than the second preset threshold, the item number included in the aforementioned target second item identification information is determined as the item identification result. Thus, when the probability value included in the first item identification information does not meet the probability value determination condition, further judgment can be made through the similarity in the second item identification information. Thus, when the similarity meets the similarity determination condition, the item number included in the target second item identification information is determined as the item identification result. Consequently, the item corresponding to the item identification result can represent the identified item. Finally, in response to determining that the similarity included in the aforementioned target second item identification information is less than or equal to the aforementioned second preset threshold, the associated item placement device is controlled to perform the item addition operation corresponding to the target item. The target item mentioned above refers to the item corresponding to the item number included in the second item identification information. Therefore, when the similarity does not meet the similarity judgment condition, it can be determined that the target item does not exist, i.e., the target item is a new product. Thus, the new product can be added. Also, because the item identification information identified by the first item identification model can identify items that meet the probability value judgment condition, when the probability value does not meet the judgment condition, items that meet the similarity judgment condition can be further identified based on the item identification information identified by the second item identification model. When the similarity does not meet the judgment condition, the item can be determined to be a new product. This improves the accuracy of item identification, thereby reducing the error rate of item placement.

[0091] Further reference Figure 2 As an implementation of the methods shown in the above figures, this disclosure provides some embodiments of an item placement device control device, which are similar to... Figure 1 Corresponding to the method embodiments shown, the device can be specifically applied to various electronic devices.

[0092] like Figure 2As shown, the item placement device control device 200 in some embodiments includes: an acquisition unit 201, a first input unit 202, a first determination unit 203, a second input unit 204, a third input unit 205, a second determination unit 206, and an execution unit 207. The acquisition unit 201 is configured to acquire an image of an item display, wherein each item is displayed in the image. The first input unit 202 is configured to input the image of the item display into a pre-trained general object detection model to obtain image coordinate information corresponding to each item as an image coordinate information set. The first determination unit 203 is configured to determine an item image set based on the image coordinate information set. The second input unit 204 is configured to input each item image in the item image set into a pre-trained first item recognition model to obtain a first item recognition information set, wherein the first item recognition information in the first item recognition information set corresponds to an item image in the item image set, and includes an item number and a probability value corresponding to the item number. The third input unit 205 is configured to input each item image in the item image set into a pre-trained second item recognition model to obtain a second item recognition information set, wherein the second item recognition information in the second item recognition information set corresponds to an item image in the item image set, and includes an item number and a similarity score corresponding to at least one target item image. The first determination unit 206 is configured to, for each first item identification information in the first item identification information set, determine the item number included in the first item identification information as an item identification result in response to determining that the probability value included in the first item identification information is greater than a first preset threshold; the execution unit 207 is configured to, for each first item identification information in the first item identification information set, perform the following steps in response to determining that the probability value included in the first item identification information is less than or equal to the first preset threshold: determine the item image corresponding to the first item identification information as a target item image; select second item identification information corresponding to the target item image from the second item identification information set as target second item identification information; determine the item number included in the target second item identification information as an item identification result in response to determining that the similarity included in the target second item identification information is greater than a second preset threshold; and control the associated item placement device to perform an item addition operation corresponding to the target item in response to determining that the similarity included in the target second item identification information is less than or equal to the second preset threshold, wherein the target item is the item corresponding to the item number included in the target second item identification information.

[0093] It is understandable that the units described in the item placement equipment control device 200 are related to the reference. Figure 1 The steps in the described method correspond to each other. Therefore, the operations, features, and beneficial effects described above for the method also apply to the device 200 and the units contained therein, and will not be repeated here.

[0094] The following is for reference. Figure 3 It shows a schematic diagram of the structure of an electronic device 300 suitable for implementing some embodiments of the present disclosure. Figure 3 The electronic device shown is merely an example and should not be construed as limiting the functionality and scope of the embodiments of this disclosure.

[0095] like Figure 3 As shown, the electronic device 300 may include a processing device (e.g., a computing device) 301, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 302 or a program loaded from a storage device 308 into a random access memory (RAM) 303. The RAM 303 also stores various programs and data required for the operation of the electronic device 300. The processing device 301, ROM 302, and RAM 303 are interconnected via a bus 304. An input / output (I / O) interface 305 is also connected to the bus 304.

[0096] Typically, the following devices can be connected to I / O interface 305: input devices 306 including, for example, touchscreens, touchpads, keyboards, mice, cameras, microphones, accelerometers, gyroscopes, etc.; output devices 307 including, for example, liquid crystal displays (LCDs), speakers, vibrators, etc.; storage devices 308 including, for example, magnetic tapes, hard disks, etc.; and communication devices 309. Communication device 309 allows electronic device 300 to communicate wirelessly or wiredly with other devices to exchange data. Although Figure 3 An electronic device 300 with various devices is shown; however, it should be understood that it is not required to implement or possess all of the devices shown. More or fewer devices may be implemented or possessed alternatively. Figure 3 Each box shown can represent a device or multiple devices as needed.

[0097] In particular, according to some embodiments of this disclosure, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, some embodiments of this disclosure include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via communication device 309, or installed from storage device 308, or installed from ROM 302. When the computer program is executed by processing device 301, it performs the functions defined in the methods of some embodiments of this disclosure.

[0098] It should be noted that, in some embodiments of this disclosure, the computer-readable medium may be a computer-readable signal medium or a computer-readable storage medium, or any combination thereof. A computer-readable storage medium may be, for example,—but not limited to—an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of a computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In some embodiments of this disclosure, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. In some embodiments of this disclosure, a computer-readable signal medium may include a data signal propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such propagated data signals may take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A computer-readable signal medium can be any computer-readable medium other than a computer-readable storage medium, which can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device. The program code contained on the computer-readable medium can be transmitted using any suitable medium, including but not limited to: wires, optical fibers, RF (radio frequency), etc., or any suitable combination thereof.

[0099] In some implementations, clients and servers can communicate using any currently known or future-developed network protocol such as HTTP (Hypertext Transfer Protocol) and can interconnect with digital data communication (e.g., communication networks) of any form or medium. Examples of communication networks include local area networks (“LANs”), wide area networks (“WANs”), the Internet (e.g., the Internet of Things), and end-to-end networks (e.g., ad hoc end-to-end networks), as well as any currently known or future-developed networks.

[0100] The aforementioned computer-readable medium may be included in the aforementioned electronic device; or it may exist independently and not assembled into the electronic device. The aforementioned computer-readable medium carries one or more programs, which, when executed by the electronic device, cause the electronic device to: acquire an image of an item display, wherein the image of the item display shows various items; input the image of the item display into a pre-trained general object detection model to obtain image coordinate information corresponding to each of the items as an image coordinate information set; determine an item image set based on the image coordinate information set; input each item image in the item image set into a pre-trained first item recognition model to obtain a first item recognition information set, wherein the first item recognition information in the first item recognition information set corresponds to an item image in the item image set, and the first item recognition information in the first item recognition information set includes an item number and a probability value corresponding to the item number; input each item image in the item image set into a pre-trained second item recognition model to obtain a second item recognition information set, wherein the second item recognition information in the second item recognition information set corresponds to an item image in the item image set, and the second item recognition information in the second item recognition information set includes an item number and an image corresponding to at least one target item. The similarity of the images, wherein at least one target item image corresponds to the item number; for each first item identification information in the first item identification information set, in response to determining that the probability value included in the first item identification information is greater than a first preset threshold, the item number included in the first item identification information is determined as the item identification result; for each first item identification information in the first item identification information set, in response to determining that the probability value included in the first item identification information is less than or equal to the first preset threshold, the following steps are performed: the item image corresponding to the first item identification information is determined as the target item image; the second item identification information corresponding to the target item image is selected from the second item identification information set as the target second item identification information; in response to determining that the similarity included in the target second item identification information is greater than a second preset threshold, the item number included in the target second item identification information is determined as the item identification result; in response to determining that the similarity included in the target second item identification information is less than or equal to the second preset threshold, the associated item placement device is controlled to perform an item addition operation corresponding to the target item, wherein the target item is the item corresponding to the item number included in the target second item identification information.

[0101] Computer program code for performing operations of some embodiments of this disclosure can be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, and C++, and conventional procedural programming languages ​​such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0102] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this disclosure. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0103] The units described in some embodiments of this disclosure can be implemented in software or hardware. The described units can also be housed in a processor; for example, a processor may be described as including an acquisition unit, a first input unit, a first determination unit, a second input unit, a third input unit, a second determination unit, and an execution unit. The names of these units do not necessarily limit the specific unit; for example, the acquisition unit may also be described as a "unit for acquiring images of an item display".

[0104] The functions described above in this document can be performed, at least in part, by one or more hardware logic components. For example, exemplary types of hardware logic components that can be used, without limitation, include: Field Programmable Gate Arrays (FPGAs), Application-Specific Integrated Circuits (ASICs), Application Standard Products (ASSPs), System-on-Chip (SoCs), Complex Programmable Logic Devices (CPLDs), and so on.

[0105] Some embodiments of this disclosure also provide a computer program product, including a computer program that, when executed by a processor, implements any of the above-described methods for controlling an item placement device.

[0106] The above description is merely a selection of preferred embodiments of this disclosure and an explanation of the technical principles employed. Those skilled in the art should understand that the scope of the invention involved in the embodiments of this disclosure is not limited to technical solutions formed by specific combinations of the above-described technical features, but should also cover other technical solutions formed by arbitrary combinations of the above-described technical features or their equivalents without departing from the above-described inventive concept. For example, technical solutions formed by substituting the above-described features with (but not limited to) technical features with similar functions disclosed in the embodiments of this disclosure.

Claims

1. A method for controlling an article placement device, comprising: obtaining an article display image, wherein each article is displayed in the article display image; inputting the article display image into a pre-trained general object detection model to obtain each image coordinate information corresponding to each article as an image coordinate information set; determining an article image set according to the image coordinate information set; inputting each article image in the article image set into a pre-trained first article recognition model to obtain a first article recognition information set, wherein first article recognition information in the first article recognition information set corresponds to an article image in the article image set, the first article recognition information in the first article recognition information set comprises an article number and a probability value corresponding to the article number, and the first article recognition model is a neural network model taking the article image as input and taking the first article recognition information corresponding to the article image as output; inputting each article image in the article image set into a pre-trained second article recognition model to obtain a second article recognition information set, wherein second article recognition information in the second article recognition information set corresponds to an article image in the article image set, the second article recognition information in the second article recognition information set comprises an article number and a similarity corresponding to at least one target article image, the at least one target article image corresponds to the article number, the target article image is an image of an article corresponding to the article number and preselected, and the second article recognition model is a neural network model taking an article image as input and taking the second article recognition information corresponding to the article image as output; for each first article recognition information in the first article recognition information set, in response to determining that the probability value included in the first article recognition information is greater than a first preset threshold, determining the article number included in the first article recognition information as an article recognition result; for each first article recognition information in the first article recognition information set, in response to determining that the probability value included in the first article recognition information is less than or equal to the first preset threshold, performing the following steps: determining the article image corresponding to the first article recognition information as a target article image; selecting second article recognition information corresponding to the target article image from the second article recognition information set as target second article recognition information; in response to determining that the similarity included in the target second article recognition information is greater than a second preset threshold, determining the article number included in the target second article recognition information as an article recognition result; and in response to determining that the similarity included in the target second article recognition information is less than or equal to the second preset threshold, controlling an associated article placement device to perform an article restocking operation corresponding to a target article, wherein the target article is an article corresponding to the article number included in the target second article recognition information. The image coordinate information in the image coordinate information set comprises an image center coordinate, an image length, and an image width; and The determining of the article image set according to the image coordinate information set comprises: ​ ​ ​ ​ ​ ​ ​ ​ ​ 2. The method of claim 1, wherein, ​ ​ For each image coordinate information in the set of image coordinate information, the following steps are performed: According to the image center coordinates included in the image coordinate information, the image position corresponding to the image coordinate information is determined; According to the image length and image width included in the image coordinate information, the image size corresponding to the image coordinate information is determined; According to the image position and the image size, the image of the corresponding article corresponding to the image coordinate information is determined.

3. The method of claim 1, wherein, The method further comprises: acquiring first article information corresponding to the target article; generating article barcode information based on the first article information; determining the inventory quantity of the target article according to the article number of the target article; controlling the article placement device to perform the article restocking operation corresponding to the target article according to the article barcode information and the inventory quantity.

4. The method of claim 3, wherein, Before the control of the article placement device to perform the article restocking operation corresponding to the target article, the method further comprises: For each target article corresponding to the article restocking operation, the following steps are performed: scanning the article barcode of the target article to obtain article barcode information; scanning the shelf barcode corresponding to the target article on the shelf on which the target article is placed to obtain shelf barcode information; in response to determining that the article barcode information matches the shelf barcode information, controlling the sound playing device to play article placement completion prompt information.

5. The method of claim 1, wherein, After the method further comprises: determining the article number included in the first article identification information as the first article number; determining the first article inventory quantity corresponding to the first article number according to the first article number; in response to determining that the first article inventory quantity is greater than a third preset threshold, controlling the associated sound playing device to play a first inventory prompt information; in response to determining that the first article inventory quantity is less than or equal to the third preset threshold, controlling the sound playing device to play a second inventory prompt information.

6. The method of claim 1, wherein, After the method further comprises: determining the article number included in the target second article identification information as the second article number; determining the second article inventory quantity corresponding to the second article number according to the second article number; in response to determining that the second article inventory quantity is greater than a fourth preset threshold, controlling the associated sound playing device to play a third inventory prompt information; in response to determining that the article inventory quantity is less than or equal to the fourth preset threshold, controlling the sound playing device to play a fourth inventory prompt information.

7. An article placement device control apparatus, comprising: An acquisition unit configured to acquire an article display image, wherein each article is displayed in the article display image; A first input unit configured to input the article display image into a pre-trained general object detection model to obtain, as an image coordinate information set, image coordinate information corresponding to each article; A first determination unit configured to determine, according to the image coordinate information set, an article image set; A second input unit configured to input each article image in the article image set into a pre-trained first article recognition model to obtain a first article recognition information set, wherein first article recognition information in the first article recognition information set corresponds to an article image in the article image set, and the first article recognition information in the first article recognition information set includes an article number and a probability value corresponding to the article number, the first article recognition model being a neural network model taking the article image as input and the first article recognition information corresponding to the article image as output; A third input unit configured to input each article image in the article image set into a pre-trained second article recognition model to obtain a second article recognition information set, wherein second article recognition information in the second article recognition information set corresponds to an article image in the article image set, and the second article recognition information in the second article recognition information set includes an article number and a similarity corresponding to at least one target article image, the at least one target article image corresponding to the article number, the target article image being a pre-selected image of an article corresponding to the article number, the second article recognition model being a neural network model taking an article image as input and the second article recognition information corresponding to the article image as output; A second determination unit configured to, for each first article recognition information in the first article recognition information set, determine, in response to determining that the probability value included in the first article recognition information is greater than a first preset threshold, the article number included in the first article recognition information as an article recognition result; An execution unit configured to, for each first article recognition information in the first article recognition information set, perform the following steps in response to determining that the probability value included in the first article recognition information is less than or equal to the first preset threshold: determining the article image corresponding to the first article recognition information as a target article image; selecting, from the second article recognition information set, second article recognition information corresponding to the target article image as target second article recognition information; in response to determining that the similarity included in the target second article recognition information is greater than a second preset threshold, determining the article number included in the target second article recognition information as an article recognition result; and in response to determining that the similarity included in the target second article recognition information is less than or equal to the second preset threshold, controlling an associated article placement device to perform an article restocking operation on a target article corresponding to the article number included in the target second article recognition information.

8. An electronic device, comprising: one or more processors; a storage device having stored thereon one or more programs; when the one or more programs are executed by the one or more processors, cause the one or more processors to carry out the method of any one of claims 1-6.

9. A computer readable medium having stored thereon a computer program, wherein, the program which when executed by the processor carries out the method of any one of claims 1-6.

10. A computer program product comprising a computer program which when executed by a processor carries out the method of any one of claims 1-6.

Images