No-load container positioning method and related product thereof

Abstract

The invention discloses a no-load container positioning method and a related product thereof. The method comprises the following steps: acquiring worker positioning information and a storage area image of a loading container; identifying a no-load container attribute set existing in the storage area image; determining a container classification label corresponding to each loading container and container position information corresponding to each loading container based on the storage area image; and determining a target no-load container based on the no-load container attribute set, the container classification label corresponding to each loading container, the container position information corresponding to each loading container and the worker positioning information. According to the invention, the no-load container in the plant can be identified and positioned, the communication cost and labor intensity of workers are reduced, the operation efficiency of a supply chain is improved, and the efficient and accurate management level of the supply chain is improved.

Description

Technical Field

[0001] The present application generally relates to the field of logistics management technology. More specifically, the present application relates to an empty container positioning method and related products. Background Art

[0002] In modern supply chain management, logistics and manufacturing environments, the organization, management and transportation of various materials place particularly strict requirements on boxes. Usually, multiple types of boxes are stored and used in factories to meet the loading needs of different materials. In this environment, finding empty boxes of the required type in a timely manner is crucial to improving production efficiency and management level.

[0003] At present, the management of empty boxes in factories mainly relies on manual operation. Workers complete related tasks by visual observation and manual handling of empty boxes. However, due to the complex factory environment, poor information transmission, and high communication costs between workers, a large amount of manpower consumption and time waste is caused, which is not only inefficient but also prone to errors, affecting the operational efficiency of the entire supply chain. Although some factories have begun to use technologies such as barcodes and RFID to assist in the management of boxes, these technologies all rely on manual scanning and registration, which not only increases the workload of workers, but also has problems such as identification errors and equipment failures, making it difficult to meet the needs of efficient and accurate management.

[0004] In view of this, there is an urgent need to provide an empty container positioning method so that the empty containers in the factory can be identified and positioned, the communication costs and labor intensity of workers can be reduced, the operational efficiency of the supply chain can be improved, and the efficient and accurate management level of the supply chain can be enhanced. Summary of the invention

[0005] In order to at least solve one or more of the technical problems mentioned above, the present application proposes an empty container positioning method and related products in multiple aspects. The empty container positioning method can identify and locate empty containers in a factory, reduce the communication cost and labor intensity of workers, improve the operational efficiency of the supply chain, and enhance the efficient and accurate management level of the supply chain.

[0006] In a first aspect, the present application provides a method for locating an empty container, comprising: acquiring worker positioning information and a storage area image of loaded containers; identifying a set of empty container attributes present in the storage area image; determining, based on the storage area image, a container classification label corresponding to each loaded container and container position information corresponding to each loaded container; and determining a target empty container based on the empty container attribute set, the container classification label corresponding to each loaded container, the container position information corresponding to each loaded container, and the worker positioning information.

[0007] In some embodiments, determining a target empty container based on a set of empty container attributes, a container classification label corresponding to each loading container, container position information corresponding to each loading container, and worker positioning information includes: determining each candidate empty container based on the set of empty container attributes and the container classification label corresponding to each loading container; determining a candidate container position corresponding to each candidate empty container based on the container position information corresponding to each loading container; and determining the target empty container based on the candidate container position corresponding to each candidate empty container and the worker positioning information.

[0008] In some embodiments, determining each candidate empty container based on the empty container attribute set and the container classification label corresponding to each loaded container includes: if the container classification label of the current loaded container matches the empty container attribute element in the empty container attribute set, determining the current loaded container as a candidate empty container.

[0009] In some embodiments, determining the target empty container based on the candidate container position and worker positioning information corresponding to each candidate empty container includes: determining each movement trajectory between the worker and each candidate empty container based on the candidate container position and worker positioning information corresponding to each candidate empty container; and determining the target empty container based on the moving distance of each movement trajectory.

[0010] In some embodiments, identifying the empty container attribute set present in the storage area image includes: identifying the empty container attribute set present in the storage area image by using an attribute recognition model.

[0011] In some embodiments, the attribute recognition model includes a backbone recognition network and an attribute classifier; wherein, identifying the empty container attribute set present in the storage area image through the attribute recognition model includes: extracting features of the storage area image through the backbone recognition network to obtain loaded container attribute features; and classifying the loaded container attribute features through the attribute classifier to obtain the empty container attribute set.

[0012] In some embodiments, determining the container classification label corresponding to each loading container and the container position information corresponding to each loading container based on the storage area image includes: determining the container classification label corresponding to each loading container and the container position information corresponding to each loading container through a container detection model.

[0013] In some embodiments, after determining the target empty container based on a set of empty container attributes, a container classification label corresponding to each loaded container, a container position information corresponding to each loaded container, and worker positioning information, the method further includes: sending the container position information of the target empty container and the moving trajectory corresponding to the target empty container to the worker task terminal.

[0014] In a second aspect, the present application provides a device for locating empty containers, comprising: a memory; and at least one processor, configured to: obtain worker positioning information and a storage area image of loaded containers; identify a set of empty container attributes present in the storage area image; determine, based on the storage area image, a container classification label corresponding to each loaded container and container position information corresponding to each loaded container; and determine a target empty container based on the empty container attribute set, the container classification label corresponding to each loaded container, the container position information corresponding to each loaded container, and the worker positioning information.

[0015] In a third aspect, the present application provides a non-volatile machine-readable medium having program code stored thereon for empty container positioning. When the program code is executed by at least one processor, the code guides the execution operation of the at least one processor, and the program code includes: obtaining worker positioning information and a storage area image of loaded containers; identifying a set of empty container attributes present in the storage area image; determining, based on the storage area image, a container classification label corresponding to each loaded container and container position information corresponding to each loaded container; and determining a target empty container based on the empty container attribute set, the container classification label corresponding to each loaded container, the container position information corresponding to each loaded container, and the worker positioning information.

[0016] The technical solution provided by this application may have the following beneficial effects:

[0017] The empty container positioning method and related products provided by the present application can determine which empty containers with certain attributes exist in the storage area image by obtaining the worker positioning information and the storage area image of the loading container, and then identifying the empty container attribute set in the storage area image. Furthermore, the container classification label corresponding to each loading container and the container position information corresponding to each loading container are determined based on the storage area image, and then the target empty container is determined based on the empty container attribute set, the container classification label corresponding to each loading container, the container position information corresponding to each loading container, and the worker positioning information. Thus, the container classification label corresponding to each loading container and the empty container attribute set are used to accurately screen out each required empty container, realize automatic, real-time and high-precision empty container detection and positioning, and the container position information corresponding to each loading container and the worker positioning information can be used to determine the empty container closest to the worker, which is conducive to reducing the communication cost and labor intensity of the workers and improving the work efficiency of the workers.

[0018] In general, this application can identify and locate empty containers in the factory, reduce the communication costs and labor intensity of workers, improve the operational efficiency of the supply chain, and enhance the efficient and accurate management level of the supply chain. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] By reading the detailed description below with reference to the accompanying drawings, the above and other purposes, features and advantages of the exemplary embodiments of the present application will become easy to understand. In the accompanying drawings, several embodiments of the present application are shown in an exemplary and non-limiting manner, and the same or corresponding reference numerals represent the same or corresponding parts, wherein:

[0020] Figure 1 An exemplary flow chart showing an empty container positioning method according to some embodiments of the present application is shown;

[0021] Figure 2 An exemplary flow chart showing an empty container positioning method according to other embodiments of the present application is shown;

[0022] Figure 3 An exemplary flow chart showing an empty container positioning method according to some other embodiments of the present application is shown;

[0023] Figure 4 A block diagram showing the hardware configuration of an empty container locating device 400 that can implement the empty container locating method according to an embodiment of the present application. DETAILED DESCRIPTION

[0024] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, rather than all of the embodiments. For the simplicity and clarity of the description, the figure marks may be repeated in the drawings to indicate corresponding or similar elements when deemed appropriate. In addition, the present application sets forth many specific details in order to provide a thorough understanding of the embodiments described herein. However, it will be understood by those of ordinary skill in the art that the embodiments described herein may be practiced without these specific details. In other cases, well-known methods, processes, and components are not described in detail to avoid blurring the embodiments described herein. Moreover, the description should not be regarded as limiting the scope of the embodiments described herein. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without making creative work are within the scope of protection of the present application.

[0025] It should be understood that the possible terms "first" or "second" etc. in the claims, specifications and drawings disclosed in this application are used to distinguish different objects, rather than to describe a specific order. The terms "include" and "comprise" used in the specification and claims of this application indicate the presence of the described features, wholes, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components and / or their collections.

[0026] It should also be understood that the terms used in this application specification are only for the purpose of describing specific embodiments and are not intended to limit the present application. As used in this application specification and claims, unless the context clearly indicates otherwise, the singular forms of "a", "an" and "the" are intended to include plural forms. It should also be further understood that the term "and / or" used in this application specification and claims refers to any combination of one or more of the associated listed items and all possible combinations, and includes these combinations.

[0027] As used in this specification and claims, the term "if" may be interpreted as "when" or "upon" or "in response to determining" or "in response to detecting," depending on the context. Similarly, the phrase "if it is determined" or "if [described condition or event] is detected" may be interpreted as meaning "upon determination" or "in response to determining" or "upon detection of [described condition or event]" or "in response to detecting [described condition or event]," depending on the context.

[0028] Due to the complex factory environment, poor information transmission, and high communication costs between workers, a large amount of manpower and time are wasted, which is not only inefficient but also prone to errors, affecting the operational efficiency of the entire supply chain. Although some factories have begun to use technologies such as barcodes and RFID to assist in the management of boxes, these technologies rely on manual scanning and registration, which not only increases the workload of workers, but also causes problems such as identification errors and equipment failures, making it difficult to meet the needs of efficient and accurate management.

[0029] In view of this, there is an urgent need to provide an empty container positioning method so that the empty containers in the factory can be identified and positioned, the communication costs and labor intensity of workers can be reduced, the operational efficiency of the supply chain can be improved, and the efficient and accurate management level of the supply chain can be enhanced.

[0030] The specific implementation of the present application is described in detail below with reference to the accompanying drawings.

[0031] Figure 1 An exemplary flow chart 100 showing an empty container positioning method according to some embodiments of the present application is shown in FIG. Figure 1 , the empty container positioning method shown in the embodiment of the present application may include:

[0032] In step S101, the worker positioning information and the storage area image of the loading container are obtained. In the embodiment of the present application, the aforementioned worker positioning information refers to the positioning information of the worker who loads the goods in the factory. The worker can manually input his positioning information into the worker task terminal, or the worker task terminal can be used to automatically locate the worker to obtain the worker positioning information, so that the system can receive the worker positioning information sent by the worker task terminal. Among them, the aforementioned worker task terminal is a terminal for distributing cargo loading tasks to workers, which can be a mobile terminal such as a mobile phone or a tablet. In actual applications, the form of the worker task terminal is diverse. In actual applications, the specific form of the worker task terminal needs to be determined according to the actual application situation. The present application does not impose any restrictions in this regard.

[0033] On the other hand, the above-mentioned loading container refers to a container used to load goods in a factory, which can be a box, a trailer compartment, a barrel, etc. The specific form of the loading container needs to be determined according to the actual application situation, and this application does not impose any restrictions in this regard. The storage area image of the loading container can be obtained by one or more high-definition cameras arranged in the factory. The imaging field of the high-definition camera covers the entire area where the loading container is stored, so that the storage area image of the loading container can be acquired in real time.

[0034] In step S102, the empty container attribute set present in the storage area image is identified. In the embodiment of the present application, the aforementioned empty container attribute set refers to a set composed of one or more empty container attribute elements, wherein the empty container attribute element can be used to describe attribute features such as appearance features and model features of the empty container. Exemplarily, assuming that there are yellow medium empty containers and blue large empty containers in the current storage area image, then the empty container attribute set has two attribute elements, yellow medium empty and blue large empty. It can be understood that the specific forms of attribute features such as appearance features and model features of empty containers can be diverse, and in actual applications can be set according to actual application conditions, and the present application does not impose any restrictions in this regard.

[0035] In step S103, the container classification label corresponding to each loading container and the container position information corresponding to each loading container are determined based on the storage area image. In an embodiment of the present application, the container classification label corresponding to each loading container and the container position information corresponding to each loading container can be further determined in the storage area image. Exemplarily, the corresponding container classification label and the corresponding container position information can be identified on the detection frame of each loading container. Among them, the container classification label can be used to describe the attribute characteristics such as the appearance characteristics and model characteristics of the loading container. Exemplarily, assuming that the current loading container is a yellow medium-sized fully loaded container, the container classification label identified by its corresponding detection frame is a yellow medium-sized fully loaded container. In addition, the container position information can be the coordinate positioning information of the loading container.

[0036] In step S104, the target empty container is determined based on the empty container attribute set, the container classification label corresponding to each loading container, the container location information corresponding to each loading container, and the worker location information. In the embodiment of the present application, the empty container attribute set and the container classification label corresponding to each loading container can be used to filter out the empty containers with consistent attributes to ensure the accuracy of the detection of the empty containers. It can be understood that if the cargo load of the current loading task is further combined, one or more empty containers to be selected that meet the current loading task can be found in these filtered empty containers. Exemplarily, for example, when the cargo load is greater than the first load threshold, all blue large empty containers in the storage area image can be filtered out, and when the cargo load is between the first load threshold and the second load threshold (the second load threshold is less than the first load threshold, and the second load threshold is greater than zero), all yellow medium empty containers in the storage area image can be filtered out.

[0037] Furthermore, after screening out all one or more empty containers to be selected that meet the current loading task in the storage area image, the container position information and worker positioning information corresponding to each loading container can be combined to select a target empty container that is most convenient for workers to pick up, which is conducive to reducing the workload of workers and improving their work efficiency.

[0038] The embodiment of the present application obtains the worker positioning information and the storage area image of the loading container, and then identifies the empty container attribute set existing in the storage area image, so as to determine which attributes of the empty container exist in the storage area image. Further, based on the storage area image, the container classification label corresponding to each loading container and the container position information corresponding to each loading container are determined, and then the target empty container is determined based on the empty container attribute set, the container classification label corresponding to each loading container, the container position information corresponding to each loading container, and the worker positioning information. Thus, the container classification label corresponding to each loading container and the empty container attribute set are used to accurately screen out each required empty container, realize automatic, real-time and high-precision empty container detection and positioning, and the container position information corresponding to each loading container and the worker positioning information can be used to determine the empty container closest to the worker, which is conducive to reducing the communication cost and labor intensity of the workers and improving the work efficiency of the workers. In general, the present application can identify and locate the empty containers in the factory, reduce the communication cost and labor intensity of the workers, improve the operational efficiency of the supply chain, and improve the efficient and accurate management level of the supply chain.

[0039] In some embodiments, the process of determining the target empty container can be further designed. Figure 2 The process of determining the target empty container is described in detail. Figure 2 An exemplary flow chart 200 showing an empty container positioning method according to other embodiments of the present application is shown. Figure 2 , the empty container positioning method shown in the embodiment of the present application may include:

[0040] In step S201, each candidate empty container is determined based on the empty container attribute set and the container classification label corresponding to each loading container. In the embodiment of the present application, if the container classification label of the current loading container matches the empty container attribute element in the empty container attribute set, for example, the container classification label of the current loading container is yellow medium empty, and the empty container attribute element yellow medium empty also exists in the empty container attribute set, it means that the same empty container attribute appears in the recognition process of the empty container attribute set and in the detection and determination process of the container classification label corresponding to each loading container, thereby accurately determining that an empty container with the attribute does exist in the storage area image, and then the current loading container can be determined as a candidate empty container, thereby improving the detection accuracy of the candidate empty container.

[0041] In step S202, the candidate container position corresponding to each candidate empty container is determined based on the container position information corresponding to each loaded container. In the embodiment of the present application, since all loaded containers include and cover the candidate empty containers, the position information of each candidate empty container can be determined based on the container position information corresponding to each loaded container, so that the candidate container position corresponding to each candidate empty container can be determined.

[0042] In step S203, the target empty container is determined based on the candidate container position corresponding to each candidate empty container and the worker positioning information. In the embodiment of the present application, each moving track between the worker and each candidate empty container can be determined based on the candidate container position corresponding to each candidate empty container and the worker positioning information, so that the moving distance of the worker along each moving track to reach each candidate empty container can be further determined, and then the target empty container can be determined based on the moving distance of each moving track. For example, the candidate empty container corresponding to the moving track with the shortest moving distance can be selected as the target empty container. It can be understood that in practical applications, in addition to the moving distance, other factors can be added to jointly determine the target empty container. For example, if the worker needs to pick up other objects along the way for loading, then it may be necessary to select a moving track that has a short moving distance and can pick up the object along the way. At this time, the candidate empty container corresponding to the moving track is determined as the target empty container. In practical applications, the determining factors for determining the target empty container need to be increased or decreased according to the actual application situation, and the present application does not impose any restrictions in this regard. Thereby, it is conducive to reducing the workload of workers and improving the work efficiency of workers.

[0043] In step S204, the container position information of the target empty container and the moving track corresponding to the target empty container are sent to the worker task terminal. It is understandable that the worker can conveniently view the moving track in the worker task terminal and reach the positioning position of the target empty container along the moving track, thereby reducing the communication cost of the worker. It is also understandable that in the next loading task of the worker, as the position of the worker and the load amount change, the target empty container will also change dynamically, thereby facilitating the worker to go to the positioning position of the next target empty container.

[0044] In some embodiments, after acquiring the storage area image, the storage area image may be preprocessed to improve the accuracy of subsequent recognition detection. Figure 3 The preprocessing process of the storage area image, the recognition process of the empty container attribute set, the detection and recognition process of the container classification label corresponding to each loaded container and the container position information corresponding to each loaded container are described in detail. Figure 3 An exemplary flow chart 300 showing an empty container positioning method according to some other embodiments of the present application is shown in FIG. Figure 3 , the empty container positioning method shown in the embodiment of the present application may include:

[0045] In step S301, worker location information and a storage area image of the loading container are obtained. In the embodiment of the present application, the content of step S301 is substantially the same as that of step S101, and will not be described in detail here.

[0046] In step S302, the storage area image is preprocessed. In the embodiment of the present application, the acquired storage area image may be preprocessed including but not limited to image denoising, image enhancement and image cutting, so as to improve the accuracy and processing efficiency of subsequent recognition detection.

[0047] In step S303, the attribute recognition model is used to identify the attribute set of empty containers present in the storage area image. In an embodiment of the present application, the aforementioned attribute recognition model includes a backbone recognition network and an attribute classifier, wherein the backbone recognition network can adopt ResNeSt50, which is an improved convolutional neural network model belonging to the ResNeSt series. ResNeSt is a highly modular image classification network architecture that enhances the expressiveness of features by introducing an attention mechanism. ResNeSt50 includes 50 layers, and the main structure includes: a convolution layer for extracting low-level features such as edges and corners in an image, a pooling layer for reducing the size of a feature map to reduce computational complexity and enhance the translation invariance of the model, a residual module for allowing gradients to be directly transferred from the subsequent convolution layer to the previous convolution layer through skip connections, and a Split-Attention module for automatically allocating more attention to more important features according to the weights of different features. The design of ResNeSt50 enables the network to better learn the feature identification of the image, so that ResNeSt50 can be used as the backbone recognition network to extract features of the storage area image and obtain the attribute characteristics of the loading container.

[0048] Furthermore, the structure of the above-mentioned attribute classifier may include an Average pooling layer, a Linear layer (i.e., a linear layer, also known as a fully connected layer (Fully Connected Layer) or a dense layer (DenseLayer)), and a BN (BatchNorm, batch normalization) layer. After the loading container attribute features are input into the attribute classifier, the loading container attribute features can be feature classified by the attribute classifier to obtain N types of attribute feature classification results, where N is a positive integer. Then, among these N types of attribute feature classification results, the classification results that meet the empty-load features are determined as empty container attribute elements, and these empty container attribute elements constitute an empty container attribute set. Exemplarily, assuming that the attribute classifier classifies them into 4 types of attribute feature classification results, specifically yellow medium empty, yellow medium full, blue large empty, and blue large full, then the empty container attribute elements in the empty container attribute set are yellow medium empty and blue large empty.

[0049] In the embodiment of the present application, in order to ensure the accuracy of the detection of empty containers, the attribute recognition model is used to identify the empty containers before the subsequent detection by the container detection model to determine whether there are empty containers with certain attributes or certain properties in the storage area image. In the recognition process of the attribute recognition model, it is not necessary to perform actual position detection on the identified empty containers. This is conducive to the subsequent detection of the required empty containers when it is determined that there are empty containers with certain attributes or certain properties in the storage area image.

[0050] In order to realize the recognition function of the attribute recognition model, during the training process of the attribute recognition model, several pictures containing loading containers can be collected as the training image set of the attribute recognition model. In actual application scenarios, the loading containers may be stacked together. Therefore, in the embodiment of the present application, it is not necessary to mark each loading container, but the entire stack of loading containers can be collectively marked, because the training data of the attribute recognition model only needs to mark whether there is a loading container with a certain attribute in this picture, rather than the specific position of each loading container. Thereby, the marking efficiency of the training image set can be improved. For example, assuming that the attribute feature classification results that need to be identified at present are 4 categories, and the marking order is "yellow medium empty, yellow medium full, blue large empty and blue large full", then if there is only a yellow medium empty container in the training image, the training image at this time can be marked as "1-0-0-0"; if there are yellow medium full and blue large empty in the training image, the training image at this time can be marked as "0-1-1-0". It is understandable that there are various ways to label the training image set, and the labeling method needs to be determined according to the actual application situation. This application does not impose any restrictions in this regard.

[0051] After completing the data labeling, the labeled training image set can be input into the initial attribute recognition model and iterated continuously. The cross entropy loss function can be used exemplarily to determine whether the attribute recognition model has completed the training. If the training is completed, the attribute recognition model is output.

[0052] In step S304, the container classification label corresponding to each loading container and the container position information corresponding to each loading container are determined by the container detection model. In the embodiment of the present application, the container detection model can adopt YOLOv8, in which a cross-stage partial network (CSPNet) can be used as the backbone detection network; a path aggregation network (PANet) can be used as the feature fusion layer; a decoupled head structure can be used as the output layer, and classification and detection are separated; a binary cross-entropy loss function (Binary Cross-Entropy Loss) can be used as the classification loss function, and Distribution Focal Loss+Complete IoU Loss can be used as the regression loss function. In terms of label matching, the Task-Aligned Assigner positive and negative sample allocation strategy can be adopted.

[0053] In order to realize the detection function of the container detection model for each loading container, during the training process of the container detection model, several pictures containing loading containers can be collected as the training data set of the container detection model. In the process of data annotation of the training data set, data annotation can be performed according to the annotation requirements of YOLOv8, that is, the detection frame of each classification of loading containers is annotated in each training picture of the training data set. For example, assuming that the attribute feature classification results that need to be identified currently are 4 categories, namely "yellow medium empty, yellow medium full, blue large empty and blue large full", then it is necessary to establish corresponding external detection frames at each yellow medium empty container, each yellow medium full container, each blue large empty container and each blue large full container. After completing the data annotation, the annotated training data set can be input into the initial container detection model, and it is continuously iterated to reduce the loss function value without overfitting, and finally output the trained container detection model. The trained container detection model has the ability to detect each loading container, and can identify the container classification label and container position information for each detected loading container.

[0054] It can be understood that in the embodiment of the present application, only when the same empty container attribute is output in the empty container attribute set and the container classification label corresponding to each loaded container, can it be determined that there is an empty container corresponding to the empty container attribute in the storage area image, thereby ensuring the accuracy of empty container detection.

[0055] Corresponding to the aforementioned application function implementation method embodiment, the present application also provides a device for positioning an empty container and a corresponding embodiment.

[0056] Figure 4 FIG. 4 is a block diagram showing the hardware configuration of an empty container positioning device 400 that can implement the empty container positioning method of an embodiment of the present application. Figure 4 As shown, the device 400 for positioning an empty container may include a processor 410 and a memory 420. Figure 4 In the device 400 for positioning an empty container, only the components related to this embodiment are shown. Therefore, it is obvious to a person skilled in the art that the device 400 for positioning an empty container may also include components related to Figure 4 The components shown in the figure are different from the common components. For example: fixed-point arithmetic units.

[0057] The device 400 for empty container location may correspond to a computing device having various processing functions, such as functions for generating a neural network, training or learning a neural network, quantizing a floating-point neural network to a fixed-point neural network, or retraining a neural network. For example, the device 400 for empty container location may be implemented as various types of devices, such as a personal computer (PC), a server device, a mobile device, etc.

[0058] The processor 410 controls all functions of the device 400 for empty container positioning. For example, the processor 410 controls all functions of the device 400 for empty container positioning by executing a program stored in the memory 420 on the device 400 for empty container positioning. The processor 410 may be implemented by a central processing unit (CPU), a graphics processing unit (GPU), an application processor (AP), an artificial intelligence processor chip (IPU), etc. provided in the device 400 for empty container positioning. However, the present application is not limited thereto.

[0059] In some embodiments, the processor 410 may include an input / output (I / O) unit 411 and a computing unit 412. The I / O unit 411 may be used to receive various data, such as worker positioning information and a storage area image of a loading container. Exemplarily, the computing unit 412 may be used to identify an empty container attribute set present in the storage area image received via the I / O unit 411; and then determine the container classification label corresponding to each loading container and the container location information corresponding to each loading container based on the storage area image; and determine the target empty container based on the empty container attribute set, the container classification label corresponding to each loading container, the container location information corresponding to each loading container, and the worker positioning information. This target empty container may be output by the I / O unit 411, for example. The output data may be provided to the memory 420 for reading and use by other devices (not shown), or may be directly provided to other devices for use.

[0060] The memory 420 is hardware for storing various data processed in the device 400 for empty container positioning. For example, the memory 420 can store processed data and data to be processed in the device 400 for empty container positioning. The memory 420 can store data sets involved in the empty container positioning method process that has been processed or is to be processed by the processor 410, such as worker positioning information and storage area images of loaded containers. In addition, the memory 420 can store applications, drivers, etc. to be driven by the device 400 for empty container positioning. For example: the memory 420 can store various programs related to the empty container positioning method to be executed by the processor 410. The memory 420 can be a DRAM, but the present application is not limited thereto. The memory 420 can include at least one of a volatile memory or a non-volatile memory. The non-volatile memory can include a read-only memory (ROM), a programmable ROM (PROM), an electrically programmable ROM (EPROM), an electrically erasable programmable ROM (EEPROM), a flash memory, a phase change RAM (PRAM), a magnetic RAM (MRAM), a resistive RAM (RRAM), a ferroelectric RAM (FRAM), and the like. The volatile memory may include dynamic RAM (DRAM), static RAM (SRAM), synchronous DRAM (SDRAM), PRAM, MRAM, RRAM, ferroelectric RAM (FeRAM), etc. In an embodiment, the memory 420 may include at least one of a hard disk drive (HDD), a solid state drive (SSD), a high-density flash memory (CF), a secure digital (SD) card, a micro secure digital (Micro-SD) card, a mini secure digital (Mini-SD) card, an extreme digital (xD) card, caches, or a memory stick.

[0061] In summary, the specific functions implemented by the memory 420 and the processor 410 of the device 400 for positioning empty containers provided in the implementation manner of this specification can be explained in comparison with the aforementioned implementation manner in this specification, and can achieve the technical effects of the aforementioned implementation manner, and will not be repeated here.

[0062] In this embodiment, the processor 410 can be implemented in any suitable manner. For example, the processor 410 can take the form of a microprocessor or processor and a computer-readable medium storing computer-readable program code (such as software or firmware) executable by the (micro)processor, a logic gate, a switch, an application specific integrated circuit (ASIC), a programmable logic controller, and an embedded microcontroller, etc.

[0063] It should also be understood that any module, unit, component, server, computer, terminal or device that executes instructions exemplified herein may include or otherwise access computer-readable media, such as storage media, computer storage media or data storage devices (removable and / or non-removable) such as disks, optical disks or tapes. Computer storage media may include volatile and non-volatile, removable and non-removable media implemented in any method or technology for storing information, such as computer-readable instructions, data structures, program modules or other data.

[0064] The foregoing content can be better understood in accordance with the following terms:

[0065] Clause A1. A method for locating an empty container, comprising: obtaining worker positioning information and a storage area image of loaded containers; identifying a set of empty container attributes present in the storage area image; determining, based on the storage area image, a container classification label corresponding to each loaded container and container position information corresponding to each loaded container; and determining a target empty container based on the empty container attribute set, the container classification label corresponding to each loaded container, the container position information corresponding to each loaded container and the worker positioning information.

[0066] Clause A2. The empty container positioning method according to Clause A1, wherein the determining the target empty container based on the empty container attribute set, the container classification label corresponding to each loading container, the container position information corresponding to each loading container and the worker positioning information comprises: determining each candidate empty container based on the empty container attribute set and the container classification label corresponding to each loading container; determining the candidate container position corresponding to each candidate empty container based on the container position information corresponding to each loading container; and determining the target empty container based on the candidate container position corresponding to each candidate empty container and the worker positioning information.

[0067] Clause A3. An empty container positioning method according to Clause A2, wherein the determining of each candidate empty container based on the empty container attribute set and the container classification label corresponding to each loaded container includes: if the container classification label of the current loaded container matches the empty container attribute element in the empty container attribute set, then determining the current loaded container as the candidate empty container.

[0068] Clause A4. An empty container positioning method according to Clause A2, wherein determining the target empty container based on the candidate container position corresponding to each candidate empty container and the worker positioning information includes: determining each movement trajectory between the worker and each candidate empty container based on the candidate container position corresponding to each candidate empty container and the worker positioning information; and determining the target empty container based on the moving distance of each movement trajectory.

[0069] Clause A5. The empty container positioning method according to clause A1, wherein the identifying the empty container attribute set present in the storage area image comprises:

[0070] The attribute set of the empty container present in the storage area image is identified by an attribute recognition model.

[0071] Clause A6. The empty container positioning method according to Clause A5, wherein the attribute recognition model includes a backbone recognition network and an attribute classifier; wherein the identifying the empty container attribute set existing in the storage area image through the attribute recognition model includes: performing feature extraction on the storage area image through the backbone recognition network to obtain loaded container attribute features; performing feature classification on the loaded container attribute features through the attribute classifier to obtain the empty container attribute set.

[0072] Clause A7. An empty container positioning method according to Clause A1, wherein the step of determining the container classification label corresponding to each loading container and the container position information corresponding to each loading container based on the storage area image includes: determining the container classification label corresponding to each loading container and the container position information corresponding to each loading container through a container detection model.

[0073] Clause A8. An empty container positioning method according to Clause A4, wherein, after determining the target empty container based on the empty container attribute set, the container classification label corresponding to each loading container, the container position information corresponding to each loading container and the worker positioning information, the method further includes: sending the container position information of the target empty container and the moving trajectory corresponding to the target empty container to the worker task terminal.

[0074] Item A9, a device for locating empty containers, comprising: a memory; and at least one processor, configured to: obtain worker positioning information and a storage area image of loaded containers; identify a set of empty container attributes present in the storage area image; determine, based on the storage area image, a container classification label corresponding to each loaded container and container position information corresponding to each loaded container; and determine a target empty container based on the empty container attribute set, the container classification label corresponding to each loaded container, the container position information corresponding to each loaded container and the worker positioning information.

[0075] Item A10, a non-volatile machine-readable medium having program code stored thereon for locating empty containers, wherein when the program code is executed by at least one processor, the code guides the execution operation of the at least one processor, and the program code includes: obtaining worker positioning information and a storage area image of loaded containers; identifying a set of empty container attributes existing in the storage area image; determining, based on the storage area image, a container classification label corresponding to each loaded container and container location information corresponding to each loaded container; and determining a target empty container based on the empty container attribute set, the container classification label corresponding to each loaded container, the container location information corresponding to each loaded container and the worker positioning information.

Claims

1. A method for positioning an empty container, characterized in that: include: Obtain worker location information and images of storage areas where loaded containers are stored; identifying a set of empty container attributes present in the storage area image; Determine, based on the storage area image, a container classification label corresponding to each loading container and container position information corresponding to each loading container; as well as The target empty container is determined based on the empty container attribute set, the container classification label corresponding to each loaded container, the container position information corresponding to each loaded container, and the worker positioning information.

2. The empty container positioning method according to claim 1, characterized in that: The determining of the target empty container based on the empty container attribute set, the container classification label corresponding to each loaded container, the container position information corresponding to each loaded container, and the worker positioning information comprises: Determine each candidate empty container based on the empty container attribute set and the container classification label corresponding to each loaded container; Determine a candidate container position corresponding to each candidate empty container based on the container position information corresponding to each loaded container; and The target empty container is determined based on the candidate container position corresponding to each candidate empty container and the worker positioning information.

3. The empty container positioning method according to claim 2, characterized in that: The determining each candidate empty container based on the empty container attribute set and the container classification label corresponding to each loaded container comprises: If the container classification label of the current loaded container matches the empty container attribute element in the empty container attribute set, the current loaded container is determined to be the candidate empty container.

4. The empty container positioning method according to claim 2, characterized in that: The determining the target empty container based on the candidate container position corresponding to each candidate empty container and the worker positioning information comprises: Determine each movement trajectory between the worker and each candidate empty container based on the candidate container position corresponding to each candidate empty container and the worker positioning information; and The target empty container is determined based on the moving distance of each moving trajectory.

5. The empty container positioning method according to claim 1, characterized in that: The identifying of the empty container attribute set present in the storage area image comprises: The attribute set of the empty container present in the storage area image is identified by an attribute recognition model.

6. The empty container positioning method according to claim 5, characterized in that: The attribute recognition model includes a backbone recognition network and an attribute classifier; wherein, identifying the attribute set of empty containers present in the storage area image by the attribute recognition model includes: Extracting features of the storage area image through the backbone recognition network to obtain attribute features of the loading container; The attribute features of the loaded container are classified by the attribute classifier to obtain the empty container attribute set.

7. The empty container positioning method according to claim 1, characterized in that: The step of determining the container classification label corresponding to each loading container and the container position information corresponding to each loading container based on the storage area image comprises: The container classification label corresponding to each loading container and the container position information corresponding to each loading container are determined through the container detection model.

8. The empty container positioning method according to claim 4, characterized in that: After determining the target empty container based on the empty container attribute set, the container classification label corresponding to each loaded container, the container position information corresponding to each loaded container, and the worker positioning information, the method further includes: The container position information of the target empty container and the movement trajectory corresponding to the target empty container are sent to the worker task terminal.

9. A device for positioning an empty container, characterized in that: include: Memory; as well as at least one processor configured to: Obtain worker location information and images of storage areas where loaded containers are stored; identifying a set of empty container attributes present in the storage area image; Determine, based on the storage area image, a container classification label corresponding to each loading container and container position information corresponding to each loading container; as well as The target empty container is determined based on the empty container attribute set, the container classification label corresponding to each loaded container, the container position information corresponding to each loaded container, and the worker positioning information.

10. A non-transitory machine-readable medium having stored thereon a program code for empty container positioning, wherein when the program code is executed by at least one processor, the code directs the execution operation of the at least one processor, the program code comprising: Obtain worker location information and images of storage areas where loaded containers are stored; identifying a set of empty container attributes present in the storage area image; Determine, based on the storage area image, a container classification label corresponding to each loading container and container position information corresponding to each loading container; as well as The target empty container is determined based on the empty container attribute set, the container classification label corresponding to each loaded container, the container position information corresponding to each loaded container, and the worker positioning information.

Images