Cargo storage management method and system, computer program product and electronic equipment

Abstract

The invention discloses a cargo storage management method and system, a computer program product and electronic equipment. Relates to the financial science and technology field or other related fields, and the method comprises the steps: obtaining goods images of a plurality of goods stored in a warehouse, carrying out the preprocessing of the goods images, and obtaining a to-be-detected goods image, the goods images being the images of the plurality of goods; a target detection model is obtained, the to-be-detected cargo image is input into the target detection model, a cargo identification result is output, the target detection model is obtained through training of a target detection algorithm, the cargo identification result comprises cargo types and cargo names of N cargoes and position information of the N cargoes, and N is a positive integer; and obtaining a cargo storage rule, and generating a cargo arrangement strategy according to the cargo identification result and the cargo storage rule. According to the invention, the problem that the management efficiency is low when the goods stored in the warehouse are managed in the prior art is solved.

Description

Technical Field

[0001] The present application relates to the field of financial technology or other related fields, and specifically, to a method, system, computer program product and electronic device for storage management of goods. Background Art

[0002] In the sorting and management of goods in the warehouses of financial institutions, the traditional manual operation method is not only inefficient but also prone to errors, especially when faced with large quantities of goods of various types that require high security protection. In recent years, with the rapid development of artificial intelligence technology, especially the maturity of target detection technology, more and more financial institutions have begun to try to introduce automated systems to optimize warehouse operation processes. In the application of financial institutions' warehouses, target detection technology, as a technology that can identify and locate specific objects from images or videos, mainly collects environmental data through sensors such as 3D laser scanning and infrared thermal imaging, and then uses deep learning models to process these data, thereby realizing real-time monitoring and identification of goods.

[0003] However, traditional target detection technology has the following limitations in the warehouse environment of financial institutions: poor environmental adaptability: the environment of financial institution warehouses is relatively complex, with a wide variety of goods and different shapes, which puts higher requirements on the environmental adaptability and recognition accuracy of target detection technology; limited data processing capabilities: there may be thousands or even tens of thousands of different goods stored in the warehouse, each of which has its own unique characteristics and storage requirements, which leads to the complexity of data processing and high demand for computing resources; the goods stored in the warehouses of financial institutions may include precious metals, important documents, etc., and it is impossible to protect the security and privacy of these sensitive information while ensuring recognition accuracy and real-time performance; although the automated target detection system can significantly improve work efficiency, its equipment procurement cost, maintenance cost and possible upgrade cost are high; real-time and reliability: in warehouse operations, the system's real-time response capability and long-term operational reliability are crucial to avoiding delays and errors.

[0004] There is a problem of low management efficiency when managing goods stored in warehouses in related technologies, and no effective solution has been proposed yet. Summary of the invention

[0005] The main purpose of the present application is to provide a cargo storage management method, system, computer program product and electronic device to solve the problem of low management efficiency in the management of cargo stored in a warehouse in the related art.

[0006] In order to achieve the above-mentioned purpose, according to one aspect of the present application, a method for storage management of goods is provided. The method comprises: obtaining cargo images of multiple goods stored in a warehouse, preprocessing the cargo images, and obtaining cargo images to be detected, wherein the cargo images refer to images of multiple goods; obtaining a target detection model, inputting the cargo images to be detected into the target detection model, and outputting cargo recognition results, wherein the target detection model is obtained by training a target detection algorithm, and the cargo recognition results include cargo types, cargo names, and location information of N cargoes, where N is a positive integer; obtaining cargo storage rules, and generating cargo sorting strategies according to the cargo recognition results and cargo storage rules.

[0007] Furthermore, the target detection model is trained in the following manner: obtaining the cargo name, cargo type and historical cargo images of historical cargo stored in the warehouse within the historical time period; obtaining the initial detection model, and replacing the first feature extraction layer in the initial detection model with the second feature extraction layer to obtain a preset detection model, wherein the second feature extraction layer is used to identify features in the historical cargo images whose area is less than the area threshold, and the first feature extraction layer is used to identify features in the historical cargo images whose area is greater than or equal to the area threshold and features whose area is less than the area threshold; a model training set is composed of cargo name, cargo type and historical cargo images, and the preset detection model is trained using the model training set to obtain a target detection model, wherein the target detection model includes a target feature extraction layer, and the target feature extraction layer is trained by the second feature extraction layer.

[0008] Furthermore, the target detection model includes a pre-classification model and a cargo recognition model. The cargo image to be detected is input into the target detection model, and the cargo recognition result is outputted, including: inputting the cargo image to be detected into the pre-classification model, extracting cargo features of the cargo image to be detected by the pre-classification model, and obtaining M cargo features, wherein each cargo feature refers to a cargo feature of at least one cargo, and M is a positive integer; inputting the M cargo features into the cargo recognition model, and the cargo recognition model recognizes each cargo feature to obtain a cargo recognition result.

[0009] Furthermore, the pre-classification model includes a second feature extraction layer, and the pre-classification model extracts cargo features of the cargo image to be detected, and obtaining M cargo features includes: obtaining an area threshold, segmenting the cargo image to be detected according to the area threshold, and obtaining M segmented cargo images; and performing feature extraction on the M segmented cargo images by the second feature extraction layer to obtain M cargo features.

[0010] Furthermore, before obtaining the cargo storage rules, the method also includes: extracting location information of N cargoes from the cargo identification results, and verifying the location information of the N cargoes according to the cargo storage rules; if there is at least one cargo whose location information does not comply with the cargo storage rules, generating prompt information, and executing the step of generating a cargo sorting strategy according to the cargo identification results and the cargo storage rules, wherein the prompt information is used to prompt the staff to sort the N cargoes; if the location information of the N cargoes complies with the cargo storage rules, stopping the step of generating a cargo sorting strategy according to the cargo identification results and the cargo storage rules.

[0011] Furthermore, generating a cargo sorting strategy according to the cargo identification results and the cargo storage rules includes: obtaining cargo types of N cargoes from the cargo identification results, obtaining N cargo types, and determining storage rules of the N cargo types from the cargo storage rules; obtaining a warehouse structure diagram of the warehouse, inputting the warehouse structure diagram and the storage rules of the N cargo types into the allocation scheduling algorithm, and outputting the cargo sorting strategy.

[0012] Furthermore, after generating a cargo sorting strategy according to the cargo identification results and the cargo storage rules, the method further includes: sending the cargo sorting strategy to the cargo sorting equipment, and controlling the cargo sorting equipment to carry and sort the N cargoes according to the cargo sorting strategy.

[0013] In order to achieve the above-mentioned purpose, according to another aspect of the present application, a cargo processing system is provided. The system includes: a camera module, which is used to capture cargo images of multiple cargoes stored in the warehouse, pre-process the cargo images, obtain cargo images to be detected, and send the cargo images to be detected to a target detection module; a target detection module, which is used to input the cargo images to be detected into a target detection model when receiving the cargo images to be detected, output cargo identification results, and also to obtain cargo storage rules, generate cargo sorting strategies according to the cargo identification results and cargo storage rules, and send the cargo sorting strategies to a management module; a management module, which is connected to the cargo sorting equipment in communication, is used to display the cargo sorting strategies to the staff, and is also used to control the cargo sorting equipment to carry and sort multiple cargoes according to the cargo sorting strategies.

[0014] In order to achieve the above-mentioned purpose, according to another aspect of the present application, a storage management device for goods is provided. The device includes: a first acquisition unit, which is used to acquire goods images of multiple goods stored in the warehouse, pre-process the goods images, and obtain the goods images to be detected, wherein the goods images refer to the images of multiple goods; a second acquisition unit, which is used to acquire a target detection model, input the goods images to be detected into the target detection model, and output the goods recognition results, wherein the target detection model is obtained by training the target detection algorithm, and the goods recognition results include the goods types, goods names and location information of N goods, where N is a positive integer; a third acquisition unit, which is used to acquire goods storage rules, and generate goods sorting strategies according to the goods recognition results and the goods storage rules.

[0015] According to another aspect of an embodiment of the present invention, there is further provided a computer storage medium, which is used to store a program, wherein when the program is run, the device where the computer storage medium is located is controlled to execute a method for storage management of goods.

[0016] According to another aspect of an embodiment of the present invention, there is also provided an electronic device comprising one or more processors and a memory; the memory stores computer-readable instructions, and the processor is used to execute the computer-readable instructions, wherein a method for storage management of goods is executed when the computer-readable instructions are executed.

[0017] According to another aspect of an embodiment of the present invention, a computer program product is provided, including a computer program, wherein when the computer program is executed by a processor, a storage management method of goods is performed.

[0018] Through the present application, the following steps are adopted: obtaining cargo images of multiple cargoes stored in a warehouse, preprocessing the cargo images, and obtaining cargo images to be detected, wherein the cargo images refer to images of multiple cargoes; obtaining a target detection model, inputting the cargo images to be detected into the target detection model, and outputting cargo recognition results, wherein the target detection model is obtained by training a target detection algorithm, and the cargo recognition results include cargo types, cargo names, and location information of N cargoes, where N is a positive integer; obtaining cargo storage rules, and generating cargo sorting strategies according to the cargo recognition results and cargo storage rules, thereby solving the problem of low management efficiency in managing cargo stored in a warehouse in the related art, by obtaining cargo images of cargo stored in a warehouse, preprocessing the cargo images to obtain cargo images to be detected, inputting the cargo images to be detected into the target detection model and outputting cargo recognition results, and finally generating cargo sorting strategies based on the cargo recognition results and preset cargo storage rules, thereby achieving the effect of improving cargo sorting efficiency and reducing labor costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] The drawings constituting a part of the present application are used to provide a further understanding of the present application. The illustrative embodiments and descriptions of the present application are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0020] Figure 1 is a flow chart of a method for storage management of goods provided according to an embodiment of the present application;

[0021] Figure 2 is a schematic diagram of an optional cargo storage management method provided according to an embodiment of the present application;

[0022] Figure 3 is a schematic diagram of a cargo handling system provided according to an embodiment of the present application;

[0023] Figure 4 is a schematic diagram of a storage management device for goods provided according to an embodiment of the present application;

[0024] Figure 5 is a schematic diagram of an electronic device provided according to an embodiment of the present application. DETAILED DESCRIPTION

[0025] It should be noted that, in the absence of conflict, the embodiments and features in the embodiments of the present application can be combined with each other. The present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0026] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present application.

[0027] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0028] It should be noted that the relevant information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data for display, data for analysis, etc.) involved in this disclosure are all information and data authorized by the user or fully authorized by all parties. For example, an interface is set between this system and the relevant user or organization. Before obtaining relevant information, it is necessary to send an acquisition request to the aforementioned user or organization through the interface, and obtain relevant information after receiving the consent information fed back by the aforementioned user or organization.

[0029] It should be noted that the collected information used in this application is information and data authorized by the user or fully authorized by all parties, and the collection, storage, use, processing, transmission, provision, disclosure and application of the relevant data comply with the relevant laws, regulations and standards of the relevant regions, take necessary confidentiality measures, do not violate public order and good morals, and provide corresponding operation entrances for users to choose to authorize or refuse to use it.

[0030] The present invention is described below in conjunction with preferred implementation steps. Figure 1 is a flow chart of a method for storage management of goods provided in accordance with an embodiment of the present application, such as Figure 1 As shown, the method comprises the following steps:

[0031] Step S101, obtaining cargo images of a plurality of cargoes stored in a warehouse, and preprocessing the cargo images to obtain cargo images to be detected, wherein the cargo images refer to images of a plurality of cargoes.

[0032] Specifically, the warehouse can be a warehouse associated with a financial institution, which can store precious metals, important documents and other items. In order to effectively manage large quantities of multi-type goods that require high security protection, cameras and other equipment can be used to first photograph the goods stored in the warehouse. After obtaining images of multiple goods, in order to improve the accuracy of subsequent target detection, image denoising, enhancement and other operations are performed to obtain images of the goods to be detected.

[0033] Step S102, obtaining a target detection model, inputting the image of the goods to be detected into the target detection model, and outputting a cargo recognition result, wherein the target detection model is obtained by training the target detection algorithm, and the cargo recognition result includes cargo types, cargo names, and location information of N cargoes, where N is a positive integer.

[0034] Specifically, the target detection model can be a trained deep learning detection algorithm. After preprocessing the cargo image, the obtained cargo image to be detected is input into the target detection model. After the model recognizes and detects the image, it outputs the cargo recognition result including the type and name of all cargo stored in the warehouse and the placement location of each cargo. The above-mentioned deep learning detection algorithm can be Faster R-CNN (Region-based Convolutional Neural Network with Regions Proposed by a Region Proposal Network) algorithm, which can make the entire detection process more efficient and close to real-time detection. In addition, the algorithm can perform fine classification and regression of candidate regions through deep learning, thereby improving the detection accuracy.

[0035] Step S103, obtaining cargo storage rules, and generating cargo sorting strategies according to cargo identification results and cargo storage rules.

[0036] Specifically, after the target detection model outputs the cargo identification result, the result is analyzed. If the analysis result indicates that there is an abnormality in the cargo placement method or placement position, the cargo storage rule can be obtained. Finally, the cargo storage rule and the cargo identification result are used to generate a cargo sorting strategy, and relevant sorting equipment is used to manage the cargo according to the cargo sorting strategy.

[0037] The storage management method of goods provided in the embodiment of the present application obtains goods images of multiple goods stored in the warehouse, pre-processes the goods images, and obtains goods images to be detected, wherein the goods images refer to images of multiple goods; obtains a target detection model, inputs the goods images to be detected into the target detection model, and outputs a goods recognition result, wherein the target detection model is obtained by training the target detection algorithm, and the goods recognition result includes the goods types and goods names of N goods and the location information of N goods, where N is a positive integer; obtains goods storage rules, and generates a goods sorting strategy according to the goods recognition results and the goods storage rules, thereby solving the problem of low management efficiency when managing goods stored in the warehouse in the related art, by obtaining goods images of the goods stored in the warehouse, pre-processing the goods images to be detected, inputting the goods images to be detected into the target detection model and outputting the goods recognition results, and finally generating the goods sorting strategy based on the goods recognition results and the preset goods storage rules, thereby achieving the effect of improving the goods sorting efficiency and reducing the labor cost.

[0038] Before using the target detection model for image analysis, the model needs to be trained. Optionally, in the storage management method for goods provided in the embodiment of the present application, the target detection model is trained in the following manner: obtaining the name of goods, the type of goods and the historical images of goods stored in the warehouse during the historical time period; obtaining an initial detection model, and replacing the first feature extraction layer in the initial detection model with the second feature extraction layer to obtain a preset detection model, wherein the second feature extraction layer is used to identify features in the historical goods images whose area is less than the area threshold, and the first feature extraction layer is used to identify features in the historical goods images whose area is greater than or equal to the area threshold and features whose area is less than the area threshold; a model training set is composed of the name of goods, the type of goods and the historical goods images, and the preset detection model is trained using the model training set to obtain a target detection model, wherein the target detection model includes a target feature extraction layer, and the target feature extraction layer is trained by the second feature extraction layer.

[0039] Specifically, first, relevant information about historical goods stored in the warehouse during the historical time period can be collected, for example, the name of the goods, the type of goods, and the images of the historical goods. Furthermore, after obtaining the initial detection model composed of the Faster R-CNN algorithm, the feature extraction layer (i.e., the FPN (Feature Pyramid Network) layer) used to identify features with an area greater than or equal to the area threshold and features with an area less than the area threshold in the historical goods image can be replaced with a feature extraction layer that can identify small targets: the S-FPN (Scale-aware Feature Pyramid Network) layer, which focuses on improving the speed of small target detection to meet real-time requirements, and then obtains a preset detection model.

[0040] Furthermore, the data of the goods collected above are used to form a model training set, and the preset detection model is trained using the model training set to obtain a target detection model. By fine-tuning the model architecture, especially enhancing the ability to capture the specific characteristics of goods of financial institutions, the model recognition accuracy and generalization ability are significantly improved. This embodiment trains the preset detection model with the feature extraction layer replaced to obtain the target detection model, which can improve the recognition accuracy of the model and lay the foundation for the subsequent accurate identification of goods.

[0041] In order to improve the recognition accuracy of the image, optionally, in the storage management method of goods provided in the embodiment of the present application, the target detection model includes a pre-classification model and a goods recognition model, the image of the goods to be detected is input into the target detection model, and the output of the goods recognition result includes: inputting the image of the goods to be detected into the pre-classification model, extracting the goods features of the image of the goods to be detected by the pre-classification model, and obtaining M goods features, wherein each goods feature refers to a goods feature of at least one goods, and M is a positive integer; inputting the M goods features into the goods recognition model, and the goods recognition model recognizes each goods feature to obtain a goods recognition result.

[0042] Specifically, when using the target detection model for image processing, since the target detection model includes a pre-classification model that can extract cargo features and a cargo recognition model that can recognize cargo features, after the target detection model receives the cargo image to be detected, the pre-classification model can be used to extract features of the image, thereby obtaining cargo features containing different labels.

[0043] Furthermore, after the cargo features are input into the cargo recognition model, the cargo recognition model is used to recognize each cargo feature, determine the type, name and other information of each cargo, and then output the cargo recognition result. This embodiment can greatly improve the efficiency and accuracy of cargo processing by using the target detection model to output the cargo recognition result.

[0044] Optionally, in the storage management method for goods provided in an embodiment of the present application, the pre-classification model includes a second feature extraction layer, and the pre-classification model extracts goods features of the goods image to be detected, and obtaining M goods features includes: obtaining an area threshold, and segmenting the goods image to be detected according to the area threshold to obtain M segmented goods images; and performing feature extraction on the M segmented goods images by the second feature extraction layer to obtain M goods features.

[0045] Specifically, when the pre-classification model is used to extract cargo features, since the model includes a second feature extraction layer that can extract small targets, after the pre-classification model receives the cargo image to be detected, the cargo image to be detected is segmented according to a set area threshold to obtain multiple segmented cargo images, and then the second feature extraction layer is used to extract features from each segmented cargo image to obtain corresponding cargo features.

[0046] In order to improve the sorting efficiency, optionally, in the storage management method for goods provided in the embodiment of the present application, before obtaining the goods storage rules, the method also includes: extracting the location information of N goods from the goods identification results, and verifying the location information of the N goods according to the goods storage rules; if there is at least one goods whose location information does not comply with the goods storage rules, generating prompt information, and executing the step of generating a goods sorting strategy according to the goods identification results and the goods storage rules, wherein the prompt information is used to prompt the staff to sort the N goods; if the location information of the N goods complies with the goods storage rules, stopping the step of generating a goods sorting strategy according to the goods identification results and the goods storage rules.

[0047] Specifically, after the target detection model outputs the cargo identification results, it is necessary to analyze the placement of each cargo, that is, it is necessary to determine the location information of each cargo based on the cargo identification results, and verify the location information based on the preset cargo storage rules. If the location information of one or more cargoes does not comply with the above cargo storage rules, it is necessary to generate a cargo sorting strategy based on the cargo identification results and other information, and it is also necessary to generate prompt information to remind the staff to sort the relevant cargo; on the contrary, if the location information of all cargoes complies with the cargo storage rules, the sorting work can be ended directly.

[0048] The cargo sorting strategy can be generated by an allocation and scheduling algorithm. Optionally, in the cargo storage management method provided in an embodiment of the present application, generating a cargo sorting strategy based on cargo identification results and cargo storage rules includes: obtaining cargo types of N cargoes from the cargo identification results, obtaining N cargo types, and determining storage rules for the N cargo types from the cargo storage rules; obtaining a warehouse structure diagram of the warehouse, inputting the warehouse structure diagram and the storage rules for the N cargo types into the allocation and scheduling algorithm, and outputting the cargo sorting strategy.

[0049] Specifically, when using the cargo identification results to generate a cargo sorting strategy, the cargo type of all cargo is first extracted from the identification results, and then the storage rules of the cargo are determined based on the cargo type. For example, when the cargo type is precious metals, the precious metals need to be stored in a fireproof and moisture-proof environment, and the temperature of the warehouse environment needs to be controlled to prevent affecting the physical properties of the precious metals; when the cargo type is a paper document, the storage rule can be to place it in the archive area for moisture-proof storage.

[0050] Furthermore, after obtaining the storage rules for each cargo, the warehouse structure diagram of the cargo warehouse can be input into the allocation and scheduling algorithm at the same time, and the algorithm performs analysis and processing to obtain the cargo sorting strategy. This embodiment uses the allocation and scheduling algorithm to output the cargo sorting strategy, thereby helping the warehouse manager to efficiently organize the storage of cargo, ensure the operational efficiency and safety of the warehouse, and not only reduce the complexity and error rate of manual operation, but also improve the overall sorting efficiency.

[0051] Optionally, in the cargo storage management method provided in an embodiment of the present application, after generating a cargo sorting strategy based on cargo identification results and cargo storage rules, the method further includes: sending the cargo sorting strategy to a cargo sorting device, and controlling the cargo sorting device to transport and sort N cargoes according to the cargo sorting strategy.

[0052] Specifically, after receiving the cargo sorting strategy, the cargo sorting equipment can summarize and sort the cargo that needs to be adjusted according to the strategy, thereby helping the staff to reduce the operation time and improve the sorting efficiency.

[0053] It should be noted that the steps shown in the flowcharts of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and that, although a logical order is shown in the flowcharts, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0054] The present application also provides a method for storage management of goods. Figure 2 is a schematic diagram of an optional cargo storage management method provided according to an embodiment of the present application, such as Figure 2 As shown, the method includes:

[0055] In order to effectively manage large quantities of multi-type goods that require high security protection, we can first collect relevant information about historical goods stored in the warehouse during the historical time period, such as the name, type and image of historical goods, and then build a comprehensive and accurate model training set based on the above information to provide a solid data foundation for subsequent model training.

[0056] After using the data of the goods collected above to form the model training set of the financial institution, the preset detection model is trained using the model training set. In this way, the target detection model is obtained. By fine-tuning the model architecture, especially enhancing the ability to capture the specific cargo characteristics of the financial institution, the model recognition accuracy and generalization ability are significantly improved.

[0057] Before using the trained target detection model for image recognition, first use a camera and other equipment to photograph the goods stored in the warehouse. After collecting multiple cargo images, in order to improve the accuracy of subsequent target detection, image denoising, enhancement and other operations are performed to obtain the image of the goods to be detected. The image of the goods to be detected is then input into the target detection model, which performs cargo detection and identification, and then outputs the cargo identification results. Finally, the cargo sorting strategy is generated and executed based on the cargo identification results and cargo storage rules, that is, the cargo sorting equipment is used to manage the goods according to the cargo sorting strategy.

[0058] This embodiment obtains cargo images of the goods stored in the warehouse, pre-processes the cargo images to obtain cargo images to be detected, inputs the cargo images to be detected into the target detection model and outputs cargo recognition results, and finally generates cargo sorting strategies based on the cargo recognition results and preset cargo storage rules, thereby achieving the effect of improving cargo sorting efficiency and reducing labor costs.

[0059] The present application also provides a cargo handling system. Figure 3 is a schematic diagram of a cargo handling system provided according to an embodiment of the present application, such as Figure 3 As shown, the system includes:

[0060] The camera module 301 is used to capture cargo images of multiple cargoes stored in the warehouse, pre-process the cargo images to obtain cargo images to be detected, and send the cargo images to be detected to the target detection module 302 .

[0061] Specifically, the camera module 301 may include multiple high-resolution cameras installed at different locations in the warehouse to ensure all-round monitoring of the goods. In order to effectively manage a large number of goods of various types that require high security protection, the module can capture the goods images of all goods stored in the warehouse, and after image denoising and enhancement, the processed goods images to be detected are input into the target detection module 302, which performs image recognition.

[0062] The target detection module 302 is used to input the image of the goods to be detected into the target detection model when receiving the image of the goods to be detected, output the goods recognition result, and obtain the goods storage rules, generate the goods sorting strategy according to the goods recognition result and the goods storage rules, and send the goods sorting strategy to the management module 303.

[0063] Specifically, after receiving the image of the goods to be processed collected by the camera module 301, the image is input into the trained target detection model. After the model recognizes and detects the image, it outputs the cargo identification result including the type and name of all the goods stored in the warehouse and the placement position of each cargo. The result is analyzed. If the analysis result indicates that there is an abnormality in the placement method or placement position of the goods, the cargo storage rule can be obtained. Finally, the cargo storage rule and the cargo identification result are used to generate a cargo sorting strategy, and the cargo sorting strategy is sent to the management module 303, thereby achieving the purpose of timely updating the cargo status and ensuring high-accuracy cargo identification and positioning.

[0064] The management module 303 is connected to the cargo sorting device for communicating with the cargo sorting device, and is used to display the cargo sorting strategy to the staff, and is also used to control the cargo sorting device to carry and sort multiple cargoes according to the cargo sorting strategy.

[0065] Specifically, the management module 303 can be connected to the cargo sorting device, and can control the cargo sorting device to manage the cargo according to the cargo sorting strategy, and can also display the above cargo sorting strategy to the staff.

[0066] The cargo processing system provided in the embodiment of the present application uses a camera module 301 to capture cargo images of multiple cargoes stored in a warehouse, pre-process the cargo images, obtain cargo images to be detected, and send the cargo images to a target detection module 302; the target detection module 302 is used to input the cargo images to be detected into a target detection model when receiving the cargo images to be detected, output cargo recognition results, and also to obtain cargo storage rules, generate cargo sorting strategies according to the cargo recognition results and cargo storage rules, and send the cargo sorting strategies to a management module 303; the management module 303 is connected to the cargo sorting equipment in communication with the workers to sort the cargo. The cargo sorting strategy is displayed to the personnel, and is also used to control the cargo sorting equipment to carry and sort multiple cargoes according to the cargo sorting strategy, which solves the problem of low management efficiency when managing the cargo stored in the warehouse in the related technology. The cargo image of the cargo stored in the warehouse is obtained through the camera module 301, and the cargo image is pre-processed to obtain the cargo image to be detected. The cargo image to be detected is input into the target detection model of the target detection module 302, and then the cargo recognition result is output. Finally, the cargo sorting strategy is generated based on the cargo recognition result and the preset cargo storage rules, and the management module 303 is used to manage the cargo, thereby achieving the effect of improving the cargo sorting efficiency and reducing the labor cost.

[0067] The embodiment of the present application also provides a storage management device for goods. It should be noted that the storage management device for goods in the embodiment of the present application can be used to execute the storage management method for goods provided in the embodiment of the present application. The storage management device for goods provided in the embodiment of the present application is introduced below.

[0068] Figure 4 is a schematic diagram of a storage management device for goods provided according to an embodiment of the present application, such as Figure 4 As shown, the device includes: a first acquisition unit 40, a second acquisition unit 41, and a third acquisition unit 42.

[0069] The first acquisition unit 40 is used to acquire cargo images of a plurality of cargoes stored in the warehouse, and pre-process the cargo images to obtain cargo images to be detected, wherein the cargo images refer to images of a plurality of cargoes;

[0070] The second acquisition unit 41 is used to acquire a target detection model, input the image of the goods to be detected into the target detection model, and output a goods recognition result, wherein the target detection model is obtained by training the target detection algorithm, and the goods recognition result includes the goods type, the goods name and the location information of N goods, where N is a positive integer;

[0071] The third acquisition unit 42 is used to acquire the cargo storage rules and generate the cargo sorting strategy according to the cargo identification result and the cargo storage rules.

[0072] The storage management device for goods provided in the embodiment of the present application obtains goods images of multiple goods stored in the warehouse through the first acquisition unit 40, pre-processes the goods images, and obtains the goods images to be detected, wherein the goods images refer to the images of multiple goods; the second acquisition unit 41 obtains the target detection model, inputs the goods images to be detected into the target detection model, and outputs the goods recognition result, wherein the target detection model is obtained by training the target detection algorithm, and the goods recognition result includes the goods types and goods names of N goods and the location information of N goods, where N is a positive integer; the third acquisition unit 42 obtains the goods storage rules, and generates the goods sorting strategy according to the goods recognition results and the goods storage rules, so as to solve the problem of low management efficiency when managing goods stored in the warehouse in the related art, by obtaining the goods images of the goods stored in the warehouse, pre-processing the goods images to obtain the goods images to be detected, inputting the goods images to be detected into the target detection model and outputting the goods recognition results, and finally generating the goods sorting strategy based on the goods recognition results and the preset goods storage rules, thereby achieving the effect of improving the goods sorting efficiency and reducing the labor cost.

[0073] Optionally, in the storage management device for goods provided in the embodiment of the present application, the second acquisition unit 41 includes: the target detection model is trained in the following manner: a first acquisition module, used to acquire the name of goods, the type of goods and the historical images of goods stored in the warehouse within the historical time period; a second acquisition module, used to acquire the initial detection model, and replace the first feature extraction layer in the initial detection model with the second feature extraction layer to obtain a preset detection model, wherein the second feature extraction layer is used to identify features in the historical goods images whose area is less than the area threshold, and the first feature extraction layer is used to identify features in the historical goods images whose area is greater than or equal to the area threshold and features whose area is less than the area threshold; a composition module, used to constitute a model training set by the name of goods, the type of goods and the historical goods images, and use the model training set to train the preset detection model to obtain the target detection model, wherein the target detection model includes a target feature extraction layer, and the target feature extraction layer is trained by the second feature extraction layer.

[0074] Optionally, in the storage management device for goods provided in the embodiment of the present application, the second acquisition unit 41 includes: a first input module, used to input the image of goods to be detected into a pre-classification model, and the pre-classification model extracts goods features of the image of goods to be detected to obtain M goods features, wherein each goods feature refers to a goods feature of at least one goods, and M is a positive integer; a second input module, used to input the M goods features into a goods recognition model, and the goods recognition model recognizes each goods feature to obtain a goods recognition result.

[0075] Optionally, in the storage management device for goods provided in the embodiment of the present application, the second acquisition unit 41 includes: a third acquisition module, used to obtain an area threshold, and segment the image of the goods to be detected according to the area threshold to obtain M segmented goods images; an extraction module, used to perform feature extraction on the M segmented goods images by the second feature extraction layer to obtain M goods features.

[0076] Optionally, in the storage management device for goods provided in the embodiment of the present application, the device also includes: an extraction unit, used to extract location information of N goods from the goods identification results before obtaining the goods storage rules, and verify the location information of the N goods according to the goods storage rules; a generation unit, used to generate prompt information if there is at least one goods whose location information does not comply with the goods storage rules, and execute the step of generating a goods sorting strategy according to the goods identification results and the goods storage rules, wherein the prompt information is used to prompt the staff to sort the N goods; an execution unit, used to stop executing the step of generating a goods sorting strategy according to the goods identification results and the goods storage rules if the location information of the N goods complies with the goods storage rules.

[0077] Optionally, in the storage management device for goods provided in the embodiment of the present application, the third acquisition unit 41 includes: a fourth acquisition module, used to acquire the goods types of N goods from the goods identification results, obtain N goods types, and determine the storage rules of the N goods types from the goods storage rules; a fifth acquisition module, used to acquire the warehouse structure diagram of the warehouse, input the warehouse structure diagram and the storage rules of N goods types into the allocation scheduling algorithm, and output the goods sorting strategy.

[0078] Optionally, in the storage management device for goods provided in an embodiment of the present application, the device also includes: a sending unit, which is used to send the cargo sorting strategy to the cargo sorting equipment after generating the cargo sorting strategy according to the cargo identification results and the cargo storage rules, and control the cargo sorting equipment to transport and sort N goods according to the cargo sorting strategy.

[0079] The storage management device for the above-mentioned goods includes a processor and a memory. The above-mentioned first acquisition unit 40, second acquisition unit 41, third acquisition unit 42, etc. are all stored in the memory as program units, and the processor executes the above-mentioned program units stored in the memory to realize corresponding functions.

[0080] The processor includes a kernel, which retrieves the corresponding program unit from the memory. One or more kernels can be set, and the problem of low management efficiency in the management of goods stored in the warehouse in the related technology can be solved by adjusting the kernel parameters.

[0081] The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM, and the memory includes at least one memory chip.

[0082] An embodiment of the present invention provides a computer storage medium, which is used to store a program, wherein when the program is running, the device where the computer storage medium is located is controlled to execute a storage management method for goods.

[0083] Figure 5 is a schematic diagram of an electronic device provided according to an embodiment of the present application, such as Figure 5 As shown, an embodiment of the present invention provides an electronic device, the electronic device 50 includes a processor, a memory, and a program stored in the memory and executable on the processor, the processor is used to execute computer-readable instructions, wherein the computer-readable instructions execute a storage management method for goods when they are executed. The device in this article can be a server, a PC, a PAD, a mobile phone, etc.

[0084] The present application also provides a computer program product, including a computer program, which, when executed by a processor, implements the steps of a method for storage management of goods in various embodiments of the present application.

[0085] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented in one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that include computer-usable program code.

[0086] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0087] These computer program instructions may also be stored in a computer readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture including an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0088] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process in the computer or other programmable device. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0089] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0090] The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.

[0091] Computer readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include temporary computer readable media (transitory media), such as modulated data signals and carrier waves.

[0092] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0093] The above are only embodiments of the present application and are not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included within the scope of the claims of the present application.

Claims

1. A method for storage management of goods, characterized in that: include: Acquire cargo images of a plurality of cargoes stored in a warehouse, and pre-process the cargo images to obtain cargo images to be detected, wherein the cargo images refer to images of the plurality of cargoes; Obtain a target detection model, input the to-be-detected cargo image into the target detection model, and output a cargo recognition result, wherein the target detection model is obtained by training a target detection algorithm, and the cargo recognition result includes cargo types, cargo names, and location information of N cargoes, where N is a positive integer; Obtain cargo storage rules, and generate cargo sorting strategies according to the cargo identification results and the cargo storage rules.

2. The method according to claim 1, characterized in that: The target detection model is trained in the following way: Obtain the cargo name, cargo type and historical cargo image of the historical cargo stored in the warehouse within the historical time period; Acquire an initial detection model, and replace a first feature extraction layer in the initial detection model with a second feature extraction layer to obtain a preset detection model, wherein the second feature extraction layer is used to identify features in the historical cargo images whose areas are less than an area threshold, and the first feature extraction layer is used to identify features in the historical cargo images whose areas are greater than or equal to the area threshold and features whose areas are less than the area threshold; The cargo name, the cargo type and the historical cargo image constitute a model training set, and the preset detection model is trained using the model training set to obtain the target detection model, wherein the target detection model includes a target feature extraction layer, and the target feature extraction layer is obtained by training the second feature extraction layer.

3. The method according to claim 1, characterized in that The target detection model includes a pre-classification model and a cargo recognition model. The cargo image to be detected is input into the target detection model, and the cargo recognition result output includes: Input the cargo image to be detected into the pre-classification model, and extract cargo features of the cargo image to be detected by the pre-classification model to obtain M cargo features, wherein each cargo feature refers to a cargo feature of at least one cargo, and M is a positive integer; The M cargo features are input into the cargo identification model, and the cargo identification model identifies each cargo feature to obtain the cargo identification result.

4. The method according to claim 3, characterized in that The pre-classification model comprises a second feature extraction layer, and the pre-classification model extracts the cargo features of the cargo image to be detected, and the obtained M cargo features include: Acquire an area threshold, and segment the image of the goods to be detected according to the area threshold to obtain M segmented goods images; The second feature extraction layer performs feature extraction on the M segmented cargo images to obtain the M cargo features.

5. The method according to claim 1, characterized in that Before obtaining the cargo storage rules, the method further includes: Extracting the location information of the N goods from the goods identification result, and verifying the location information of the N goods according to the goods storage rule; If the location information of at least one cargo does not conform to the cargo storage rule, a prompt message is generated, and the step of generating a cargo sorting strategy according to the cargo identification result and the cargo storage rule is performed, wherein the prompt message is used to prompt the staff to sort the N cargoes; If the location information of the N goods conforms to the goods storage rule, the step of generating a goods sorting strategy according to the goods identification result and the goods storage rule is stopped.

6. The method according to claim 1, characterized in that Generating a cargo sorting strategy according to the cargo identification result and the cargo storage rule includes: Acquire the cargo types of the N cargoes from the cargo identification results to obtain N cargo types, and determine storage rules for the N cargo types from the cargo storage rules; A warehouse structure diagram of the warehouse is obtained, the warehouse structure diagram and the storage rules of the N types of goods are input into an allocation and scheduling algorithm, and the goods sorting strategy is output.

7. The method according to claim 1, characterized in that After generating a cargo sorting strategy according to the cargo identification result and the cargo storage rule, the method further includes: The cargo sorting strategy is sent to the cargo sorting equipment, and the cargo sorting equipment is controlled to carry and sort the N cargoes according to the cargo sorting strategy.

8. A cargo handling system, characterized in that: include: A camera module is used to capture images of multiple goods stored in the warehouse, pre-process the images to obtain images of goods to be detected, and send the images of goods to be detected to the target detection module; The target detection module is used to input the image of the goods to be detected into the target detection model and output the goods recognition result when receiving the image of the goods to be detected, and is also used to obtain the goods storage rules, generate the goods sorting strategy according to the goods recognition result and the goods storage rules, and send the goods sorting strategy to the management module; The management module is in communication with the cargo sorting device and is used to display the cargo sorting strategy to the staff and to control the cargo sorting device to carry and sort the multiple cargoes according to the cargo sorting strategy.

9. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the cargo storage management method according to any one of claims 1 to 7 is implemented.

10. An electronic device, characterized in that: It includes one or more processors and a memory, wherein the memory is used to store one or more programs, wherein when the one or more programs are executed by the one or more processors, the one or more processors implement the storage management method of goods described in any one of claims 1 to 7.

Images