Warehouse monitoring method, device, electronic equipment and storage medium

Abstract

The present invention discloses a warehouse monitoring method, device, electronic device and storage medium. The warehouse monitoring method comprises: firstly obtaining scene image information of on-duty personnel in the warehouse; and inputting the scene image information into a first neural network model for screening to determine the behavior image information of the on-duty personnel; wherein the neural network model is trained by a plurality of the behavior image information in a historical time; then determining the behavior scoring result of the behavior image information according to the search behavior image information, and determining the behavior management plan of the on-duty personnel according to the behavior scoring result; finally, outputting the behavior management plan to schedule and manage the on-duty personnel. Thus, when monitoring warehouse personnel, the irregular operations of various personnel can be reduced, the damage to goods can be avoided, and the safety and efficiency of monitoring can be improved.

Description

Technical Field

[0001] The present invention relates to the technical field of warehouse management, and in particular to a warehouse monitoring method, device, electronic equipment and storage medium. Background Art

[0002] At present, with the rapid development of the national economy, China's logistics industry has maintained a rapid growth rate, the logistics system has been continuously improved, and the industry operation has become increasingly mature and standardized. With the development of the logistics industry, the task of monitoring and managing warehouses has become increasingly arduous, especially for the various behaviors of warehouse personnel, which are difficult to accurately monitor, such as cargo handling, parcel sorting, etc., resulting in the current monitoring of warehouse personnel. Due to the irregular operations of various personnel, not only are goods easily damaged, but also safety hazards will be caused. Summary of the invention

[0003] In a first aspect, the main purpose of the present invention is to provide a warehouse monitoring method, comprising:

[0004] Obtain scene image information of personnel on duty in the warehouse;

[0005] Inputting the scene image information into a first neural network model for screening to determine the behavior image information of the on-duty personnel; wherein the neural network model is trained by a plurality of behavior image information in a historical period;

[0006] Determine a behavior scoring result of the behavior image information according to the behavior image information, and determine a behavior management plan for the on-duty personnel according to the behavior scoring result;

[0007] The behavior management plan is output to perform scheduling management on the on-duty personnel.

[0008] Optionally, the step of inputting the scene image information into a first neural network model for feature extraction and screening to determine the behavior image information of the on-duty personnel includes:

[0009] Detecting the human body region coordinates of the on-duty personnel in the scene image information according to the first neural network model;

[0010] The scene image information is cropped according to the human body region coordinates to obtain a human body region image;

[0011] The human body region image is screened to determine the behavior image information.

[0012] Optionally, determining a behavior scoring result of the behavior image information according to the behavior image information, and determining a behavior management plan for the on-duty personnel according to the behavior scoring result, includes:

[0013] Determine the corresponding behavior scoring result according to the behavior image information of the on-duty personnel, and construct a scoring matrix according to the behavior scoring result; the scoring matrix includes a plurality of scoring sets of the on-duty personnel;

[0014] Calculating the similarity between the score set of the on-the-job personnel and the standard score set according to the score matrix;

[0015] A behavior management plan for the on-duty personnel is determined based on the similarity.

[0016] Optionally, determining the behavior scoring result of the on-duty personnel according to the behavior image information of the on-duty personnel, and constructing a scoring matrix according to the behavior scoring result, includes:

[0017] Inputting the behavior image information into the second neural network model to determine the three-dimensional posture corresponding to the on-duty personnel and the key points of the objects associated with the on-duty personnel;

[0018] Determining a behavior score result of the on-duty personnel according to the three-dimensional posture and the key points of the object;

[0019] The scoring matrix is ​​constructed according to the behavior scoring results.

[0020] Optionally, calculating the similarity between the score set of the on-the-job personnel and the standard score set according to the score matrix includes:

[0021] Determining a score set for the on-the-job personnel according to the score matrix;

[0022] The information entropy formula is used to calculate the information entropy value of the score set corresponding to the on-the-job personnel;

[0023] The similarity between the score set of the on-the-job personnel and the standard score set is determined according to the information entropy value.

[0024] Optionally, determining the similarity between the score set of the on-the-job personnel and the standard score set according to the information entropy value includes:

[0025] Determining whether the information entropy value is greater than a preset entropy value;

[0026] When the information entropy value is greater than the preset entropy value, the similarity between the score set of the on-the-job personnel and the standard score set is the standard similarity;

[0027] When the information entropy value is less than the preset entropy value, the similarity between the score set of the on-the-job personnel and the standard score set is a non-standard similarity.

[0028] Optionally, determining a behavior management plan for the on-duty personnel according to the similarity includes:

[0029] Determine the similarity type corresponding to the on-the-job personnel;

[0030] When the similarity type corresponding to the on-duty personnel is the standard similarity, determining the behavior management plan of the on-duty personnel to be the standard management plan;

[0031] When the similarity type corresponding to the on-duty personnel is a non-standard similarity, it is determined that the behavior management plan for the on-duty personnel is a violation warning plan.

[0032] In a third aspect, an embodiment of the present invention provides a warehouse monitoring device, including:

[0033] An acquisition module is used to obtain scene image information of personnel on duty in the warehouse;

[0034] A screening module, used for inputting the scene image information into a first neural network model for screening to determine the behavior image information of the on-duty personnel; wherein the neural network model is trained by a plurality of behavior image information in a historical period;

[0035] A determination module, used to determine a behavior scoring result of the behavior image information according to the behavior image information, and determine a behavior management plan for the on-duty personnel according to the behavior scoring result;

[0036] The output module is used to output the behavior management plan to perform scheduling management on the on-duty personnel.

[0037] In a third aspect, an embodiment of the present invention provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the above-mentioned warehouse monitoring method when executing the computer program.

[0038] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the warehouse monitoring method as described above are implemented.

[0039] The above solution of the present invention includes at least the following beneficial effects:

[0040] The warehouse monitoring method provided by the present invention first obtains the scene image information of the on-duty personnel in the warehouse; and inputs the scene image information into the first neural network model for screening to determine the behavior image information of the on-duty personnel; wherein the neural network model is trained by multiple pieces of the behavior image information in the historical time; then the behavior scoring result of the behavior image information is determined according to the behavior image information, and the behavior management plan of the on-duty personnel is determined according to the behavior scoring result; finally, the behavior management plan is output to schedule and manage the on-duty personnel. Therefore, when monitoring the warehouse personnel, the irregular operations of various personnel can be reduced, the damage to the goods can be avoided, and the safety and efficiency of the monitoring can be improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the structures shown in these drawings without paying creative work.

[0042] Figure 1 A schematic diagram of the overall process of the warehouse monitoring method provided by an embodiment of the present invention;

[0043] Figure 2 A schematic diagram of a specific flow chart of step S20 provided in an embodiment of the present invention;

[0044] Figure 3 A schematic diagram of a specific flow chart of step S30 provided in an embodiment of the present invention;

[0045] Figure 4 A schematic diagram of a specific flow chart of step S31 provided in an embodiment of the present invention;

[0046] Figure 5 A schematic diagram of a specific flow chart of step S32 provided in an embodiment of the present invention;

[0047] Figure 6 Another schematic diagram of the process of step S30 provided in an embodiment of the present invention;

[0048] Figure 7 A schematic diagram of a specific flow chart of step S33 provided in an embodiment of the present invention;

[0049] Figure 8 A structural block diagram of a warehouse monitoring device provided by an embodiment of the present invention;

[0050] Fig. 9 This is a structural block diagram of an electronic device provided by an embodiment of the present invention.

[0051] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0052] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0053] The terms "first", "second", "third", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish different objects, rather than to describe a specific order. In addition, the term "comprising" and any variation thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device comprising a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units that are not listed, or may optionally include other steps or units inherent to these processes, methods, products or devices.

[0054] First, the scheme of the embodiment of the present application is introduced by way of examples with reference to the relevant drawings.

[0055] like Figure 1 As shown, a specific embodiment of the present invention provides a warehouse monitoring method, comprising:

[0056] S10: Obtain scene image information of on-duty personnel in the warehouse.

[0057] In this embodiment, scene image information can be acquired through an image acquisition device (such as a camera, etc.), and scene image information in the warehouse can be acquired in real time or at preset time intervals, and the scene image information can be acquired for multiple different images of each on-duty personnel at different times, or can be acquired at the same time; the image acquisition device can be remotely controlled and connected to the server through the Internet of Things and other means, and can output the acquired scene image information to the server, so that the server can analyze the on-duty personnel according to the scene image information, and after the analysis, it can be output to a visualization interface for display for management personnel to view; it can be understood that on-duty personnel can be represented as couriers, sorters, porters, etc. in the warehouse, and therefore, the scene image information can include a variety of scene image information of on-duty personnel in the warehouse, such as cargo handling images, package sorting images, as well as overweight handling images, violent sorting images, etc.

[0058] S20, inputting the scene image information into the first neural network model for screening to determine the behavior image information of the on-duty personnel; wherein the neural network model is trained by multiple behavior image information in historical time.

[0059] In this embodiment, the neural network model can be a ResNet residual network model or other neural network models, and the historical time can be within two months or within three months. Multiple behavioral image information determined in the historical time is input into the above-mentioned neural network model for training. After the training is completed, the neural network model can be used to extract features of the scene image information collected at the current time, thereby determining the behavioral image information of the on-duty personnel; optionally, the behavioral image information can be images of on-duty personnel carrying goods or images of on-duty personnel sorting goods. The neural network model contains multiple semantic information. After feature extraction of the scene image information, behavioral image information with different semantic information can be determined.

[0060] like Figure 2 As shown, the scene image information is input into the first neural network model for screening to determine the behavior image information of the on-duty personnel, including:

[0061] S21, detecting the human body region coordinates of the on-duty personnel in the scene image information according to the first neural network model;

[0062] S22, cutting the scene image information according to the human body region coordinates to obtain a human body region image;

[0063] S23, screening the human body region image to determine the behavior as image information.

[0064] Among them, the scene image information includes a scene image of a warehouse and images of multiple on-duty personnel. The first neural network model can be a ResNet-50 neural network model. The body area coordinates of the on-duty personnel are determined by the first neural network model. Since the bodies of the on-duty personnel are in contact with the goods when they are carrying or sorting goods, the body area coordinates may also include the cargo area coordinates. The pre-trained neural network model can be used to determine the coordinate information of the human body area and the cargo area as a whole, thereby cropping the corresponding human body area image through the human body area coordinates. Therefore, the human body area image also includes the corresponding cargo area image. By extracting features from the human body area image, a variety of different behavioral image information can be determined.

[0065] For example, a certain scene image collected includes four on-duty personnel A, B, C, and D. A carries 1 piece of cargo on his back, B carries 1 piece of cargo on his shoulders, C carries 1 piece of cargo with both hands, and D carries 5 pieces of cargo. Therefore, when cropping, the corresponding areas of A, B, C, and D can be cropped respectively to obtain the human body area images of A, B, C, and D respectively. Through the semantic information in the neural network model, it can be determined that A is the back running behavior image, B is the shoulder running behavior image, C is the holding running behavior image, and D is the overweight carrying image. Therefore, the corresponding scoring results can be determined for the behavioral image information of the four on-duty personnel A, B, C, and D respectively.

[0066] S30. Determine a scoring result of the behavior image information according to the information entropy index, and determine a behavior management plan for on-the-job personnel according to the scoring result.

[0067] In this embodiment, information entropy is expressed as a quantitative indicator of information content, that is, the probability of occurrence of specific information; when the information entropy index is higher, the corresponding on-the-job personnel's scoring result is higher, and when the information entropy index is lower, the corresponding on-the-job personnel's scoring result is lower; therefore, the scoring result of each on-the-job personnel's behavior image information can be determined by the information entropy index. When there are multiple behavior image information of an on-the-job personnel, the corresponding information entropy index can be determined through the behavior image information; it can be understood that each on-the-job personnel can only determine a unique scoring result, and determine the corresponding behavior management plan based on the unique scoring result; the behavior management plan means that the behavior of the on-the-job personnel is managed by the corresponding plan to avoid irregular operating behavior of the on-the-job personnel.

[0068] like Figure 3 As shown, the above-mentioned determining the behavior scoring result of the behavior image information according to the behavior image information, and determining the behavior management plan of the on-duty personnel according to the behavior scoring result, includes:

[0069] S31, determining corresponding behavior scoring results according to the behavior image information of the on-duty personnel, and constructing a scoring matrix according to the behavior scoring results; the scoring matrix includes a score set of multiple on-duty personnel;

[0070] S32. Calculate the similarity between the scoring set of the on-the-job personnel and the standard scoring set according to the scoring matrix;

[0071] S33. Determine a behavior management plan for on-the-job personnel based on the similarity.

[0072] Among them, the scoring matrix is ​​represented as a set of complex numbers or real numbers arranged in a rectangular array. Each on-duty personnel may correspond to multiple behavioral image information, and each behavioral image information may obtain a behavioral score. By determining the scoring set of each on-duty personnel, the scoring sets of multiple on-duty personnel can be used to construct a scoring matrix. When determining the behavior of a certain on-duty personnel, the scoring set of a certain on-duty personnel can be determined through the scoring matrix, and the similarity between the corresponding scoring set and the standard scoring set can be calculated, thereby determining the corresponding behavior management plan, and then the behavior of the on-duty personnel can be corrected by the management personnel; it can be understood that the standard scoring set is pre-set, and can be obtained by obtaining the standard behavioral image information of the on-duty personnel for scoring, and then determining the obtained standard scoring set.

[0073] For example, if the score set k1 of a certain on-the-job employee A is determined to be [40, 60, 70, 30, 50, 80, 90, 70], the standard score set k v is [100, 100, 100, 100, 100, 100, 90, 100]. By determining k1 and k v The similarity between them can be used to determine the behavior of on-the-job personnel A, and then determine the corresponding behavior management plan; optionally, when determining the similarity between the score set of on-the-job personnel and the standard score set, the cosine theorem can be used to calculate the similarity; the cosine theorem indicates that the cosine value of the angle between two vectors in the vector space is used as a measure of the size of the difference between two individuals. The closer the cosine value is to 1, the closer the angle is to 0 degrees, and the higher the similarity.

[0074] like Figure 4 As shown, the behavior scoring results of the on-duty personnel are determined based on the behavior image information of the on-duty personnel, and a scoring matrix is ​​constructed based on the behavior scoring results, including:

[0075] S311, inputting the behavior image information into the second neural network model to determine the corresponding three-dimensional posture of the on-duty personnel and the key points of the objects associated with the on-duty personnel;

[0076] S312, determining the behavior score result of the on-duty personnel according to the three-dimensional posture and the key points of the object;

[0077] S313. Construct a scoring matrix based on the behavior scoring results.

[0078] Among them, the three-dimensional posture represents the movement posture characteristics of the body, the object key points are represented as the location points of the goods, and the second neural network model can be a DensePose-RCNN neural network model. The second neural network model can determine the score value of each three-dimensional posture and the score value of the object key point. By performing weighted average calculation on the score value of the three-dimensional posture and the score value of the object key point, the behavior score result corresponding to the behavior image information can be determined; it can be understood that each behavior image information can correspond to a unique behavior score result, and by determining the behavior score results corresponding to multiple behavior image information of each on-the-job personnel, a scoring matrix corresponding to multiple on-the-job personnel can be constructed.

[0079] Specifically, the three-dimensional posture may have a variety of different motion posture features, and the key points of the object may also have different position points. When the weighted average is used to calculate the behavior scoring result, the weight of the three-dimensional posture may be made higher, or the weight of the key points of the object may be made higher. It can be understood that each behavior image information generates a corresponding behavior scoring result. When an on-duty personnel has multiple behavior image information, multiple corresponding behavior scoring results may be generated. By calculating multiple behavior scoring results of all on-duty personnel in the warehouse, the constructed scoring matrix is ​​more comprehensive, and the scoring set of each on-duty personnel can be accurately determined based on the scoring matrix in the future.

[0080] Alternatively, the scoring matrix can be taken as follows:

[0081]

[0082] Where n represents the number of on-duty personnel, m represents the number of behaviors, and the set of on-duty personnel is represented by l = {l1, l2..., l n}, the behavior set is represented by k = {k1, k2..., k m}, the behavior score result is expressed as w mn Thus, through the above-mentioned scoring matrix, the scoring set of each on-the-job personnel can be determined, so that the similarity between the scoring set of the on-the-job personnel and the annotated scoring set can be determined later according to the scoring matrix.

[0083] For example, if a worker l1 throws a cargo, we can determine the corresponding three-dimensional posture and object key points, determine the action posture and cargo location, and obtain the throwing behavior k1 of the worker l1, thereby determining the behavior score result w of the throwing behavior 11 , different behavior scoring results w can be determined through the different behavior image information of the on-the-job personnel l1 mn , thus obtaining the corresponding score set.

[0084] Furthermore, the similarity between the score set of the on-the-job personnel and the standard score set is calculated based on the score matrix, including:

[0085] S321. Determine the score set of the on-the-job personnel according to the score matrix;

[0086] S322, using the information entropy formula to calculate the information entropy value of the corresponding score set of the on-the-job personnel;

[0087] S323. Determine the similarity between the on-the-job personnel's score set and the standard score set based on the information entropy value.

[0088] In this embodiment, each on-the-job personnel has a corresponding scoring set. The scoring set of the on-the-job personnel is input into the information entropy formula for calculation to determine the corresponding information entropy value, thereby the similarity between the scoring set of the on-the-job personnel and the standard scoring set can be determined by the information entropy value. It can be understood that the information entropy values ​​of all on-the-job personnel in the scoring matrix can be calculated simultaneously through the scoring matrix, and the corresponding similarity of each on-the-job personnel can be determined at the same time, thereby the corresponding behavior management plans can be determined and output by the similarities of all on-the-job personnel; it can be understood that the smaller the information entropy value, the smaller the difference between the scoring set and the standard scoring set, and the greater the similarity between the two; the larger the information entropy value, the larger the difference between the scoring set and the standard scoring set, and the smaller the similarity between the two.

[0089] Optionally, the above-mentioned determining the similarity between the score set of the on-the-job personnel and the standard score set according to the information entropy value includes:

[0090] 301. Determine whether the information entropy value is greater than a preset entropy value;

[0091] 302. When the information entropy value is greater than the preset entropy value, the similarity between the on-the-job personnel's score set and the standard score set is a non-standard similarity;

[0092] 303. When the information entropy value is less than the preset entropy value, the similarity between the on-the-job personnel's score set and the standard score set is the standard similarity.

[0093] Among them, the standard similarity can be a range value of 0.5-1, and the non-standard similarity can be a range value of 0-0.5; from the above, it can be seen that the similarity between the score set of the on-the-job personnel and the standard score set can be calculated by the cosine theorem, and can also be judged by information entropy. The preset entropy value can be pre-set, that is, when the information entropy value of a certain on-the-job personnel is calculated to be greater than the preset entropy value, it can be determined that the similarity of the on-the-job personnel is non-standard similarity, indicating that the behavior of the on-the-job personnel is normal behavior information; when the information entropy value of a certain on-the-job personnel is calculated to be less than the preset entropy value, it can be determined that the similarity of the on-the-job personnel is standard similarity, indicating that the behavior of the on-the-job personnel is abnormal behavior information; it can be understood that when the number of abnormal behaviors of the on-the-job personnel is relatively large, it means that the probability of abnormal behavior is relatively large, so the corresponding information entropy value calculated is larger; when the number of normal behaviors of the on-the-job personnel is relatively large, it means that the probability of normal behavior is relatively large, so the corresponding information entropy value calculated is smaller.

[0094] Optionally, a behavior management plan for on-the-job personnel is determined based on the similarity, including:

[0095] S331, determining the similarity type corresponding to the on-the-job personnel;

[0096] S332: When the similarity type corresponding to the on-duty personnel is the standard similarity, determining the behavior management plan of the on-duty personnel as the standard management plan;

[0097] S333. When the similarity type corresponding to the on-duty personnel is non-standard similarity, the behavior management plan for the on-duty personnel is determined to be a violation warning plan.

[0098] In this embodiment, the similarity types include standard similarity and non-standard similarity, the standard management plan may include a reward plan, etc., and the violation warning plan may include a persuasion plan, a correction plan, etc.; when it is determined that the on-the-job personnel are of standard similarity, the behavior management plan of the on-the-job personnel can be set as the standard management plan, so that the standard management plan can be used to enable the management personnel to give real-time rewards to the on-the-job personnel in the future; when it is determined that the on-the-job personnel are of non-standard similarity, the behavior management plan of the on-the-job personnel can be set as the violation warning plan, so that the standard management plan can be used to enable the management personnel to correct or persuade the on-the-job personnel in the future.

[0099] For example, when a certain on-duty personnel often carries a lot of goods at one time and many scene image information are captured, after the behavioral image information is determined through feature extraction, the corresponding information entropy value calculated through information entropy is relatively large, so it can be determined that the difference between his behavior and the standard behavior is relatively large. Therefore, it can be determined that the corresponding behavior management plan is a violation warning plan, which can be pushed to the management platform, etc., so that the on-duty personnel can be corrected or persuaded by the management personnel to avoid safety hazards and improve the standardized operations of on-duty personnel.

[0100] S40. Output the behavior management plan to perform scheduling management on-duty personnel.

[0101] In this embodiment, the behavior management plan includes a standard management plan and a violation warning plan. Therefore, the server can output the behavior management plan determined by each on-duty personnel to a visual interface (such as a display, mobile phone app), so that the management personnel can view the behavior management plan of the on-duty personnel through the visual interface, and adopt different behavior management plans to manage each on-duty personnel, thereby improving the monitoring and management efficiency of the warehouse, and can monitor each on-duty personnel in real time more timely, improve the safety of on-duty personnel, and avoid the problem of damage to goods.

[0102] The warehouse monitoring method provided by the present invention first obtains the scene image information of the on-duty personnel in the warehouse; and inputs the scene image information into the first neural network model for screening to determine the behavior image information of the on-duty personnel; wherein the neural network model is trained by multiple behavior image information in historical time; then the behavior scoring result of the behavior image information is determined according to the behavior image information, and the behavior management plan of the on-duty personnel is determined according to the behavior scoring result; finally, the behavior management plan is output to schedule and manage the on-duty personnel. Therefore, when monitoring the warehouse personnel, the irregular operations of various personnel can be reduced, the damage to the goods can be avoided, and the safety and efficiency of the monitoring can be improved.

[0103] In a third aspect, an embodiment of the present invention provides a warehouse monitoring device 10, comprising:

[0104] An acquisition module 11 is used to acquire scene image information of on-duty personnel;

[0105] The first determination module 12 is used to input the scene image information into the first neural network model for screening, and determine the behavior image information of the on-duty personnel; wherein the neural network model is trained by multiple behavior image information in historical time;

[0106] The second determination module 13 determines the behavior scoring result of the behavior image information according to the behavior image information, and determines the behavior management plan of the on-duty personnel according to the behavior scoring result;

[0107] The output module 14 is used to output the behavior management plan to perform scheduling management on-duty personnel.

[0108] The warehouse monitoring method provided by the present invention first obtains the scene image information of the on-duty personnel in the warehouse; and inputs the scene image information into the first neural network model for screening to determine the behavior image information of the on-duty personnel; wherein the neural network model is trained by multiple behavior image information in historical time; then the behavior scoring result of the behavior image information is determined according to the behavior image information, and the behavior management plan of the on-duty personnel is determined according to the behavior scoring result; finally, the behavior management plan is output to schedule and manage the on-duty personnel. Therefore, when monitoring the warehouse personnel, the irregular operations of various personnel can be reduced, the damage to the goods can be avoided, and the safety and efficiency of the monitoring can be improved.

[0109] It should be noted that the warehouse monitoring device 10 provided in the specific embodiment of the present invention is a device corresponding to the above-mentioned warehouse monitoring method, and all embodiments of the above-mentioned warehouse monitoring method are applicable to the warehouse monitoring device 10. The above-mentioned warehouse monitoring device 10 embodiments all have corresponding modules corresponding to the steps in the above-mentioned warehouse monitoring method, and can achieve the same or similar beneficial effects. In order to avoid excessive repetition, each module in the warehouse monitoring device 2 will not be described in detail here.

[0110] like Fig. 9 As shown, a specific embodiment of the present invention further provides an electronic device 20, including a memory 202, a processor 201, and a computer program stored in the memory 202 and executable on the processor 201, wherein the processor 201 implements the steps of the above-mentioned warehouse monitoring method when executing the computer program.

[0111] Specifically, the processor 201 is used to call the computer program stored in the memory 202 to perform the following steps:

[0112] Obtain scene image information of personnel on duty in the warehouse;

[0113] Input the scene image information into the first neural network model for screening to determine the behavior image information of the on-duty personnel; wherein the neural network model is trained by multiple behavior image information in historical time;

[0114] Determine a behavior scoring result of the behavior image information according to the behavior image information, and determine a behavior management plan for the on-the-job personnel according to the behavior scoring result;

[0115] Output the behavior management plan to schedule and manage on-duty personnel.

[0116] Optionally, the processor 201 inputs the scene image information into a first neural network model for screening to determine the behavior image information of the on-duty personnel, including:

[0117] Detecting the human body region coordinates of the on-duty personnel in the scene image information according to the first neural network model;

[0118] The scene image information is cropped according to the human body region coordinates to obtain a human body region image;

[0119] The human body region images are screened to determine the behaviors as image information.

[0120] Optionally, the processor 201 determines a behavior scoring result of the behavior image information according to the behavior image information, and determines a behavior management plan for the on-duty personnel according to the behavior scoring result, including:

[0121] Determine the comprehensive behavior score result corresponding to the on-duty personnel according to the behavior image information of the on-duty personnel, and construct a score matrix according to the behavior score result and the comprehensive behavior score; the score matrix includes the score set of multiple on-duty personnel;

[0122] According to the rating matrix, calculate the similarity between the rating set of the on-the-job personnel and the standard rating set;

[0123] Determine the behavior management plan for on-the-job personnel based on the similarity.

[0124] Optionally, the processor 201 determines the behavior score result of the on-duty personnel according to the behavior image information of the on-duty personnel, and constructs a score matrix according to the behavior score result, including:

[0125] Inputting the behavior image information into the second neural network model to determine the corresponding three-dimensional posture of the on-duty personnel and the key points of the objects associated with the on-duty personnel;

[0126] Determine the behavior score results corresponding to the on-duty personnel based on the three-dimensional posture and key points of the object;

[0127] A scoring matrix is ​​constructed based on the behavior scoring results.

[0128] Optionally, the processor 201 calculates the similarity between the score set of the on-the-job personnel and the standard score set according to the score matrix, including:

[0129] According to the scoring matrix, determine the scoring set of the on-the-job personnel;

[0130] The information entropy formula is used to calculate the information entropy value of the corresponding score set of the on-the-job personnel;

[0131] The similarity between the on-the-job personnel's scoring set and the standard scoring set is determined based on the information entropy value.

[0132] Optionally, the processor 201 determines the similarity between the score set of the on-the-job personnel and the standard score set according to the information entropy value, including:

[0133] Determine whether the information entropy value is greater than a preset entropy value;

[0134] When the information entropy value is greater than the preset entropy value, the similarity between the on-the-job personnel's score set and the standard score set is a non-standard similarity;

[0135] When the information entropy value is less than the preset entropy value, the similarity between the on-the-job personnel's score set and the standard score set is the standard similarity.

[0136] Optionally, the processor 201 determines the behavior management plan of the on-duty personnel according to the similarity, including:

[0137] Determine the similarity type corresponding to the on-the-job personnel;

[0138] When the similarity type corresponding to the on-the-job personnel is the standard similarity, the behavior management plan for the on-the-job personnel is determined to be the standard management plan;

[0139] When the similarity type corresponding to the on-duty personnel is non-standard similarity, the behavior management plan for the on-duty personnel is determined to be a violation warning plan.

[0140] That is, in a specific embodiment of the present invention, the processor 201 of the electronic device 20 implements the steps of the above-mentioned warehouse monitoring method when executing a computer program, thereby reducing irregular operations of various personnel when monitoring warehouse personnel, avoiding damage to goods, and improving the safety and efficiency of monitoring.

[0141] It should be noted that, since the processor 201 of the electronic device 20 implements the steps of the above-mentioned warehouse monitoring method when executing a computer program, all embodiments of the above-mentioned warehouse monitoring method are applicable to the electronic device 20 and can achieve the same or similar beneficial effects.

[0142] The computer-readable storage medium provided in the embodiment of the present invention stores a computer program. When the computer program is executed by the processor, it implements the various processes of the warehouse monitoring method or the application-side warehouse monitoring method provided in the embodiment of the present invention, and can achieve the same technical effect. To avoid repetition, it will not be repeated here.

[0143] A person skilled in the art can understand that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through a computer program, and the program can be stored in a computer-readable storage medium, and when the program is executed, it can include the processes of the embodiments of the above-mentioned methods. The storage medium can be a disk, an optical disk, a read-only memory (ROM) or a random access memory (RAM).

[0144] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "examples", "specific examples", or "some examples" means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner.

[0145] The above description is only a preferred embodiment of the present invention, and does not limit the patent scope of the present invention. All equivalent structural changes made by using the contents of the present invention specification and drawings under the concept of the present invention, or directly / indirectly applied in other related technical fields are included in the patent protection scope of the present invention.

Claims

1. A warehouse monitoring method, characterized in that: include: Obtain scene image information of personnel on duty in the warehouse; Inputting the scene image information into a first neural network model for screening to determine the behavior image information of the on-duty personnel; wherein the neural network model is trained by a plurality of behavior image information in a historical period; Determine a behavior scoring result of the behavior image information according to the behavior image information, and determine a behavior management plan for the on-duty personnel according to the behavior scoring result; Outputting the behavior management plan to perform scheduling management on the on-duty personnel; Determining a behavior scoring result of the behavior image information according to the behavior image information, and determining a behavior management plan for the on-duty personnel according to the behavior scoring result, includes: Determine the corresponding behavior scoring result according to the behavior image information of the on-duty personnel, and construct a scoring matrix according to the behavior scoring result; the scoring matrix includes a plurality of scoring sets of the on-duty personnel; Calculating the similarity between the score set of the on-the-job personnel and the standard score set according to the score matrix; A behavior management plan for the on-duty personnel is determined based on the similarity.

2. The warehouse monitoring method according to claim 1, characterized in that: The step of inputting the scene image information into the first neural network model for feature extraction and screening to determine the behavior image information of the on-duty personnel includes: Detecting the human body region coordinates of the on-duty personnel in the scene image information according to the first neural network model; The scene image information is cropped according to the human body region coordinates to obtain a human body region image; The human body region image is screened to determine the behavior image information.

3. The warehouse monitoring method according to claim 1, characterized in that: Determining the behavior scoring result of the on-duty personnel according to the behavior image information of the on-duty personnel, and constructing a scoring matrix according to the behavior scoring result, includes: Inputting the behavior image information into a second neural network model to determine the three-dimensional posture corresponding to the on-duty personnel and key points of objects associated with the on-duty personnel; Determining a behavior score result of the on-duty personnel according to the three-dimensional posture and the key points of the object; The scoring matrix is ​​constructed according to the behavior scoring results.

4. The warehouse monitoring method according to claim 1, characterized in that: The calculating, according to the scoring matrix, the similarity between the scoring set of the on-the-job personnel and the standard scoring set comprises: Determining a score set for the on-the-job personnel according to the score matrix; The information entropy formula is used to calculate the information entropy value of the score set corresponding to the on-the-job personnel; The similarity between the score set of the on-the-job personnel and the standard score set is determined according to the information entropy value.

5. The warehouse monitoring method according to claim 4, characterized in that: The determining the similarity between the score set of the on-the-job personnel and the standard score set according to the information entropy value includes: Determining whether the information entropy value is greater than a preset entropy value; When the information entropy value is greater than the preset entropy value, the similarity between the score set of the on-the-job personnel and the standard score set is a non-standard similarity; When the information entropy value is less than the preset entropy value, the similarity between the score set of the on-the-job personnel and the standard score set is the standard similarity.

6. The warehouse monitoring method according to claim 5, characterized in that: Determining the behavior management plan for the on-duty personnel according to the similarity includes: Determine the similarity type corresponding to the on-the-job personnel; When the similarity type corresponding to the on-duty personnel is the standard similarity, determining the behavior management plan of the on-duty personnel to be the standard management plan; When the similarity type corresponding to the on-duty personnel is a non-standard similarity, it is determined that the behavior management plan for the on-duty personnel is a violation warning plan.

7. A warehouse monitoring device, characterized in that: include: An acquisition module is used to obtain scene image information of personnel on duty in the warehouse; A screening module, used for inputting the scene image information into a first neural network model for screening to determine the behavior image information of the on-duty personnel; wherein the neural network model is trained by a plurality of behavior image information in a historical period; A determination module, used to determine a behavior scoring result of the behavior image information according to the behavior image information, and determine a behavior management plan for the on-duty personnel according to the behavior scoring result; An output module, used for outputting the behavior management plan to perform scheduling management on the on-duty personnel; The determination module is further used to determine the corresponding behavior scoring result according to the behavior image information of the on-duty personnel, so as to construct a scoring matrix according to the behavior scoring result; the scoring matrix includes a plurality of scoring sets of the on-duty personnel; The determination module is further used to calculate the similarity between the score set of the on-the-job personnel and the standard score set according to the score matrix; The determination module is further used to determine a behavior management plan for the on-duty personnel based on the similarity.

8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the steps of the warehouse monitoring method according to any one of claims 1 to 6 are implemented.

9. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the steps of the warehouse monitoring method according to any one of claims 1 to 6 are implemented.

Images