A method and system for intelligently identifying the entry and exit of a human shadow bomb equipment

Abstract

The application discloses a kind of human shadow ammunition equipment warehouse-in and warehouse-out intelligent identification method and system, it is related to intelligent warehousing management technical field, by the inventory registration of each human shadow ammunition equipment in equipment warehouse, based on warehouse-out demand table, inventory information, inventory state and warehouse-in application information, carry out warehouse-out, warehouse-in operation, the visual image tracking of several human shadow ammunition equipment of warehouse-in and warehouse-out, based on image tracking result obtains several frames equipment directional picture, obtains trace information and analyzes according to the equipment directional picture of total frame number, extracts and obtains warehouse-out feature set and warehouse-in feature set, and compares with the warehouse-in and warehouse-out template configured in advance, based on the identification of abnormal warehouse-in and warehouse-out behavior according to comparison result, and information extraction is carried out to warehouse-in and warehouse-out behavior, establishes topological information tree and carries out linkage information push, realizes the safe, efficient, accurate ammunition equipment warehouse-in and warehouse-out management.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of intelligent warehouse management, in particular to a human shadow ammunition equipment warehouse-in and warehouse-out intelligent identification method and system. BACKGROUND

[0002] In the traditional warehouse management method, especially the warehouse-in and warehouse-out management of special materials (such as human shadow ammunition equipment), most of them rely on manual recording, barcode scanning or simple visual identification technology for inventory registration and tracking, which is easy to miss or misrecord, and difficult to synchronize with the real warehouse-in and warehouse-out operation in time, resulting in inaccurate inventory information. The existing visual identification is often limited to local node shooting, lacks continuous tracking capability, and it is difficult to obtain complete warehouse-in and warehouse-out trajectory information. Moreover, it lacks multi-dimensional feature extraction and comparison of warehouse-in and warehouse-out behaviors, resulting in that various information generated in the warehouse-in and warehouse-out process cannot be effectively associated and integrated, leading to difficulty in tracing and insufficient linkage response capability, thereby affecting the efficiency of equipment warehouse-in and warehouse-out. Therefore, there is an urgent need for a warehouse-in and warehouse-out management method that can realize full-process visual tracking, feature extraction, abnormal behavior automatic identification and information linkage push, so as to improve the safety, accuracy and efficiency of human shadow ammunition equipment management. SUMMARY

[0003] The purpose of the present application is to provide a human shadow ammunition equipment warehouse-in and warehouse-out intelligent identification method and system to solve the problems in the background art.

[0004] In order to achieve the above-mentioned purpose, the present application provides the following technical scheme: a human shadow ammunition equipment warehouse-in and warehouse-out intelligent identification method, comprising the following steps:

[0005] Step S1: inventory registration is performed for each human shadow ammunition equipment in the equipment warehouse, warehouse-out operation is performed based on the warehouse-out demand table and inventory information, and warehouse-in operation is performed based on the inventory state and warehouse-in application information;

[0006] Step S2: visual image tracking is performed on the several pieces of human shadow ammunition equipment for warehouse-in and warehouse-out, several frames of equipment directional pictures are obtained based on the image tracking results, and warehouse-in and warehouse-out trace information is obtained according to the equipment directional pictures of all frames;

[0007] Step S3: the warehouse-in and warehouse-out trace information is analyzed, the warehouse-out feature set and the warehouse-in feature set are extracted, and comparison is made with the pre-configured warehouse-in and warehouse-out template, abnormal warehouse-in and warehouse-out behaviors are identified based on the comparison results, information extraction is performed on the warehouse-in and warehouse-out behaviors, a topological information tree is established, and linkage information push is performed.

[0008] In a preferred embodiment, the process of inventory registration for each human shadow ammunition equipment in the equipment warehouse, warehouse-out operation based on the warehouse-out demand table and inventory information, and warehouse-in operation based on the inventory state and warehouse-in application information comprises:

[0009] The equipment identity information of the shadow ammunition equipment is used to create a respective registration text, all registration texts are stored in a database, the registration texts are audited, and based on the audit results, the inventory registration of the shadow ammunition equipment is completed, and the rejection processing of the registration texts is completed;

[0010] The shadow ammunition equipment required for delivery is recorded in the delivery demand table, data monitoring is performed on all registration texts stored in the database and completed audit, the delivery and inventory records of the shadow ammunition equipment under each equipment area are obtained, and the inventory information is obtained based on the delivery and inventory records;

[0011] The shadow ammunition equipment required for delivery is selected from the equipment warehouse according to the delivery demand table, and the whole delivery process is supervised, and the shadow ammunition equipment is transported to the equipment warehouse according to the storage demand table, and the whole storage process is supervised.

[0012] In a preferred embodiment, the process of visually tracking a plurality of shadow ammunition equipment for delivery and storage includes:

[0013] A plurality of storage location key nodes are set on the delivery path channel and the storage path channel, the coverage range includes the warehouse door area, the shelf channel and the loading and unloading point, the warehouse door area is used as the visual starting point, and the loading and unloading point is used as the visual ending point, the visual image tracking of the whole delivery operation is performed, the positions of the visual starting point and the visual ending point are exchanged, and the visual image tracking of the whole storage operation is performed;

[0014] A plurality of delivery tracking nodes are set, a plurality of image recording devices are deployed at the visual starting point, the visual ending point and the delivery tracking nodes, a plurality of frames of image segments of the shadow ammunition equipment delivered from the equipment warehouse are photographed by the image recording devices, the delivery scene animation obtained by splicing different image segments is used as the image tracking result of the delivery operation;

[0015] A plurality of storage tracking nodes are set, the mechanical arm operation points corresponding to the storage tracking nodes are set for deploying the mechanical arm, the image recording devices are deployed at the storage tracking nodes, the imaging interception area is set to obtain the storage image frames of the shadow ammunition equipment at each position, when the shadow ammunition equipment at any storage tracking node is not in the imaging interception area, the mechanical arm at the current storage tracking node adjusts the position of the shadow ammunition equipment to be in the imaging interception area;

[0016] The shadow ammunition equipment is loaded from the loading and unloading point and stored in the equipment warehouse, all storage image frames of the whole storage operation are integrated as the image tracking result of the storage operation.

[0017] In a preferred embodiment, the process of obtaining a plurality of equipment orientation pictures based on the image tracking result includes:

[0018] Set the first processing window and the second processing window;

[0019] The first processing window is used for image preprocessing of the several frame image segments and the warehouse image frame, including image denoising, image enhancement and layered distortion correction; a standard noise image is obtained through image denoising, the standard noise image is converted into a standard fitting image through image enhancement, the standard fitting image is subjected to layered distortion correction, and an equipment framing interface image is obtained, wherein the equipment framing interface image covers a complete human shadow ammunition equipment picture and a surrounding environment picture;

[0020] The second processing window is used for picture reconstruction of the equipment framing interface image, the equipment framing interface image is divided into several reconstruction regions, image features under each reconstruction region are identified, and the image features include directional element features and non-directional element features;

[0021] When there is only directional element feature in a reconstruction region, the reconstruction region is retained as a splicing object, when there is only non-directional element feature, the reconstruction region is removed, and when there are both directional element feature and non-directional element feature, the reconstruction region is stripped of non-directional element feature, and the reconstruction region including only directional element feature is taken as a splicing object; all the splicing objects are spliced in time sequence and position sequence to obtain several frames of equipment directional pictures.

[0022] In a preferred embodiment, the process of obtaining the warehouse trace information according to the equipment directional pictures of the total number of frames includes:

[0023] A plurality of trace marking points are selected in each frame of equipment directional picture, trace marking points at the same equipment position on different frames of equipment directional picture are connected to obtain a warehouse trace path, and the warehouse trace path is divided into a plurality of information nodes based on operation time stamps;

[0024] Operation related information of the human shadow ammunition equipment at the current time and position is imported for each information node, operation related information and evidence snapshots of all information nodes on the warehouse trace path are linked and integrated as warehouse trace information, the warehouse trace information is encapsulated as a data packet, and the address of the data packet is mapped to a pre-deployed virtual machine.

[0025] In a preferred embodiment, the process of image enhancement and layered distortion correction includes:

[0026] A plurality of to-be-enhanced regions in the standard noise image are set, pixel contrast, illumination intensity and color saturation are taken as enhancement type items, any enhancement type item that does not meet the requirements is taken as a to-be-enhanced pixel block, and is marked as a to-be-enhanced point in the to-be-enhanced region;

[0027] Group all to-be-enhanced point positions under the same enhancement type item into one type point set, set several gradient enhancement intervals, integrate to-be-enhanced point positions under the same gradient enhancement interval of the same type point set into one gradient point cluster, perform pixel adjustment on several gradient point clusters, and convert the standard noise image into a standard fitting image;

[0028] Divide the standard fitting image into different focusing plane layers based on the depth of field size, integrate the standard fitting image of the same focusing plane layer into one layered image set, set a sliding correction window to perform window traversal on each layered image set, and cut the standard fitting image into several to-be-corrected sub-regions;

[0029] Mark a number of distortion points in each to-be-corrected sub-region, and mark the distortion type of the distortion points, classify distortion points of the same distortion type into one distortion set, set a corresponding number of correction parameters based on the number of distortion sets, and each correction parameter is used to correct the distortion points in the corresponding distortion set, thereby converting the distortion points into normal points, and after completing the distortion correction of all layered image sets, converting all standard fitting images into equipment framing interface images.

[0030] In a preferred embodiment, the process of analyzing the in-and-out warehouse trace information, extracting the out-of-warehouse feature set and the in-warehouse feature set, and comparing with the pre-configured in-and-out warehouse template includes:

[0031] Setting several types of feature extraction rules to extract features from the in-and-out warehouse trace information, obtaining different types of in-and-out warehouse related features, including operation context features, spatio-temporal trajectory features, equipment identity features, and behavior sequence features;

[0032] Based on the in-and-out warehouse trace information, the operation type of the current human shadow ammunition equipment is determined, when the operation type is in-warehouse, all in-and-out warehouse related features related to the current in-warehouse are extracted to generate an in-warehouse feature set, and when the operation type is out-of-warehouse, all in-and-out warehouse related features related to the current out-of-warehouse are extracted to generate an out-of-warehouse feature set;

[0033] Pre-configure the in-and-out warehouse template, the in-and-out warehouse template includes several text identification words, deploy the rule engine to access several feature extraction rules, in-and-out warehouse related features, and in-and-out warehouse template, and compare each operation behavior of each human shadow ammunition equipment in the in-and-out warehouse process in turn, when the in-and-out warehouse features of the operation behavior are not successfully compared with the text identification words of the in-and-out warehouse template, the current operation behavior is screened as an abnormal in-and-out warehouse behavior, and all abnormal in-and-out warehouse behaviors are screened.

[0034] In a preferred embodiment, the process of extracting information from the in-and-out warehouse behavior includes:

[0035] The preset information rule library is used for storing rule forms and identification forms;

[0036] The information rule library is used for type grouping and event labeling of all the times of warehouse-in and warehouse-out behaviors, the type grouping includes primary grouping and secondary grouping, and the rule forms include preliminary screening forms and rescreening forms;

[0037] The preliminary screening forms are used for primary grouping of the warehouse-in and warehouse-out behaviors, screening all abnormal warehouse-in and warehouse-out behaviors, the rescreening forms are used for secondary grouping of the abnormal warehouse-in and warehouse-out behaviors, obtaining abnormal detail types of each abnormal warehouse-in and warehouse-out behavior, and the normal warehouse-in and warehouse-out behaviors are not processed, and the abnormal detail types include quantity abnormality, time sequence abnormality and frequency abnormality;

[0038] The identification forms are used for event labeling of the abnormal warehouse-in and warehouse-out behaviors with different abnormal detail types, labeling each abnormal warehouse-in and warehouse-out behavior as a standardized abnormal structure event, and labeling event information, the event information includes event ID, abnormal type, time stamp, core entity and event detail content.

[0039] In a preferred embodiment, the process of establishing a topological information tree for linkage information pushing includes:

[0040] All the event nodes and event edges are connected based on the mutual relationship of mapping elements, and then an information sub-tree is built, a root node is set, all the information sub-trees are connected to the root node to obtain the topological information tree, and when a standardized abnormal structure event is added, an information sub-tree corresponding to the current standardized abnormal structure event is established and mounted to the topological information tree;

[0041] When information retrieval of a person, ammunition, equipment shadow in and out of a warehouse is performed, the corresponding event ID and retrieval words are input, the corresponding information sub-tree is retrieved from the topological information tree based on the event ID, and the corresponding event node on the information sub-tree is retrieved based on the retrieval words;

[0042] The event information of the directly retrieved event node is taken as direct demand information;

[0043] All the other event nodes on the event edges of the information sub-tree corresponding to the retrieved event node are traversed, the event information of the other event nodes is taken as linkage information, and the linkage information is divided into depth levels, and the linkage information is folded into a plurality of depth level data linkage layers;

[0044] The folded linkage information in the data linkage layers is unfolded in order from low to high based on the depth levels, whether to pick the linkage information in the data linkage layers is determined by the object of information retrieval, and the abnormal warehouse-in and warehouse-out behaviors are processed based on the direct retrieval information and the linkage information.

[0045] The application also provides a human shadow ammunition equipment warehouse in and out intelligent identification system, which comprises:

[0046] The warehouse in and out module is used for inventory registration of each human shadow ammunition equipment in the equipment warehouse, warehouse out operation based on the warehouse out demand table and the inventory information, and warehouse in operation based on the inventory state and the warehouse in application information.

[0047] The visual tracking module is used for visual image tracking of a plurality of human shadow ammunition equipment in and out, and a plurality of frames of equipment directional pictures are obtained based on the image tracking results, and the in and out trace information is obtained according to the equipment directional pictures of all frames.

[0048] The intelligent identification and information pushing module is used for analyzing the in and out trace information, extracting the warehouse out feature set and the warehouse in feature set, comparing with the pre-configured in and out template, identifying the abnormal in and out behavior based on the comparison result, extracting the in and out behavior information, establishing a topological information tree, and pushing the linked information.

[0049] In the above technical solution, the application provides the technical effects and advantages:

[0050] 1. The application can accurately capture each frame of picture of the equipment in the in and out process, form a complete equipment directional picture sequence, realize accurate restoration and digital storage of the real operation process, analyze the in and out trace information, extract the warehouse out feature set and the warehouse in feature set, intelligently compare with the pre-set in and out template, automatically identify the abnormal behaviors such as quantity abnormality, path abnormality and identity inconsistency, and improve the safety management level to a certain extent.

[0051] 2. The application establishes a topological information tree, hierarchically organizes and associates the abnormal events and related information, and pushes the information, so that the user can obtain direct demand information and expand associated linked information when performing information retrieval, thereby providing comprehensive and structured data support for abnormal behavior analysis, tracing and decision-making, effectively improving the decision-making efficiency, and indirectly improving the management efficiency of the ammunition equipment in and out. DETAILED DESCRIPTION

[0052] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the drawings needed in the embodiments will be briefly introduced. Obviously, the drawings in the following description are only some embodiments described in the present application, and other drawings can also be obtained by those skilled in the art based on these drawings.

[0053] Figure 1 The method flowchart of the present application.

[0054] Figure 2System block diagram of the present application. DETAILED DESCRIPTION

[0055] In order to make the objects, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are some but not all of the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by a person of ordinary skill in the art without creative work fall within the protection scope of the present application.

[0056] Embodiment 1, please refer to Figure 1 The human shadow ammunition equipment warehouse-in and warehouse-out intelligent identification method described in the embodiment includes the following steps:

[0057] Step S1: Inventory registration is performed for each human shadow ammunition equipment in the equipment warehouse, warehouse-out operation is performed based on a warehouse-out demand table and inventory information, and warehouse-in operation is performed based on an inventory state and warehouse-in application information;

[0058] Step S2: Visual image tracking is performed on a plurality of human shadow ammunition equipment that is warehouse-in or warehouse-out, a plurality of frames of equipment directional pictures are obtained based on the image tracking result, and warehouse-in and warehouse-out trace information is obtained according to the equipment directional pictures of all the frames;

[0059] Step S3: The warehouse-in and warehouse-out trace information is analyzed, a warehouse-out feature set and a warehouse-in feature set are extracted, and comparison is performed with a preconfigured warehouse-in and warehouse-out template, abnormal warehouse-in and warehouse-out behaviors are identified based on the comparison result, information extraction is performed on the warehouse-in and warehouse-out behaviors, a topology information tree is established, and linkage information is pushed.

[0060] It needs to be further explained that, in the specific implementation process, the process of inventory registration performed for each human shadow ammunition equipment in the equipment warehouse, warehouse-out operation performed based on a warehouse-out demand table and inventory information, and warehouse-in operation performed based on an inventory state and warehouse-in application information includes:

[0061] The equipment identity information of all the human shadow ammunition equipment in the equipment warehouse is counted, and the equipment identity information specifically includes the storage position of the human shadow ammunition equipment, the warehouse-in time, different equipment types, and the number of equipment corresponding to each equipment type;

[0062] A registration text corresponding to each human shadow ammunition equipment is created based on the equipment identity information, the registration text is used to record the warehouse-in and warehouse-out situation of the human shadow ammunition equipment, and all the registration texts are stored in a database deployed in the equipment warehouse;

[0063] The registration text is audited by the database, if the audit is passed, the current inventory registration of the human shadow ammunition equipment is completed, the corresponding equipment identity information is written into the registration text, if the audit is not passed, the current registration text is rejected;

[0064] The audit is based on the IP address corresponding to the associated registration text, if the IP address is not in the preset address list, the audit is not passed, otherwise, the audit is passed;

[0065] The warehouse-out requirement table is used to record all human shadow ammunition equipment required to be taken out of the warehouse;

[0066] The update frequency is set, and all registration texts stored in the database and completed audit are monitored in real time based on the update frequency, and then the in-out warehouse record of the human shadow ammunition equipment in each equipment area in the current equipment warehouse is obtained;

[0067] The inventory information is obtained based on all in-out warehouse records, and the inventory information is used to indicate the storage situation of the human shadow ammunition equipment in all equipment areas in the equipment warehouse at a certain time, including equipment out of warehouse and equipment in warehouse;

[0068] The human shadow ammunition equipment required to be taken out of the warehouse is selected from the equipment warehouse according to the warehouse-out requirement table, the warehouse-out operation is carried out and the whole warehouse-out process is supervised, and the human shadow ammunition equipment from outside is transported into the equipment warehouse according to the warehouse-in requirement table, the warehouse-in operation is carried out, and the whole warehouse-in process is supervised.

[0069] The inventory state is described as follows:

[0070] The inventory alarm threshold is set, and the number of human shadow ammunition equipment stored in the current equipment warehouse is obtained based on the inventory information, if the number is less than or equal to the inventory alarm threshold, the inventory state is updated to be empty, otherwise, the inventory state is identified as normal.

[0071] It needs to be further explained that in the specific implementation process, the process of visually tracking a plurality of human shadow ammunition equipment in and out of the warehouse includes:

[0072] A plurality of storage location key nodes are set on the warehouse-out path channel and the warehouse-in path channel of the human shadow ammunition equipment, and the specific storage location key node coverage range includes the warehouse door area, the shelf channel and the loading and unloading point;

[0073] The storage location key node of the warehouse door area is taken as the visual starting point, and the storage location key node of the loading and unloading point is taken as the visual ending point, the visual image tracking of the whole process corresponding to the warehouse-out operation is carried out, the positions of the visual starting point and the visual ending point are exchanged, and the visual image tracking of the whole process corresponding to the warehouse-in operation is carried out;

[0074] The content of the visual image tracking of the whole process corresponding to the warehouse-out operation is as follows:

[0075] A plurality of out-of-warehouse tracking nodes are arranged between the visual starting point and the visual ending point, the plurality of out-of-warehouse tracking nodes are numbered, and the number is denoted as i, i.e. i = 1, 2, 3, …, n, where n is a natural number greater than 0. A plurality of image recording devices are deployed at the visual starting point, the visual ending point, and the out-of-warehouse tracking nodes. The image recording devices are started, and the image recording devices capture a plurality of frames of image segments of the mannequin ammunition equipment out of the equipment warehouse;

[0076] The connections between different image segments are performed, and an out-of-warehouse scene animation is constructed. The out-of-warehouse scene animation is used to represent the real-time pictures of the mannequin ammunition equipment at each position when the mannequin ammunition equipment is operated to be out of the warehouse. All the real-time pictures are combined to serve as the image tracking result of the whole process corresponding to the out-of-warehouse operation;

[0077] The content of the visual image tracking of the whole process corresponding to the in-warehouse operation is as follows:

[0078] A plurality of in-warehouse tracking nodes are arranged between the visual starting point and the visual ending point, and a mechanical arm operation point is associated with each in-warehouse tracking node. A mechanical arm for controlling the mannequin ammunition equipment is deployed at the mechanical arm operation point. An image recording device is deployed at each in-warehouse tracking node, and an imaging interception area corresponding to each image recording device is set.

[0079] The image recording device is operated based on the imaging interception area. The image recording device obtains an in-warehouse image frame of the mannequin ammunition equipment at each position in the imaging interception area. When the mannequin ammunition equipment at any in-warehouse tracking node is not in the imaging interception area, the mechanical arm of the mechanical arm operation point associated with the current in-warehouse tracking node is manipulated to adjust the position of the mannequin ammunition equipment so that the mannequin ammunition equipment is in the imaging interception area. When the mannequin ammunition equipment is in the imaging interception area, no operation is performed.

[0080] A cargo transport robot is enabled at the visual starting point and the visual ending point. The cargo transport robot performs a loading operation on the mannequin ammunition equipment from the loading and unloading point and stores the mannequin ammunition equipment in the equipment warehouse. All in-warehouse image frames of the mannequin ammunition equipment in the whole process of the in-warehouse operation are integrated to serve as the image tracking result of the whole process corresponding to the in-warehouse operation.

[0081] It needs to be further explained that, in the specific implementation process, the process of obtaining the in-and-out warehouse trace information based on the image tracking result to obtain a plurality of frames of equipment directional pictures includes:

[0082] A first processing window and a second processing window are set.

[0083] The first processing window is used for processing the image tracking results of the whole process corresponding to the out-of-warehouse operation and the in-warehouse operation of the human shadow ammunition equipment respectively, and performing image preprocessing on the several frame image segments of the human shadow ammunition equipment out of the warehouse and the several frame in-warehouse image frames ready to be stored in the warehouse.

[0084] The image preprocessing includes image denoising, image enhancement and hierarchical distortion correction.

[0085] The image denoising includes the following steps: the first processing window selects a denoising area composed of several pixel blocks, the pixel value of the pixel block at the center of the rectangular area is taken as a pixel reference value, and the pixel values of other pixel blocks around the center of the rectangular area are uniformly taken as the pixel reference value; after adjusting the pixel values of all pixel blocks, a standard noise image is obtained.

[0086] The image enhancement includes the following steps: setting several to-be-enhanced areas in the standard noise image, detecting the pixel contrast, illumination intensity and color saturation of each to-be-enhanced area, obtaining a pixel block that does not meet the requirements of any of the pixel contrast, illumination intensity and color saturation, and marking it as a to-be-enhanced point in the to-be-enhanced area.

[0087] The pixel contrast, illumination intensity and color saturation are taken as enhancement type items.

[0088] All to-be-enhanced point positions under the same enhancement type item are grouped into a type point set, several gradient enhancement intervals of different numerical ranges are set, and to-be-enhanced point positions in the same gradient enhancement interval under the same type point set are integrated into a gradient point cluster.

[0089] According to the gradient enhancement intervals set according to the type point sets corresponding to different enhancement type items, several different gradient point clusters are obtained, and pixel adjustment of the gradient point clusters is performed in parallel. The pixel adjustment includes increasing the pixel contrast value, increasing the illumination intensity value and fitting the color saturation. After the pixel adjustment, the standard noise image is converted into a standard fitting image.

[0090] It should be noted that the enhancement type items of several pixel blocks in the same gradient interval have the common feature of the same deviation numerical range, and they are grouped into a gradient point cluster. Processing the image enhancement of all pixel blocks represented by the same gradient point cluster realizes batch processing and effectively improves the processing efficiency of image enhancement.

[0091] The layered distortion correction includes: dividing the standard fitting image into different focus plane layers based on the depth of field size, integrating the standard fitting images in the same focus plane layer into a layered atlas, and then dividing to obtain the layered atlas corresponding to different focus plane layers. When the photographed human shadow ammunition equipment is not in the same focus plane layer (i.e., different distances from the camera), a set of correction parameters cannot achieve perfect correction for all objects at different depths, so it is necessary to divide according to the number of depth of field to correct the distortion.

[0092] A sliding correction window is set, each layered atlas is windowed by the sliding correction window, the sliding correction window covers each standard fitting image included in the layered atlas, and the standard fitting image is divided into a plurality of to-be-corrected sub-regions;

[0093] A plurality of distortion points are marked in each to-be-corrected sub-region, and the corresponding distortion type of the distortion point is marked. The distortion type includes radial distortion, tangential distortion, and shutter distortion. The distortion points of the same distortion type in all to-be-corrected sub-regions are classified into a distortion set;

[0094] A corresponding number of correction parameters is set based on the number of distortion sets. Each correction parameter is used to correct the distortion points in a corresponding distortion set, and then the distortion points are converted into normal points. After the sliding correction window completes the distortion correction of all different layered atlases, all standard fitting images are converted into corresponding equipment framing interface images, and the equipment framing interface images cover complete human shadow ammunition equipment pictures and surrounding environment pictures;

[0095] The second processing window is used to reconstruct the picture of the equipment framing interface image. The equipment framing interface image is equally and sized divided into a plurality of reconstruction regions. The image features in each reconstruction region are identified in sequence. The image features include directional element features and non-directional element features.

[0096] The directional element feature is used to express the outline appearance feature of the artificial ammunition equipment.

[0097] The non-directional element feature is used to express the surrounding environment feature of the artificial ammunition equipment.

[0098] When there is only a directional element feature in a reconstruction region, the current reconstruction region is retained as a to-be-spliced object. When there is only a non-directional element feature in a reconstruction region, the current reconstruction region is removed. When there are directional element features and non-directional element features in a reconstruction region, the corresponding reconstruction region is stripped of non-directional element features to obtain a reconstruction region including only directional element features as a to-be-spliced object.

[0099] The all to-be-spliced objects are spliced according to time sequence and position sequence, and a plurality of frames of equipment directional pictures are constructed, each frame of equipment directional picture being used to represent real-time conditions of the shadow ammunition equipment at an in-out warehouse position;

[0100] A plurality of trace calibration points are selected on each frame of equipment directional picture, the trace calibration points including a plurality of contour boundary points, center points and corner points on a corresponding contour of the shadow ammunition equipment, and the trace calibration points at the same equipment position on different frames of equipment directional picture are connected to obtain an in-out warehouse trace path;

[0101] The in-out warehouse trace path is divided into a plurality of information nodes based on operation time stamps, and operation-related information of the shadow ammunition equipment at a current time and position is imported for each information node, the operation-related information including equipment identity information, space-time trajectory information, context information and audit information;

[0102] The space-time trajectory information describes when, where and how the shadow ammunition equipment moves, and specifically includes operation time, spatial position, event label and belonging camera ID, the event label is used to record different operation behaviors of the equipment, such as picking up from a shelf of the equipment warehouse, passing through a warehouse door, loading and being scanned, which correspond to the concept of how to move, and the belonging camera ID indicates that the current equipment is captured by which camera;

[0103] The context information describes how all operation behaviors are completed and surrounding environment information when the operation behaviors are performed, and specifically includes operator information, operation behavior sequence, interactive device information and environment data, the operator information includes an operator ID (such as staff_007) associated through face recognition or ID card recognition, and records an activity range in the whole operation process; the operation behavior sequence records specific actions of the personnel, including approaching a shelf, bending, scanning equipment code, holding up equipment and moving equipment, the interactive device information involves related information of each device in operation, such as a used goods transportation robot ID, a handheld terminal ID and a mechanical arm ID; and the environment data records real-time temperature and humidity in the equipment warehouse during the in-out warehouse process;

[0104] The audit information is used to describe whether there is suspected abnormality in the operation behavior at different times and positions, and to perform evidence reservation for the suspected abnormal operation behavior, when there is suspected abnormality in a certain operation behavior, an abnormal flag bit is associated with a current information node, and image segments of a plurality of seconds before and after an equipment directional picture corresponding to the current information node are stored to a cloud server path as evidence snapshot links;

[0105] Link the operation-related information of all information nodes on the warehouse entry and exit trace path and the evidence snapshot, integrate them into warehouse entry and exit trace information, encapsulate the warehouse entry and exit trace information into a data packet, and map the address of the data packet to a pre-deployed virtual machine.

[0106] It should be further explained that, in the specific implementation process, the warehouse entry and exit trace information is analyzed, the warehouse-out feature set and the warehouse-in feature set are extracted, and the pre-configured warehouse-in and warehouse-out template is compared, and the process of identifying abnormal warehouse-in and warehouse-out behavior based on the comparison result includes:

[0107] A plurality of types of feature extraction rules are set, and feature extraction is performed on the warehouse entry and exit trace information based on all the feature extraction rules. After the feature extraction is completed, different types of warehouse-in and warehouse-out related features are obtained, including operation context features, space-time trajectory features, equipment identity features, and behavior sequence features.

[0108] The plurality of types of feature extraction rules include the following:

[0109] Rule 1: Quantity consistency rule;

[0110] Rule 2: Path compliance rule;

[0111] Rule 3: Identity consistency rule;

[0112] Rule 4: Process completeness rule;

[0113] The corresponding relationship between different feature extraction rules and corresponding types of warehouse-in and warehouse-out related features is as follows:

[0114] Rule 1: Quantity consistency rule—context feature;

[0115] Rule 2: Path compliance rule—space-time trajectory feature;

[0116] Rule 3: Identity consistency rule—equipment identity feature;

[0117] Rule 4: Process completeness rule—behavior sequence feature;

[0118] Based on the warehouse entry and exit trace information, the operation type of the person-image ammunition equipment is determined. When the operation type is warehouse-in, all warehouse-in and warehouse-out related features related to the current warehouse-in are extracted to generate a warehouse-in feature set. When the operation type is warehouse-out, all warehouse-in and warehouse-out related features related to the current warehouse-out are extracted to generate a warehouse-out feature set.

[0119] A pre-configured warehouse-in and warehouse-out template is configured, and the warehouse-in and warehouse-out template defines a plurality of text identification words for identifying a plurality of warehouse-in and warehouse-out features. The text identification word is used to represent the behavior feature content of the person-image ammunition equipment in the correct execution of warehouse-in and warehouse-out.

[0120] The rule engine is deployed, a plurality of feature extraction rules, warehouse in / out related features and warehouse in / out templates are accessed by the rule engine, and then all operation behaviors of each person missile ammunition equipment in the warehouse in / out process are sequentially compared. When the warehouse in / out features corresponding to the operation behaviors are not successfully compared with the text identification words of the warehouse in / out templates, the operation behaviors are screened as abnormal warehouse in / out behaviors, and all abnormal warehouse in / out behaviors are screened.

[0121] It should be noted that the content of the different feature extraction rules for extracting the corresponding warehouse in / out related features is as follows: the rule target of the quantity consistency rule is to judge whether the equipment quantity of the planned warehouse in / out is consistent with the equipment quantity of the actual operation; the data source of the execution rule is the field of the warehouse in / out trace information associated with the business order number and the actual identified equipment list, the warehouse in / out demand table or the warehouse in / out application table is queried according to the field, the calculation quantity is obtained, and when the calculation quantity is not consistent with the actual quantity, the quantity difference is taken as the context feature;

[0122] The rule target of the path compliance rule is to judge whether the actual track of the equipment moving in the warehouse area is consistent with the predetermined safe and efficient path; the data source of the execution rule is the field related to the "spatial and temporal track sequence" in the warehouse in / out trace information, all data points (such as image coordinates, actual coordinates and area labels) with position information in the spatial and temporal track sequence are extracted, the check of the must-pass point is performed, the must-pass point is a key node that must be passed on the standard path, such as a scanning point, a weighing point and an exit, whether the fields of the spatial and temporal track sequence cover all the must-pass points is checked, and then the check of the forbidden area is performed, the forbidden area is a pre-set forbidden entry area (such as other shelf channels), whether the fields of the spatial and temporal track sequence include the forbidden area is checked;

[0123] The rule target of the identity consistency rule is to judge whether the actual operation equipment is completely consistent with the planned operation equipment, whether there is "package adjustment" or "wrong taking"; the data source of the execution rule is the field of the warehouse in / out trace information associated with the business order number and the actual identified equipment list, the planned equipment ID set in the database is queried based on the business order number, the actual identified equipment ID set is obtained, and the set operation is performed to obtain the different calculation set;

[0124] The calculation set includes the following:

[0125] Calculation difference set 1: the equipment in the plan but not in the actual, i.e. the missing equipment;

[0126] Calculation difference set 2: the equipment in the actual but not in the plan, i.e. the redundant equipment;

[0127] Calculation intersection: the equipment in the actual and the plan, i.e. the correct equipment operation;

[0128] The computing set is taken as an equipment identity feature of the human shadow ammunition equipment;

[0129] The rule target of the process completeness rule is to determine whether an operation is completed strictly according to a pre-set standard operation process (SOP), whether there is a step omission or a sequence error; the data source for executing the rule is a field corresponding to an event label sequence and an operation action sequence in the warehouse-in and warehouse-out trace information; the event label sequence in the track sequence is merged with the operation action sequence in the behavior sequence according to the time sequence to obtain a complete actual behavior sequence; the actual behavior sequence is compared with a pre-configured standard behavior sequence template through a sequence matching algorithm to calculate a missing step and an error step sequence, and the calculation result is taken as a behavior sequence feature.

[0130] It needs to be further explained that, in the specific implementation process, the process of extracting information from the warehouse-in and warehouse-out behaviors, establishing a topological information tree and pushing linkage information includes:

[0131] A pre-set information rule library stores a rule form and an identification form;

[0132] The information rule library is used for type grouping and event labeling of all warehouse-in and warehouse-out behaviors based on the information rule library, the type grouping includes primary grouping and secondary grouping, and the rule form includes a primary screening form and a secondary screening form;

[0133] The primary screening form is used for primary grouping of the warehouse-in and warehouse-out behaviors, the primary screening form records behavior numbers of all abnormal warehouse-in and warehouse-out behaviors and behavior numbers of normal warehouse-in and warehouse-out behaviors, all abnormal warehouse-in and warehouse-out behaviors are screened through the primary grouping, and the abnormal warehouse-in and warehouse-out behaviors are secondary grouped based on the secondary screening form to obtain an abnormal detail type of each abnormal warehouse-in and warehouse-out behavior, and the normal warehouse-in and warehouse-out behaviors are not processed;

[0134] The abnormal detail types recorded in the secondary screening form include quantity abnormality, time sequence abnormality and frequency abnormality, for example, when the warehouse-in and warehouse-out quantity of the human shadow ammunition equipment does not match a document, it is quantity abnormality; when there is no order first warehouse-out or warehouse-in without quality inspection, it is time sequence abnormality; when the same human shadow ammunition equipment is warehouse-in and warehouse-out in a time period more than a pre-set number threshold of the current time period, it indicates that the frequency abnormality occurs;

[0135] The identification form is used for event labeling of the abnormal warehouse-in and warehouse-out behaviors corresponding to different abnormal detail types, each abnormal warehouse-in and warehouse-out behavior is labeled as a standardized abnormal structure event, and event information is labeled, the event information includes an event ID, an abnormal type, a timestamp, a core entity and event detail content;

[0136] Wherein, the event ID is the unique identification of the standardized abnormal structure event, the abnormal type is the quantity abnormality, the time sequence abnormality and the frequency abnormality, the time stamp represents the specific time point of the abnormality occurrence, the core entity is used for recording the materials, personnel, equipment and storage locations involved in the event, and the event details content is used for recording the specific content in the whole process of the event occurrence, such as the actual number of the human shadow ammunition equipment when executing the warehouse in and out, which is inconsistent with the specified number recorded in the document.

[0137] Each standardized abnormal structure event is instantiated as an event node, the core entity of the standardized abnormal structure event is taken as the mapping element of the event node, the event edges between the event nodes are connected to each other, and the event edges are used to represent the mutual relationship between the mapping elements in different event nodes, for example, personnel A belongs to the loading and unloading group C, material B is stored in the storage location D, and personnel E operates the equipment F;

[0138] All the current event nodes and event edges are connected based on the mutual relationship of the mapping elements, and then an information sub-tree is built, each information sub-tree is used to represent all the event information of a human shadow ammunition equipment performing warehouse in and out, a root node is set, all the information sub-trees are connected to the root node, and a topological information tree is constructed;

[0139] When a new standardized abnormal structure event is added, the information sub-tree corresponding to the current standardized abnormal structure event is established, and the information sub-tree is mounted on the topological information tree, when the information retrieval of a human shadow ammunition equipment performing warehouse in and out is performed, the corresponding event ID and the retrieval word are input, the corresponding information sub-tree is retrieved from the topological information tree based on the event ID, and the corresponding event node on the information sub-tree is retrieved based on the retrieval word;

[0140] The event information of the directly retrieved event node is taken as the direct demand information;

[0141] All the other event nodes on the event edges of the information sub-tree corresponding to the retrieved event node are traversed, the event information of the other event nodes is taken as the linkage information, and the linkage information is divided into several depth levels of data linkage layers by depth level division;

[0142] Wherein, the event information of the event node is obtained from the corresponding information node in the virtual machine, the operation related information and the evidence snapshot link of the information node related to the event node are taken together as the event information;

[0143] Specifically, event information of an event node corresponding to an event edge associated with a directly retrieved event node is taken as linkage information with a depth of 1, and a data linkage layer with a depth of 1 is constructed for folding the corresponding linkage information; event information of an event node on an event edge associated with an event node corresponding to the linkage information with a depth of 1 is taken as linkage information with a depth of 2, and a data linkage layer with a depth of 2 is constructed for folding the linkage information with a depth of 2; and the process is repeated until all event nodes in an information sub-tree are constructed with corresponding data linkage layers.

[0144] The linkage information folded in the data linkage layers is unfolded in sequence based on the low-to-high order of the depth levels, for linkage information pushing; whether the linkage information pushed in the corresponding data linkage layer is selected is determined by an object of information retrieval, and a decision is made on abnormal in-and-out warehouse behaviors based on the directly retrieved information and the linkage information.

[0145] Embodiment 2, please refer to Figure 2 As shown in the figure, the present application further provides an intelligent identification system for human shadow ammunition equipment in-and-out warehouse, which comprises:

[0146] An in-and-out warehouse module is configured to register the inventory of each human shadow ammunition equipment in the equipment warehouse, perform an out-of-warehouse operation based on an out-of-warehouse demand table and inventory information, and perform an in-warehouse operation based on inventory status and in-warehouse application information;

[0147] A visual tracking module is configured to track the visual images of a plurality of human shadow ammunition equipment in-and-out warehouse, obtain a plurality of frames of equipment orientation pictures based on the image tracking results, and obtain in-and-out warehouse trace information according to the equipment orientation pictures of all frames;

[0148] An intelligent identification and information pushing module is configured to analyze the in-and-out warehouse trace information, extract an out-of-warehouse feature set and an in-warehouse feature set, compare the feature sets with pre-configured in-and-out warehouse templates, identify abnormal in-and-out warehouse behaviors based on the comparison results, extract information on the in-and-out warehouse behaviors, and establish a topological information tree for linkage information pushing.

[0149] The above is only a specific embodiment of the present application, but the protection scope of the present application is not limited thereto, any person skilled in the art can easily think of changes or replacements within the technical scope disclosed in the present application, which should be covered in the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. A method for intelligently identifying the entry and exit of human shadow ammunition equipment into a warehouse, characterized in that, Comprise the following steps: Step S1: inventory registration is performed for each shadow ammunition equipment in the equipment warehouse, warehouse-out operation is performed based on a warehouse-out demand table and inventory information, and warehouse-in operation is performed based on an inventory state and warehouse-in application information; Step S2: visual image tracking is performed on a plurality of pieces of shadow ammunition equipment that are in and out of the warehouse, a plurality of frames of equipment orientation pictures are obtained based on image tracking results, and in and out warehouse trace information is obtained according to the equipment orientation pictures of all frames; Step S3: the in and out warehouse trace information is analyzed, a warehouse-out feature set and a warehouse-in feature set are extracted, and comparison is performed with preconfigured in and out warehouse templates, abnormal in and out warehouse behaviors are identified based on comparison results, information extraction is performed on the in and out warehouse behaviors, a topology information tree is established, and linkage information is pushed; The process of obtaining the in and out warehouse trace information according to the equipment orientation pictures of all frames comprises: a plurality of trace calibration points are selected on each frame of the equipment orientation pictures, trace points on the same equipment position in different frames of the equipment orientation pictures are connected, an in and out warehouse trace path is obtained, the in and out warehouse trace path is divided into a plurality of information nodes based on operation time stamps, operation related information of the shadow ammunition equipment at a current time and position is imported for each information node, operation related information and evidence snapshots of all information nodes on the in and out warehouse trace path are linked and integrated as in and out warehouse trace information, the in and out warehouse trace information is encapsulated as a data packet, and an address of the data packet is mapped to a predeployed virtual machine; The process of analyzing the in and out warehouse trace information, extracting a warehouse-out feature set and a warehouse-in feature set, and comparing the extracted feature sets with preconfigured in and out warehouse templates comprises: a plurality of types of feature extraction rules are set to extract features from the in and out warehouse trace information, different types of in and out warehouse related features are obtained, including operation context features, space-time trajectory features, equipment identity features, and behavior sequence features; operation types of the current shadow ammunition equipment are determined based on the in and out warehouse trace information, when the operation type is warehouse-in, in and out warehouse related features related to the current warehouse-in are extracted to generate a warehouse-in feature set, and when the operation type is warehouse-out, in and out warehouse related features related to the current warehouse-out are extracted to generate a warehouse-out feature set; in and out warehouse templates are preconfigured, the in and out warehouse templates comprise a plurality of text identification words, a rule engine is deployed to access a plurality of feature extraction rules, in and out warehouse related features, and the in and out warehouse templates, all operation behaviors of each shadow ammunition equipment in the in and out warehouse process are sequentially compared, when in and out warehouse features of the operation behaviors do not match the text identification words of the in and out warehouse templates, the operation behaviors are screened as abnormal in and out warehouse behaviors, and all abnormal in and out warehouse behaviors are screened.

2. The method of claim 1, wherein the method is characterized by, The process of inventory registration for each shadow ammunition equipment in the equipment warehouse, warehouse-out operation based on a warehouse-out demand table and inventory information, and warehouse-in operation based on an inventory state and warehouse-in application information comprises: The equipment identity information of the statistical mannequin ammunition equipment is used to create a respective registration text, all registration texts are stored in a database, the registration texts are audited, and based on the audit results, the inventory registration of the mannequin ammunition equipment is completed, and the registration text is rejected; The mannequin ammunition equipment required to be discharged is recorded in the discharge demand table, all registration texts stored in the database and completed are audited, the storage information of the mannequin ammunition equipment under each equipment area is obtained, and the storage information is obtained based on the in-out record of the mannequin ammunition equipment; The mannequin ammunition equipment required to be discharged is selected from the equipment warehouse according to the discharge demand table, and the whole process of discharging is supervised, and the mannequin ammunition equipment is transported to the equipment warehouse according to the storage demand table, and the whole process of storage is supervised.

3. The intelligent identification method for the entry and exit of manned ammunition equipment according to claim 2, characterized in that, The process of tracking the visual image of a plurality of mannequin ammunition equipment in and out of the warehouse includes: A plurality of key nodes are set on the discharge path channel and the storage path channel, the coverage range includes the warehouse door area, the shelf channel and the loading and unloading point, the warehouse door area is used as the visual starting point, the loading and unloading point is used as the visual ending point, the visual image tracking of the whole process of discharging is carried out, the positions of the visual starting point and the visual ending point are exchanged, and the visual image tracking of the whole process of storage is carried out; A plurality of discharge tracking nodes are set, a plurality of image recording devices are deployed at the visual starting point, the visual ending point and the discharge tracking nodes, a plurality of frames of image segments of the mannequin ammunition equipment discharged from the equipment warehouse are photographed by the image recording devices, the discharge scene animation obtained by splicing different image segments is used as the image tracking result of the discharging operation; A plurality of storage tracking nodes are set, the corresponding mechanical arm operation points of the storage tracking nodes are set for deploying the mechanical arm, the image recording devices are deployed at the storage tracking nodes, and the imaging interception area is set to obtain the storage image frame of the mannequin ammunition equipment at each position, when the mannequin ammunition equipment at any storage tracking node is not in the imaging interception area, the position of the mannequin ammunition equipment is adjusted by the mechanical arm at the current storage tracking node to be in the imaging interception area; The mannequin ammunition equipment is loaded from the loading and unloading point and stored in the equipment warehouse, and all the storage image frames of the whole storage operation process are integrated as the image tracking result of the storage operation.

4. The method of claim 3, wherein the method further comprises: determining whether the person is a person of interest based on the determined face feature information and the determined body feature information. The process of obtaining a plurality of equipment directional pictures based on the image tracking result includes: A first processing window and a second processing window are set; The first processing window is used for image preprocessing of a plurality of image segments and storage image frames, including image denoising, image enhancement and layered distortion correction; a standard noise image is obtained by image denoising, the standard noise image is converted into a standard fitting image by image enhancement, the layered distortion correction is performed on the standard fitting image, and an equipment frame interface picture is obtained, the equipment frame interface picture covers a complete mannequin ammunition equipment picture and a surrounding environment picture; The second processing window is used for picture reconstruction of the equipment frame interface picture, the equipment frame interface picture is divided into a plurality of reconstruction regions, and the image features in each reconstruction region are identified, the image features include directional element features and non-directional element features; When there is only a directional element feature in a certain reconstruction area, the reconstruction area is reserved as a splicing object; when there is only a non-directional element feature, the reconstruction area is removed; when there are both directional element features and non-directional element features, the reconstruction area is stripped of non-directional element features, and the reconstruction area including only directional element features is taken as a splicing object; and all splicing objects are spliced in time sequence and position sequence to obtain a number of frames of directional pictures.

5. The human shadow ammunition equipment warehouse in and out intelligent identification method according to claim 4, characterized in that, The image enhancement and layered distortion correction process includes: Setting a number of to-be-enhanced regions in the standard noise image, taking pixel contrast, illumination intensity, and color saturation as enhancement type items, obtaining a pixel block that does not meet the requirements of any enhancement type item, and marking it as a to-be-enhanced point in the to-be-enhanced region; Grouping all to-be-enhanced point positions under the same enhancement type item into a type point set, setting a number of gradient enhancement intervals, integrating to-be-enhanced point positions under the same gradient enhancement interval in the same type point set into a gradient point cluster, performing pixel adjustment on a number of gradient point clusters, and converting the standard noise image into a standard fitting image; Dividing the standard fitting image into different focus plane layers based on the depth of field, integrating the standard fitting image of the same focus plane layer into a layered image set, setting a sliding correction window to traverse each layered image set, and cutting the standard fitting image into a number of to-be-corrected sub-regions; Marking a number of distortion points in each to-be-corrected sub-region and marking the distortion type of the distortion points, classifying distortion points of the same distortion type into a distortion collection, setting a corresponding number of correction parameters based on the number of distortion collections, and using each correction parameter to correct the distortion points in the corresponding distortion collection, thereby converting the distortion points into normal points; after completing the distortion correction of all layered image sets, converting all standard fitting images into a frame bounding interface image.

6. The human shadow ammunition equipment warehouse in and out intelligent identification method according to claim 5, characterized in that, The process of extracting information on warehouse entry and exit behaviors includes: The preset information rule library is used to store rule forms and identification forms; Based on the information rule library, all warehouse entry and exit behaviors are grouped by type and annotated by event, type grouping includes primary grouping and secondary grouping, and rule forms include preliminary screening forms and rescreening forms; The preliminary screening form is used to group the warehouse entry and exit behaviors once, and all abnormal warehouse entry and exit behaviors are screened out; the rescreening form groups the abnormal warehouse entry and exit behaviors twice to obtain the abnormal detail type of each abnormal warehouse entry and exit behavior, and does not process the normal warehouse entry and exit behaviors; the abnormal detail type includes quantity anomaly, time sequence anomaly, and frequency anomaly; The identification form is used to annotate the abnormal warehouse entry and exit behaviors of different abnormal detail types by event, mark each abnormal warehouse entry and exit behavior as a standardized abnormal structure event, and mark event information, including event ID, abnormal type, timestamp, core entity, and event detail content.

7. The intelligent identification method for the entry and exit of manned ammunition equipment according to claim 6, characterized in that, The process of establishing a topological information tree for linked information pushing includes: All event nodes and event edges are connected based on the mutual relationship of mapping elements, and then an information subtree is built, a root node is set, all information subtrees are connected to the root node to obtain a topology information tree, and when a standardized abnormal structure event is added, the information subtree corresponding to the current standardized abnormal structure event is established and hung on the topology information tree; When information retrieval of a certain in-out warehouse figure ammunition equipment is performed, the corresponding event ID and retrieval word are input, the corresponding information subtree is retrieved from the topology information tree based on the event ID, and the corresponding event node on the information subtree is retrieved based on the retrieval word; The event information of the directly retrieved event node is taken as direct demand information; All other event nodes on the event edges of the information subtree corresponding to the retrieved event node are traversed, the event information of the other event nodes is taken as linkage information, the linkage information is divided into a plurality of depth levels, and the linkage information is folded into a plurality of data linkage layers of depth levels; The folded linkage information in the data linkage layers is unfolded in order from low to high based on the depth level, whether to pick the linkage information in the data linkage layer is determined by the object of information retrieval, and the abnormal in-out warehouse behavior is processed based on the direct retrieval information and the linkage information.

8. A human shadow ammunition equipment warehouse in and out intelligent identification system for implementing the human shadow ammunition equipment warehouse in and out intelligent identification method of any one of claims 1 to 7. The system comprises: An in-out warehouse module for inventory registration of each figure ammunition equipment in the equipment warehouse, out-of-warehouse operation based on the out-of-warehouse demand table and the inventory information, and in-warehouse operation based on the inventory state and the in-warehouse application information; A visual tracking module for visual image tracking of a plurality of figure ammunition equipment in the in-out warehouse, obtaining a plurality of equipment orientation pictures based on the image tracking result, and obtaining in-out warehouse trace information according to the equipment orientation pictures of all frames; An intelligent identification and information pushing module for analyzing the in-out warehouse trace information, extracting an out-of-warehouse feature set and an in-warehouse feature set, comparing with a pre-configured in-out warehouse template, identifying abnormal in-out warehouse behavior based on the comparison result, extracting information of the in-out warehouse behavior, establishing a topology information tree, and pushing linkage information.

Images