View label cataloging implementation and application method, system, device and medium

Abstract

The application provides a view label cataloging implementation and application method, system, device and medium, including the following steps: acquiring view information; performing feature analysis on the view information to obtain and store analysis label information associated with the view; taking time information in the stored analysis label information as a hinge label line, taking any remaining feature label in the analysis label information as another hinge label line, performing cleaning, collision and combination to obtain a label combination array; in response to a search condition, filtering all the label combination arrays, and outputting filtering information to enable a client to call associated view information. AI analysis labeling and manual labeling are combined, the view label is more perfect, and the manual label is not limited to AI capability, and can better meet the personalized needs of users. Applications are different due to scenes, but the labeling process and label storage and retrieval do not change with applications.

Description

Technical Field

[0001] This invention relates to the field of video surveillance, specifically to a method, system, device, and medium for implementing and applying view label cataloging. Background Technology

[0002] In the field of video surveillance, most users still rely on manual monitoring for video applications. Although intelligent recognition technology based on video has gradually matured in recent years, it is still mainly used for single-point applications, such as single-point recognition and deployment.

[0003] Existing technologies primarily offer general query, statistics, and control scenarios, such as face and vehicle queries, pedestrian / vehicle traffic statistics, and blacklist control, but lack practical application capabilities. Current AI scenarios only construct data resource pools, lacking data governance, with low data standardization and inconsistent dictionaries. This makes retrieval unreliable as applications change, requiring multi-platform collaboration for analysis, which is inconvenient. Summary of the Invention

[0004] This application provides a method, system, device, and medium for implementing and applying view label cataloging to facilitate retrieval.

[0005] The first aspect of this application provides a method for implementing and applying view label cataloging.

[0006] A method for implementing and applying view label cataloging includes the following steps:

[0007] Get view information;

[0008] Perform feature parsing on the view information to obtain and store the parsed label information associated with the view;

[0009] The time information in the stored parsed label information is used as one hinge label line, and any other feature label in the parsed label information is used as another hinge label line. The labels are then cleaned, collided, and combined to obtain a label combination array.

[0010] In response to search criteria, all tag combination arrays are filtered, and the filtered information is output so that the client can retrieve the associated view information.

[0011] By adopting the above technical solution, the view information is converted into tags through feature parsing, so that the retrieval does not change with the application, and the retrieval is more accurate by associating the information in the parsed tag information.

[0012] Preferably, feature parsing is performed on the view information to obtain parsed label information associated with the view, including:

[0013] The view information is input into the AI ​​view parsing model for feature parsing to obtain parsing label information associated with the view.

[0014] Preferably, after inputting the view information into the AI ​​view parsing model for feature parsing to obtain the parsing label information associated with the view, the process includes:

[0015] Obtain manually labeled information and supplement the parsed label information with the manually labeled information.

[0016] Preferably, in response to search criteria, the tag combination array is filtered, including:

[0017] The super-tag search engine generates field search criteria in response to external search information;

[0018] By combining the field search criteria with the built-in reliability search algorithm, reliable tag combinations within a preset reliability range are selected from all tag combination arrays.

[0019] Preferably, in response to search criteria, after filtering all tag combination arrays and outputting the filtered information so that the client can retrieve the associated view information, the method further includes:

[0020] The location information in the parsed label information corresponding to the reliable label combination array is matched into the GIS to generate the motion trajectory of the filtered object in the filter information.

[0021] Preferably, in response to search criteria, after filtering all tag combination arrays and outputting the filtered information so that the client can retrieve the associated view information, the method further includes:

[0022] Based on the preset classifier, obtain the view information related to the same attribute and the view information related to the same location of the filtered objects in the filtering information;

[0023] Generate attribute-related maps and location-related maps based on view information related to the same attribute and view information related to the same location.

[0024] Preferably, after filtering all tag combination arrays in response to search criteria and outputting the filtered information to enable the client to retrieve related view information, the method further includes:

[0025] Output the parsed label information and associated view information to the intelligent interaction system so that the intelligent interaction system can generate knowledge based on the parsed label information and associated view information.

[0026] The second aspect of this application provides a system for implementing and applying a cataloging method based on view tags.

[0027] Systems based on view label cataloging implementation and application methods include:

[0028] The information acquisition module is used to acquire view information;

[0029] The parsing module is used to perform feature parsing on the view information, obtain the parsed label information associated with the view, and store it.

[0030] The combined array generation module is used to take the time information in the stored parsed tag information as one hinge tag line and any other feature tag in the parsed tag information as another hinge tag line, and perform cleaning, collision and combination to obtain a tag combined array.

[0031] The search and filtering module is used to filter all tag combination arrays in response to search criteria and output the filtered information so that the client can retrieve the associated view information.

[0032] An electronic device is provided in the third aspect of this application.

[0033] An electronic device includes a processor, a memory, and a transceiver. The memory is used to store instructions, the transceiver is used to communicate with other devices, and the processor is used to execute the instructions stored in the memory to cause the electronic device to perform a view label cataloging implementation and application method.

[0034] A computer-readable storage medium is provided in the fourth aspect of this application.

[0035] A computer-readable storage medium includes a processor, a memory, and a transceiver, wherein the memory is used to store instructions, the transceiver is used to communicate with other devices, and the processor is used to execute the instructions stored in the memory to cause the electronic device to perform a view label cataloging implementation and application method.

[0036] In summary, this application includes at least one of the following beneficial technical effects:

[0037] 1. The integration of AI-based parsing and manual tagging makes view tagging more complete, and manual tagging is not limited by AI capabilities, thus better meeting users' personalized needs.

[0038] 2. Applications vary depending on the scenario, but the tagging process, tag storage, and retrieval remain consistent across applications. This decoupling of tagging from the application, along with component-based design, facilitates application expansion.

[0039] 3. Improve the accuracy of view retrieval and its practical application capabilities. Attached Figure Description

[0040] Figure 1 This is a flowchart illustrating a view label cataloging implementation and application method according to Embodiment 1 of this application;

[0041] Figure 2 This is a flowchart illustrating step S2 of a view label cataloging implementation and application method according to Embodiment 1 of this application;

[0042] Figure 3 This is a schematic diagram of the architecture of a view label cataloging implementation and application method according to Embodiment 2 of this application;

[0043] Figure 4 This is a schematic diagram of the architecture of a view label cataloging implementation and application method according to Embodiment 3 of this application;

[0044] Figure 5 This is a schematic diagram of the system architecture of a view label cataloging implementation and application method according to Embodiment 3 of this application;

[0045] Figure 6 This is a schematic diagram of the structure of an electronic device according to an embodiment of this application.

[0046] Explanation of the attached diagram labels: 1. Information acquisition module; 2. Parsing module; 3. Combined array generation module; 4. Search and filtering module. Detailed Implementation

[0047] To enable those skilled in the art to better understand the technical solutions in this specification, the technical solutions in the embodiments of this specification will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments.

[0048] In the description of the embodiments in this application, words such as "illustrative," "for example," or "for example" are used to indicate examples, illustrations, or explanations. Any embodiment or design described as "illustrative," "for example," or "for example" in the embodiments of this application should not be construed as being more preferred or advantageous than other embodiments or designs. Rather, the use of words such as "illustrative," "for example," or "for example" is intended to present the relevant concepts in a specific manner.

[0049] In the description of the embodiments of this application, the term "and / or" is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, B existing alone, or A and B existing simultaneously. Furthermore, unless otherwise stated, the term "multiple" means two or more. For example, multiple systems refer to two or more systems, and multiple screen terminals refer to two or more screen terminals. In addition, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the indicated technical features. Thus, a feature defined with "first" or "second" may explicitly or implicitly include one or more of that feature. The terms "comprising," "including," "having," and their variations all mean "including but not limited to," unless otherwise specifically emphasized.

[0050] Scenario: This invention is applied to the field of video surveillance. Through AI view parsing technology and manual labeling, views are tagged and stored in big data. Big data technology is used to fuse and combine these tags; a super search engine enables applications such as target relationship graphs, knowledge generation, and target trajectory tracking based on business scenarios, thereby constructing a big data intelligent system oriented towards practical business operations.

[0051] The proper nouns used in this application include:

[0052] IoT (Internet of Things): A network of a large number of interconnected devices and systems that can collect, exchange, and process data to enable intelligent decision-making and control.

[0053] AI-powered visual analytics is a technology that utilizes artificial intelligence to perform deep analysis and processing of data such as videos and images. This technology can automatically extract useful information, patterns, and structures from large amounts of visual data, and generate understanding and reasoning about video content. AI-powered visual analytics has a wide range of applications, covering data intelligence platforms, intelligent video analytics, edge computing and multimodal analysis, enterprise space digital management, behavioral analysis and early warning systems, video content analysis, and product recognition and consumer behavior insights in the retail sector.

[0054] WebGL: A JavaScript API for rendering interactive 2D and 3D graphics in a web browser without requiring plugins.

[0055] Tag-based super search engines: a system that uses tags to organize, retrieve, and discover content. In such systems, the accuracy of data matching is crucial for providing relevant and personalized search results.

[0056] Tag storage service: A method for managing and organizing resources that allows users to categorize and filter resources using tags, thereby reducing the complexity of resource management. This service typically provides a range of functions, including but not limited to tag creation, binding, retrieval, modification, and deletion. In cloud platforms, tag services can support tag management for various resource types, such as cloud servers, cloud disks, and network resources. Users can categorize and organize these resources using tags, enabling rapid resource retrieval and management. Furthermore, to improve security and management efficiency, cloud platforms usually provide tag quota management, limiting the number of tags that can be created for each project and the number of tags that can be bound to each resource.

[0057] A search algorithm framework based on time series and feature labels: a data retrieval method that combines the time dimension and multi-dimensional features. This framework can be applied to various scenarios, such as financial market analysis, network security monitoring, and user behavior analysis. Key components include data preprocessing, label generation, collision detection, label combination, reliability assessment, result ranking and display, and optimization and iteration. It can efficiently process and analyze large-scale time series datasets, helping users quickly identify important events and patterns. By combining the time dimension and multi-dimensional features, this framework can provide richer and deeper data insights.

[0058] SaaS: A software service delivered via cloud technology that allows users to use software applications without local installation, maintenance, or hardware investment. Users can access the service from anywhere in the world via the internet.

[0059] A classifier is a machine learning algorithm whose main task is to assign a given data item to a predefined category. Classifiers can be used in various fields, such as text classification, image recognition, and sentiment analysis.

[0060] Geographic Information System (GIS): An integrated tool for capturing, storing, analyzing, and managing geospatial data. The core functions of GIS include data creation, management, analysis, and visualization. Through GIS, users can integrate location data (the location of things) with descriptive information (the state of things at that location), providing a foundation for mapping and analysis in the natural sciences and virtually every industry.

[0061] Knowledge generation refers to the process of extracting, integrating, and innovating new knowledge or information from large amounts of data using artificial intelligence technology. This process can be achieved in various ways, including but not limited to natural language processing (NLP), machine learning algorithms, and generative AI.

[0062] Large Language Models (LLMs): These models primarily focus on processing and understanding human language, including tasks such as text analysis, sentiment analysis, language translation, and speech recognition. Large Language Models (LLMs) are built using deep neural networks containing hundreds of billions of weights and trained on large amounts of unlabeled text using self-supervised learning methods.

[0063] Large-scale visual models focus on processing image and video data, enabling tasks such as image recognition, object detection, and semantic segmentation. These models are based on the Transformer architecture and possess powerful feature extraction and recognition capabilities. For example, the EVA model combines the strongest semantic learning (CLIP) with the strongest geometric structure learning (MIM), requiring only the standard ViT model and scaling it to higher dimensions.

[0064] Multimodal large models: Multimodal large models combine information from multiple modalities, such as language and vision, and exhibit stronger capabilities in performing multimodal tasks. These models utilize powerful large language models as their core to perform multimodal tasks, such as image-based writing, the text-image large model LaVIT, and the text-video large model Video-LaVIT. Research and applications of multimodal large models encompass encoders, input projections, large model backbones, and multimodal generators for various modalities, including image, speech, video, and 3D.

[0065] Large-scale computational models focus on how to efficiently train and deploy these massive models. As large-scale models evolve, the demand for computing resources also increases, driving the evolution of computing architectures. The principle behind large-scale models is based on deep learning, utilizing massive amounts of data and computing resources to train neural network models with a large number of parameters.

[0066] Example 1:

[0067] Reference Figure 1 and Figure 2 A method for implementing and applying view label cataloging includes the following steps:

[0068] S1: Get view information;

[0069] Specifically: the view information includes online views and offline views. The online / offline views are accessed to the smart terminal system through the IoT access service to realize the access of view resources. In this embodiment, the smart terminal can be a computer, and in other embodiments it can also be a MUC.

[0070] S2: Perform feature parsing on the view information to obtain and store the parsed label information associated with the view;

[0071] Includes the following steps:

[0072] S21: Input the view information into the AI ​​view parsing model for feature parsing to obtain the parsing label information associated with the view;

[0073] Specifically: The view is forwarded to the AI ​​parsing service via the IoT forwarding service. The AI ​​view parsing model in the AI ​​view parsing technology is used to automatically parse the view and tag the view information, such as tagging the time, location, type, and features of the view target to obtain parsed tag information.

[0074] S22: Obtain manually labeled information and supplement the parsed label information with the manually labeled information;

[0075] It should be noted that, since current AI view analysis technology can only recognize a limited number of objects, it cannot identify specific related features in areas not covered by AI view analysis technology, such as recognizing shooting scores in images. To solve this problem, this application adopts a manual annotation method, in which staff manually annotate the relevant features in the view that cannot be recognized by AI, obtaining manual annotation information, which is then added to the analysis label information to update the analysis label information.

[0076] In other embodiments, the AI ​​view resolution model can be trained and iteratively strengthened to enable the AI ​​view resolution technology to recognize these relevant features.

[0077] It's important to further explain here that while the smart terminal performs feature parsing on the view information, the IoT forwards the view to the playback plugin. After receiving the streaming data, the playback plugin performs video frame decoding and sends the decoded streaming data back to the client. Upon receiving the decoded streaming data, the client calls the WebGL interface to render the image.

[0078] S3: Stores and parses tag information;

[0079] Specifically: Tag storage is used to store tags. Tag storage services can store parsed tag information to the cloud platform system, making it easy for various clients to retrieve and parse the tag information online. At the same time, it can aggregate resources from various places, complete resource integration, and facilitate the full utilization of resources.

[0080] Tag storage services enable unified management of storage space across all centralized storage resources. Based on resource information reported by network-connected storage resources, the system calculates the total and free storage capacity of the cluster. This can be understood as follows: if a monitored storage space can store a certain number of hours of view resources, the cloud platform system stores the parsed tag information corresponding to the total amount of view resources at that location. It also allocates corresponding storage space based on the number of views within those view resources to store the parsed tag information. When the monitored storage space is full and new view resources are received, causing the oldest stored view resources to be deleted, the cloud platform system also performs corresponding operations on the stored parsed tag information, thereby allocating appropriate storage resources for tag storage. Simultaneously, the size of the storage pool is dynamically adjusted, allowing for flexible expansion and contraction.

[0081] It should be noted that the cloud platform system uses a discrete storage algorithm to store the parsed tag information, storing it on different nodes in a sharded manner, thereby effectively ensuring data security.

[0082] After storing the parsed tag information in the storage space, the cloud platform system associates the parsed tag information with the corresponding view information. Specifically, this association can be achieved through pointers or by adding information related to the storage location of the view information to the parsed tag information.

[0083] S4: Take the time information in the stored parsed label information as one hinge label line, and the remaining feature labels in the parsed label information as another hinge label line, and perform cleaning, collision and combination to obtain a label combination array.

[0084] Specifically: After parsing and storing the tag information, a tag super search engine cluster is used to manage all nodes. These nodes can be understood as monitoring the uploaded parsed tag information, concurrently responding to user searches, and providing integrated, high-concurrency, and fast-response indexing services. Simultaneously, the tag super search engine also provides image indexes, tag indexes, time indexes, and audio indexes for target searches. To make the search results more accurate, this application arrays the parsed tag information. Specifically, the time information in the stored parsed tag information is used as one hinge tag line, and the remaining feature tags in the parsed tag information (such as user behavior, device information, etc.) are used as another hinge tag line. The time information mentioned here, in this embodiment, refers to the specific time the view information was captured, where the time information is in seconds. Each time unit (i.e., each second) is considered an independent tag, representing the position of the data point in the time dimension. For example, if considering a one-hour time range, there will be 3600 time tags. The cleaning, collision, and combination between the two hinge label lines can be processed by a search algorithm framework based on time series and feature labels to obtain N sets of label combination arrays, which are then stored in the storage space of the cloud platform.

[0085] S5: In response to search criteria, filter all tag combination arrays and output the filtered information so that the client can retrieve the associated view information;

[0086] Specifically: The original search criteria are sent by personnel through a client and input into the cloud platform system via the tag super search engine; the super tag search engine generates field search criteria in response to external search information; the field search criteria are combined with the built-in reliability retrieval algorithm to filter out reliable tag combination arrays within a preset reliability range from all stored tag combination arrays; the built-in reliability retrieval algorithm can be understood as the built-in reliability algorithm of the tag super search engine, and the preset reliability range can be understood as selecting a preset number of tag combination arrays as reliable tag combination arrays in descending order of reliability, the preset number of which can be set by personnel according to actual conditions; then, filtering information is generated based on the relevant information of the storage location of the corresponding view information in the filtered reliable tag combination arrays, and the client receives the filtering information and retrieves the relevant view for display.

[0087] S6: Match the location information in the parsed label information corresponding to the reliable label combination array into the GIS to generate the motion trajectory of the filtered object in the filter information.

[0088] Specifically: After retrieving the reliable tag combination array, the platform system will find the location tag information in the parsed tag information corresponding to the reliable tag combination array. It should be noted that the location tag information can be location coordinates or address text information. The location information is imported into GIS to obtain the specific location of the search target in GIS. Then, based on the time information of multiple reliable tag combination arrays, the coordinate positions are concatenated in chronological order to obtain the coordinate transformation trajectory of the target in GIS, i.e., the target trajectory. The system will also find the corresponding view information associated with multiple reliable tag combination arrays and send the target trajectory and the view information corresponding to each coordinate in the target trajectory to the client.

[0089] To facilitate understanding, the following scenario is illustrated: For example, when monitoring a car with a certain license plate, the car passes through multiple monitoring points along its route from the origin to the destination. The origin and destination are also monitored. When searching for the car with that license plate, the client's GIS will display the car's trajectory from the origin to the destination. Clicking on a coordinate point in the trajectory will display a view of the car passing through that monitoring point or a video composed of consecutive views.

[0090] Example 2:

[0091] Reference Figure 3 The difference between this embodiment and Embodiment 1 is that:

[0092] After filtering all the tag combination arrays in response to the search criteria, the following steps are performed: Based on the preset classifier, obtain the same attribute related view information and same location related view information of the filtered objects in the filtered information. Based on the same attribute related view information and same location related view information, generate the same attribute related map and same location related map.

[0093] Specifically: In the SaaS application interface, the smart device hosting the SaaS application interface can be either the device hosting the cloud platform system or an external device. Based on the same location of the target within the system, such as a 100-meter boundary (this range can be customized), a corresponding target relationship graph view is generated and displayed. Specifically, based on the location tag information of the filtered objects, the parsed tag information of the same location is found, and related view information is searched to obtain related view information for the same location; similarly, related views with the same attribute are obtained in the same way. It should be noted that related graphs with the same attribute may include political figures of a certain level, equipment of a certain level, etc. If the current search target is a certain aircraft, the generated related view with the same attribute may include views of similar aircraft in the system. It should also be noted that graph generation technology is an existing technology that converts structured and unstructured data into computer-understandable knowledge representations. It constructs knowledge graphs by describing entities, attributes, and relationships between entities. The generated related graphs with the same attribute and related graphs with the same location will be displayed in the SaaS application interface for users to view.

[0094] Example 3:

[0095] Reference Figure 4 The difference between this embodiment and embodiment one is that, in response to the search conditions, after filtering all the tag combination arrays, the following steps are performed: outputting the parsed tag information and associated view information to the intelligent interaction system so that the intelligent interaction system can generate a knowledge graph based on the parsed tag information and associated view information.

[0096] Specifically: After the cloud platform system outputs parsed tag information and associated view information to the intelligent interaction system, the intelligent interaction system, based on large models such as language large model, visual large model, multimodal large model, and computational large model, generates knowledge from the returned parsed tag information and associated view information to obtain the corresponding knowledge graph. It should be noted that the intelligent interaction system can communicate with the client. The associated view information can be the storage location information of related views, so that the intelligent interaction system can retrieve the corresponding views. After the user inputs the requirements into the SaaS application in the client, the client communicates with the intelligent interaction system. The intelligent interaction system calls its own general corpus and domain corpus to analyze the user's business corpus, obtains the search conditions, and searches on the cloud platform system through the super search engine. After obtaining the parsed tag information and associated view information, the intelligent interaction system generates the knowledge graph and transmits it to the SaaS application in the client for personnel to view.

[0097] Reference Figure 5 A system based on view label cataloging implementation and application methods includes:

[0098] Information acquisition module 1 is used to acquire view information;

[0099] Parsing module 2 is used to perform feature parsing on view information, obtain parsing label information associated with the view, and store it.

[0100] The combined array generation module 3 is used to take the time information in the stored parsed tag information as a hinge tag line and any other feature tag in the parsed tag information as another hinge tag line, and perform cleaning, collision and combination to obtain a tag combined array.

[0101] The search and filtering module 4 is used to filter all tag combination arrays in response to search criteria and output the filtered information so that the client can retrieve related view information.

[0102] See Figure 6 The present application provides a schematic diagram of the structure of an electronic device. The electronic device 1000 may include: at least one processor 1001, at least one network interface 1004, a user interface 1003, a memory 1005, and at least one communication bus 1002.

[0103] The communication bus 1002 is used to realize the connection and communication between these components.

[0104] The user interface 1003 may include a display screen and a camera. Optionally, the user interface 1003 may also include a standard wired interface and a wireless interface.

[0105] The network interface 1004 may optionally include a standard wired interface or a wireless interface (such as a Wi-Fi interface).

[0106] The processor 1001 may include one or more processing cores. The processor 1001 connects to various parts within the server 1000 using various interfaces and lines, and performs various functions and processes data by running or executing instructions, programs, code sets, or instruction sets stored in the memory 1005, and by calling data stored in the memory 1005. Optionally, the processor 1001 may be implemented using at least one hardware form of Digital Signal Processing (DSP), Field-Programmable Gate Array (FPGA), or Programmable Logic Array (PLA). The processor 1001 may integrate one or a combination of several of the following: Central Processing Unit (CPU), Graphics Processing Unit (GPU), and modem. The CPU primarily handles the operating system, user interface, and applications; the GPU is responsible for rendering and drawing the content required for display; and the modem handles wireless communication. It is understood that the modem may also be implemented as a separate chip without being integrated into the processor 1001.

[0107] The memory 1005 may include random access memory (RAM) or read-only memory. Optionally, the memory 1005 may include a non-transitory computer-readable storage medium. The memory 1005 may be used to store instructions, programs, code, code sets, or instruction sets. The memory 1005 may include a program storage area and a data storage area, wherein the program storage area may store instructions for implementing an operating system, instructions for at least one function (such as touch function, sound playback function, image playback function, etc.), instructions for implementing the above-described method embodiments, etc.; the data storage area may store data involved in the above-described method embodiments, etc. Optionally, the memory 1005 may also be at least one storage device located remotely from the aforementioned processor 1001. The memory 1005, as a computer storage medium, may include an operating system, a network communication module, a user interface module, and an application program for a view label cataloging implementation and application method.

[0108] It should be noted that the above embodiments of the apparatus are only illustrated by the division of the above functional modules. In practical applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above. In addition, the apparatus and method embodiments provided in the above embodiments belong to the same concept, and the specific implementation process can be found in the method embodiments, which will not be repeated here.

[0109] In the electronic device 1000 shown in the figure, the user interface 1003 is mainly used to provide an input interface for the user and obtain the user input data; while the processor 1001 can be used to call an application program stored in the memory 1005 that implements and applies a view label cataloging method. When executed by one or more processors, the electronic device performs one or more of the methods described in the above embodiments.

[0110] An electronic device readable storage medium stores instructions. When executed by one or more processors, these instructions cause the electronic device to perform a view label cataloging implementation and application method as described in one or more of the above embodiments.

[0111] Those skilled in the art will clearly understand that the technical solutions of this application can be implemented using software and / or hardware. In this specification, "unit" and "module" refer to software and / or hardware capable of independently performing or cooperating with other components to perform specific functions. Hardware may include, for example, a Field-Programmable Gate Array (FPGA), an Integrated Circuit (IC), etc.

[0112] It should be noted that, for the sake of simplicity, the foregoing method embodiments are all described as a series of actions. However, those skilled in the art should understand that this application is not limited to the described order of actions, as some steps may be performed in other orders or simultaneously according to this application. Furthermore, those skilled in the art should also understand that the embodiments described in the specification are preferred embodiments, and the actions and modules involved are not necessarily essential to this application.

[0113] In the above embodiments, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions in other embodiments.

[0114] In the several embodiments provided in this application, it should be understood that the disclosed apparatus can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some service interface; the indirect coupling or communication connection between devices or units may be electrical or other forms.

[0115] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0116] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0117] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage device (CMD). Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a memory and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned memory includes various media capable of storing program code, such as USB flash drives, read-only memory (ROM), random access memory (RAM), portable hard drives, magnetic disks, or optical disks.

[0118] Those skilled in the art will understand that all or part of the steps in the various methods of the above embodiments can be implemented by a program instructing related hardware. The program can be stored in a computer-readable storage medium, which may include: a flash drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk, etc.

[0119] The foregoing description is merely an exemplary embodiment of this disclosure and should not be construed as limiting the scope of this disclosure. Any equivalent changes and modifications made in accordance with the teachings of this disclosure shall still fall within the scope of this disclosure. Other embodiments of this disclosure will be readily apparent to those skilled in the art upon consideration of the specification and practice of the disclosure herein. This application is intended to cover any variations, uses, or adaptations of this disclosure that follow the general principles of this disclosure and include common knowledge or customary techniques in the art not described herein. The specification and embodiments are to be considered exemplary only, and the scope and spirit of this disclosure are defined by the claims.

Claims

1. A method for implementing and applying view label cataloging, characterized in that, Includes the following steps: Get view information; The view information is parsed to obtain and store the parsed label information associated with the view; the parsed label information includes time information and several other feature labels. The time information in the stored parsed label information is used as one hinge label line, and the remaining feature labels in the parsed label information are used as another hinge label line. The two hinge label lines are cleaned, collided, and combined to obtain a label combination array and stored. In response to search criteria, the system filters all stored arrays of tag combinations and outputs the filtered information so that the client can retrieve the associated view information. The view information is analyzed for features to obtain the parsed label information associated with the view, including: The view information is input into the AI ​​view parsing model for feature parsing to obtain parsing label information associated with the view; After inputting view information into the AI ​​view parsing model for feature parsing and obtaining parsed label information associated with the view, the following are included: Obtain manually labeled information and supplement the parsed label information with the manually labeled information; In response to search criteria, the array of tag combinations is filtered, including: The super-tag search engine generates field search criteria in response to external search information; Based on the field search criteria and the built-in reliability search algorithm, reliable tag combination arrays within the preset reliability range are selected from all tag combination arrays. In response to search criteria, all tag combination arrays are filtered, and the filtered information is output so that the client can retrieve the associated view information. This also includes: The location information in the parsed label information corresponding to the reliable label combination array is matched into the GIS to generate the motion trajectory of the filtered object in the filter information. In response to search criteria, all tag combination arrays are filtered, and the filtered information is output so that the client can retrieve the associated view information. This also includes: Based on the preset classifier, obtain the view information related to the same attribute and the view information related to the same location of the filtered objects in the filtering information; Generate attribute-related maps and location-related maps based on view information related to the same attribute and view information related to the same location; After filtering all tag combination arrays in response to search criteria and outputting the filtered information so that the client can retrieve the associated view information, it also includes: Output the parsed label information and associated view information to the intelligent interaction system so that the intelligent interaction system can generate knowledge based on the parsed label information and associated view information.

2. A view label cataloging implementation and application system, characterized in that, include: Information acquisition module (1), used to acquire view information; The parsing module (2) is used to perform feature parsing on the view information, obtain the parsing label information associated with the view, and store it. The combined array generation module (3) is used to take the time information in the stored parsed tag information as a hinge tag line and any other feature tag in the parsed tag information as another hinge tag line, and clean, collide and combine them to obtain a tag combined array. The retrieval and filtering module (4) is used to filter all the tag combination arrays in response to the retrieval conditions and output the filtering information so that the client can retrieve the associated view information; The view information is analyzed for features to obtain the parsed label information associated with the view, including: The view information is input into the AI ​​view parsing model for feature parsing to obtain parsing label information associated with the view; After inputting view information into the AI ​​view parsing model for feature parsing and obtaining parsed label information associated with the view, the following are included: Obtain manually labeled information and supplement the parsed label information with the manually labeled information; In response to search criteria, the array of tag combinations is filtered, including: The super-tag search engine generates field search criteria in response to external search information; Based on the field search criteria and the built-in reliability search algorithm, reliable tag combination arrays within the preset reliability range are selected from all tag combination arrays. In response to search criteria, all tag combination arrays are filtered, and the filtered information is output so that the client can retrieve the associated view information. This also includes: The location information in the parsed label information corresponding to the reliable label combination array is matched into the GIS to generate the motion trajectory of the filtered object in the filter information. In response to search criteria, all tag combination arrays are filtered, and the filtered information is output so that the client can retrieve the associated view information. This also includes: Based on the preset classifier, obtain the view information related to the same attribute and the view information related to the same location of the filtered objects in the filtering information; Generate attribute-related maps and location-related maps based on view information related to the same attribute and view information related to the same location; After filtering all tag combination arrays in response to search criteria and outputting the filtered information so that the client can retrieve the associated view information, it also includes: Output the parsed label information and associated view information to the intelligent interaction system so that the intelligent interaction system can generate knowledge based on the parsed label information and associated view information.

3. An electronic device, characterized in that, The device includes a processor, a memory, and a transceiver. The memory is used to store instructions, the transceiver is used to communicate with other devices, and the processor is used to execute the instructions stored in the memory to cause the electronic device to perform the method as described in claim 1.

4. A computer-readable storage medium, characterized in that, The device includes a processor, a memory, and a transceiver. The memory is used to store instructions, the transceiver is used to communicate with other devices, and the processor is used to execute the instructions stored in the memory to cause the electronic device to perform the method as described in claim 1.

Images