Data management method and apparatus, data retrieval method and apparatus, and storage medium

Abstract

The present disclosure relates to the technical field of data processing, and provides a data management method and apparatus, a data retrieval method and apparatus, and a storage medium. The data management method of the present disclosure comprises: acquiring a visual feature in environmental data; storing the visual feature as a first index into an ontology database; and storing, as a second index, a non-visual feature in the environmental data into the ontology database.

Description

Data management, retrieval method and device and storage medium

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS

[0002] This application is based on the application with CN application number 202311819794.X and application date December 27, 2023, and claims its priority. The disclosed content of the CN application is hereby introduced as a whole into this application. Technical Field

[0003] The present disclosure relates to the field of data processing technology, and in particular to a data management and retrieval method and device and a storage medium. Background Art

[0004] When a computer interacts with the environment in the same environment, it first needs to collect full image data of the environment, then perform object recognition on the image data before proceeding to the next step between the computer and the object. This operation method places high demands on the computer's computing power.

[0005] Summary of the Invention

[0006] One purpose of the present disclosure is to improve the efficiency of object recognition in the environment and reduce the computing power required.

[0007] According to one aspect of some embodiments of the present disclosure, a data management method is proposed, including: obtaining visual features in environmental data; storing the visual features as a first index in an ontology database; and storing non-visual features in the environmental data as a second index in the ontology database.

[0008] In some embodiments, the visual feature includes at least one of a color feature and a shape feature.

[0009] In some embodiments, obtaining visual features in environmental data includes: performing color extraction on the collected image data using a color space model in an image sensor used to collect environmental data to obtain color features, wherein the visual features include color features.

[0010] In some embodiments, obtaining visual features in environmental data includes: extracting the shape of an object in the environmental data using a predetermined contour recognition algorithm to obtain shape features, wherein the visual features include shape features.

[0011] In some embodiments, obtaining the visualization feature in the environmental data includes: extracting the supplementary visualization feature from the environmental data according to a preset supplementary visualization feature type, wherein the visualization feature includes the supplementary visualization feature.

[0012] In some embodiments, storing the visualization feature as the first index in the ontology database includes: storing each visualization feature as the first index in the ontology database, wherein the number of types of the visualization features is greater than one.

[0013] In some embodiments, storing each type of visualization feature as a first index in the ontology database includes: sorting the categories in order from low to high according to the feature complexity coefficient of the visualization feature of each category; generating a first index based on the sorting result, wherein the retrieval priority of the visualization features of the categories with higher sorting is higher than that of the visualization features of the categories with lower sorting.

[0014] In some embodiments, storing the visualization feature as the first index in the ontology database further includes: updating the visualization feature according to the number of visualization features of each category in the visualization feature, wherein the categories are sorted in order from low to high according to the feature complexity coefficient for the updated visualization feature.

[0015] In some embodiments, updating the visualization feature according to the number of visualization features of each category in the visualization feature includes updating the visualization feature by deleting the visualization feature of the category where the number of visualization features is 1 in the visualization feature.

[0016] In some embodiments, if the number of each type of visualization feature in the visualization feature is 1, the first index does not exist, and the second index constitutes a single-type element index sequence for storage.

[0017] In some embodiments, the feature complexity coefficient of a category is determined according to the number of different visual features of the corresponding category. The greater the number, the higher the feature complexity coefficient.

[0018] In some embodiments, the second index has a lower retrieval priority than the first index.

[0019] According to one aspect of some embodiments of the present disclosure, a retrieval method is proposed, comprising: obtaining feature information of a retrieval object; querying the feature information in an ontology database to obtain retrieval results, wherein the ontology database is generated according to any one of the data management methods mentioned above.

[0020] In some embodiments, querying the feature information in the ontology database to obtain the search results includes: querying the feature information according to the index with the highest priority in the ontology database to obtain a candidate range; within the candidate range, narrowing the candidate range by querying the feature information in descending order of the priority of the index in the ontology database until the search results are obtained.

[0021] According to one aspect of some embodiments of the present disclosure, a data management device is proposed, including: a visual feature acquisition unit, configured to acquire visual features in environmental data; a first index determination unit, configured to store the visual features as a first index in an ontology database; and a second index determination unit, configured to store non-visual features in the environmental data as a second index in the ontology database.

[0022] According to one aspect of some embodiments of the present disclosure, a retrieval device is proposed, comprising: an information acquisition unit, configured to acquire feature information of a retrieval object; and a query unit, configured to query the feature information in an ontology database to obtain a retrieval result, wherein the ontology database is generated according to any one of the data management methods mentioned above.

[0023] According to one aspect of some embodiments of the present disclosure, a data processing device is proposed, including: a memory; and a processor coupled to the memory, wherein the processor is configured to execute any one of the above-mentioned data management methods based on instructions stored in the memory.

[0024] According to one aspect of some embodiments of the present disclosure, a computer-readable storage medium is provided, on which computer program instructions are stored. When the instructions are executed by a processor, any one of the data management methods mentioned above is implemented.

[0025] According to one aspect of some embodiments of the present disclosure, a computer program is proposed, configured to enable a processor to execute any one of the data management methods mentioned above.

[0026] According to one aspect of some embodiments of the present disclosure, a computer program product is provided, comprising a computer program or instructions, wherein when the computer program or instructions are executed by a processor, any one of the data management methods mentioned above is implemented. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] The drawings described herein are used to provide a further understanding of the present disclosure and constitute a part of the present disclosure. The exemplary embodiments of the present disclosure and their descriptions are used to explain the present disclosure and do not constitute an improper limitation of the present disclosure. In the drawings:

[0028] FIG1 is a flowchart of some embodiments of the data management method disclosed herein.

[0029] FIG2 is a flowchart of other embodiments of the data management method disclosed herein.

[0030] FIG3 is a schematic diagram of some embodiments of the data management method disclosed herein.

[0031] FIG4 is a schematic diagram of some further embodiments of the data management method disclosed herein.

[0032] FIG5 is a flowchart of some embodiments of the retrieval method of the present disclosure.

[0033] FIG6 is a schematic diagram of some embodiments of the data management device disclosed herein.

[0034] FIG7 is a schematic diagram of some embodiments of the retrieval device disclosed herein.

[0035] FIG8 is a schematic diagram of some embodiments of a data processing device according to the present disclosure.

[0036] FIG9 is a schematic diagram of some other embodiments of the data processing device disclosed herein. DETAILED DESCRIPTION

[0037] The technical solution of the present disclosure is further described in detail below through the accompanying drawings and examples.

[0038] A flowchart of some embodiments of the data management method of the present disclosure is shown in FIG1 .

[0039] In step S12, visual features are obtained from the environmental data. In some embodiments, visual features are features that can be seen in the environment video or image, such as the color or shape of the object itself, or can be extracted from the video, image, or other information. In some embodiments, visual features can be obtained during the feature extraction process using a preset HSV color space model or image feature extraction tools such as approxPolyDP.

[0040] In some embodiments, the types of visualization features may be preset to facilitate targeted feature extraction and storage.

[0041] In some embodiments, the visual feature includes at least one of a color feature and a shape feature.

[0042] In some embodiments, color features are extracted from the collected image data using a color space model in an image sensor used to collect environmental data. This method allows the image sensor to perform color feature extraction, reducing the burden of subsequent data processing.

[0043] In some embodiments, the shape of the object in the environmental data is extracted by a predetermined contour recognition algorithm (eg, approxPolyDP algorithm) to obtain shape features, wherein the visualization features include shape features.

[0044] In some embodiments, the visualization features may also include other features with visualization characteristics (hereinafter referred to as supplementary visualization features). In some embodiments, the specific feature types in the supplementary visualization features can be expanded and supplemented as needed to improve the scalability of data management. In some embodiments, supplementary visualization features are extracted from the environmental data based on the preset supplementary visualization feature types.

[0045] In step S14, the visual feature is stored as a first index in the ontology database. In some embodiments, the visual feature can be encoded and stored, for example, using Huffman coding, so that the feature encoding table includes the visual feature.

[0046] In some embodiments, each visualization feature is stored in the ontology database as a first index. In some embodiments, if the number of visualization feature types is greater than 1, the number of first indexes is greater than 1, and the number of first indexes matches the number of visualization feature types.

[0047] In some embodiments, the feature complexity coefficient of each type of visualization feature is first determined, and then the categories are sorted in order from low to high according to the feature complexity coefficient of each category of visualization features, and a first index is generated based on the sorting result. The retrieval priority of the visualization features of the categories with higher ranking is higher than that of the visualization features of the categories with lower ranking. Through such a method, the features with lower retrieval difficulty can be ranked higher, and the features with higher retrieval difficulty can be ranked lower. During retrieval, the features with lower retrieval difficulty are given priority, thereby reducing the retrieval complexity and the amount of computation, and further reducing the computing power requirements.

[0048] In some embodiments, the feature complexity coefficient of the above categories is determined according to the number of different visual features of the corresponding categories, and the greater the number, the higher the feature complexity coefficient. In some embodiments, the feature complexity coefficient is a positive integer greater than or equal to 1. For example, if there are three colors in the environment, the feature load coefficient of the color feature is 3. In some embodiments, taking the shape feature as an example, if divided according to regular shapes and irregular shapes, for regular shapes, it is divided according to the number of regular shapes; if it is an irregular shape, the number of edges of the irregular shape is determined, and the ones with the same number of edges are classified into one type; if the objects in the current environment are all irregular shapes, and the number of edges of the irregular shapes is the same, the feature complexity coefficient of the shape feature is 1; if the current environment is all irregular shapes, and the number of irregular shape edges is inconsistent (for example, there are 3 figures with 5 edges and 2 figures with 8 edges), the irregular shapes with fewer edges are ranked first and the ones with more edges are ranked last, and then the shape feature coefficient = 1 + 2 (there are two irregular shapes with the same number of edges) + 2 (the number of irregular shapes with the least number of edges) = 5. By using this method, the retrieval complexity of the visual features in the environment can be quantified, thereby improving the reliability of the first index sorting.

[0049] In some embodiments, a flowchart of generating a first index based on visualization features is shown in FIG2 .

[0050] In step 241, the visualization features are updated by deleting the visualization features of the type with a number of 1. In some embodiments, if the visualization features do not include a type with a number of 1, step 241 is skipped and step 242 is executed. In some embodiments, if the number of features of all types in the visualization features is 1, the process of generating the first index is skipped and step S16 is executed.

[0051] In step 242 , the categories are sorted in descending order of the feature complexity coefficient according to the feature complexity coefficient of each category of visualization features.

[0052] In step 243, a first index is generated based on the sorting results. In some embodiments, the visual features of the corresponding categories are encoded and stored in descending order of the sorting results. When the first index is subsequently used, the visual features of the categories ranked higher in the sorting results are retrieved with higher priority than the visual features of the categories ranked lower in the sorting results.

[0053] In some embodiments, in steps 242 and 243, if it is determined that there are species with a feature complexity coefficient greater than 1, the species corresponding to the feature complexity coefficient closest to 1 is found and used as the first index of the first category; based on the feature complexity coefficient corresponding to the first index of the first category, the species that is greater than and closest to the basis is found and used as the first index of the second category; and according to the rule of increasing feature complexity coefficients, the first index of the third category is gradually determined, and the first index of the nth category is gradually determined, where n is a positive integer. The first indexes of each category are stored in the ontology database in the order in which the first indexes were created.

[0054] Through this method, we can first exclude feature types that have no retrieval value, which is beneficial to improving the subsequent retrieval efficiency; when the feature complexity coefficient is greater than 1, according to the binary tree calculation logic, the smaller the coefficient, the higher the calculation efficiency. By setting the features with small feature complexity coefficients to be ranked higher, they can be retrieved first, which is beneficial to further improve the retrieval efficiency.

[0055] In step S16, the remaining features after extracting the visual features are treated as non-visual features, and the non-visual features in the environmental data are stored as a second index in the ontology database. Non-visual features are features that cannot be extracted from visual information such as video images, such as refrigerator model and water cup volume.

[0056] In some embodiments, the non-visual features may be encoded and stored, for example, using Huffman coding, so that the feature encoding table includes the non-visual features. In some embodiments, the retrieval priority of the second index is lower than that of the first index.

[0057] In some embodiments, if the number of visual features of each type in the visual features is 1, the first index does not exist, and the second index constitutes a single-category element index sequence storage, and subsequent retrieval relies on non-visual features to avoid affecting the retrieval results when the visual features are insufficient.

[0058] Through the method in the embodiment shown above, the object features in the environment ontology can be distinguished between visualization and non-visualization, and the visualization features and non-visualization features can be encoded separately and indexed and stored together in unstructured data, so that the environment data has a classification index divided into visualization features and non-visualization features. Since the types and quantities of objects in the same environment are limited, by using visualization features to quickly classify and distinguish objects in the same environment, while expanding and improving the environment ontology database, the efficiency of the interaction between the computer and the environment is improved and the computing power requirement is reduced.

[0059] In some embodiments, as shown in FIG3 and FIG4, after collecting environmental data through the environmental data collector, the data is transcoded and pre-processed, and color features, shape features, and other visual features are obtained through visual feature extraction (for example, using a feature recognition model). The types and quantities of other visual features are not limited and can be set or adjusted as needed. The number of color feature types COL is obtained based on the extracted features. num , the number of shape feature types SH num , and other visual features, the number of types OT num , the number of types is used as the characteristic complexity coefficient.

[0060] As shown in Figure 3, in COL num SH num ,OT num When both are greater than 1, in the environmental ontology database, the features in the first index are stored in ascending order of number of types, and the second index follows the first index.

[0061] As shown in Figure 4, in COL num SH num ,OT num If there is an item equal to 1 in , then in the environmental ontology database, the feature with the number of types of 1 is not stored, and the other features in the first index are stored in the order of the number of types from small to large, and the second index follows the first index.

[0062] Based on the method in the embodiment shown above, it is possible to construct linearized multivariate features based on the constructed general environmental knowledge base and specific environmental scenarios as expanded knowledge of the general environmental knowledge base; and encode and store them in combination with visual and non-visual features, where non-visual features serve as a supplement to visual features, which is conducive to improving the accuracy of subsequent retrieval.

[0063] FIG5 is a flowchart of some embodiments of the retrieval method disclosed herein.

[0064] In step 501, characteristic information of the search object is obtained, for example, a water cup, white, 200 ml; a table, black, 100 cm×60 cm. In some embodiments, the characteristic information of the search object can be obtained through request information from a search requester.

[0065] Furthermore, the feature information is queried in the ontology database to obtain search results. The ontology database is the ontology database mentioned above that stores the first index and the second index. In this ontology database, the matching priority of the first index is higher than that of the second index. In some embodiments, the specific operation steps can be shown as steps 502 to 504 in Figure 5.

[0066] In step 502, feature information is searched according to the index with the highest priority in the ontology database to obtain a candidate range.

[0067] In step 503, it is determined whether the search object is locked. If the search object is locked, step 505 is executed; if the search object is not locked, step 504 is executed.

[0068] In step 504, within the candidate range, the candidate range is narrowed down by querying feature information in descending order of priority of the indexes in the ontology database.

[0069] In some embodiments, in step 502, the candidate range is obtained based on the query feature information of the first category first index in the ontology database. If the candidate range is large and the search target cannot be narrowed down, in step 504, the candidate range is narrowed down based on the query feature information of the second category first index in the ontology database. If the search target still cannot be narrowed down, the search is continued in the order of the third category first index, ..., the nth first index, and the second index until the search target is narrowed down or the search based on the second index is completed, at which point the search ends.

[0070] In step 505, the search result is fed back to the search requester, and the search result includes the locked search object information.

[0071] Through the method in the embodiment shown above, the linear computational complexity O(n) can be converted into a binary search computational complexity O(log n), thereby reducing the difficulty of data query and improving the efficiency of object recognition in image data; during retrieval, objects in the current environment can be quickly classified and retrieved through visual feature coding. If the same item has multiple specifications, it can be quickly retrieved through non-visual specification parameter coding, etc., so as to achieve the purpose of quickly locking the target object.

[0072] A schematic diagram of some embodiments of the data management device disclosed herein is shown in FIG6 .

[0073] Visual feature acquisition unit 611 can acquire visual features from environmental data. In some embodiments, the types of visual features can be pre-set to facilitate targeted feature extraction and storage. In some embodiments, the visual features include at least one of color features and shape features. In some embodiments, the visual features can be acquired according to the method of any of the embodiments in step S12.

[0074] The first index determination unit 612 can store the visual feature as the first index in the ontology database. In some embodiments, the visual feature can be encoded and stored, for example, using Huffman coding, so that the visual feature is included in the feature encoding table. In some embodiments, the first index can be determined and stored in the ontology database according to step S14 or the method in any of the embodiments shown in FIG. 2 .

[0075] The second index determination unit 613 can store the non-visual features in the environmental data as second indexes in the ontology database. In some embodiments, the second index determination unit 613 can store the remaining features after extracting the visual features as non-visual features in the ontology database. Non-visual features are features that cannot be extracted from visual information such as video images. In some embodiments, the non-visual features can be encoded and stored, for example, using Huffman coding.

[0076] Such a data management device can distinguish the features of objects in the environment ontology into visual and non-visual, encode the visual features and non-visual features separately, create indexes and store them together in unstructured data, so that the environment data has a classification index divided into visual features and non-visual features. Since the types and quantities of objects in the same environment are limited, by using visual features to quickly classify and distinguish objects in the same environment, the efficiency of interaction between computers and the environment is improved while expanding and improving the environment ontology database.

[0077] FIG7 shows a schematic diagram of some embodiments of the retrieval device disclosed herein.

[0078] The information acquisition unit 721 can acquire feature information of a search target.

[0079] The query unit 722 can query the feature information in the ontology database to obtain search results. The ontology database is generated according to any of the data management methods mentioned above. The ontology database includes a first index and a second index. In some embodiments, the first index in the ontology database has a higher matching priority than the second index. In some embodiments, the query unit 722 can obtain search results by executing the method of steps 502 to 504 above.

[0080] Such a retrieval device can convert linear computational complexity into a binary search computational complexity problem, thereby reducing the difficulty of data query and improving the efficiency of object recognition in image data; during retrieval, objects in the current environment can be quickly classified and retrieved through visual feature coding. If the same item has multiple specifications, it can be quickly retrieved through non-visual specification parameter coding, so as to achieve the purpose of quickly locking the target object.

[0081] A structural diagram of an embodiment of a data processing device disclosed in the present invention is shown in FIG8 . The data processing device includes a memory 801 and a processor 802 . The memory 801 may be a disk, a flash memory, or any other non-volatile storage medium. The memory is used to store instructions in the corresponding embodiments of the data management method or the retrieval method described above. The processor 802 is coupled to the memory 801 and may be implemented as one or more integrated circuits, such as a microprocessor or a microcontroller. The processor 802 is used to execute instructions stored in the memory, which can improve the efficiency of object recognition in the environment and reduce the amount of computing power required.

[0082] In one embodiment, as shown in FIG9 , a data processing device 900 includes a memory 901 and a processor 902. The processor 902 is coupled to the memory 901 via a BUS 903. The data processing device 900 can also be connected to an external storage device 905 via a storage interface 904 to access external data, and can also be connected to a network or another computer system (not shown) via a network interface 906. A detailed description is omitted here.

[0083] In this embodiment, by storing data instructions in a memory and then processing the instructions through a processor, the efficiency of object recognition in the environment can be improved and the computing power requirement can be reduced.

[0084] In another embodiment, a computer-readable storage medium stores computer program instructions thereon, which, when executed by a processor, implement the steps of the method in the corresponding embodiment of the data management method or retrieval method. Those skilled in the art will appreciate that the embodiments of the present disclosure may be provided as methods, devices, or computer program products. Therefore, the present disclosure may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Furthermore, the present disclosure may take the form of a computer program product implemented on one or more computer-usable non-transient storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0085] The present disclosure is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems) and computer program products according to the embodiments of the present disclosure. It should be understood that each process and / or box in the flowchart and / or block diagram and the combination of the processes and / or boxes in the flowchart and / or block diagram can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate a device for implementing the functions specified in one or more processes in the flowchart and / or one or more boxes in the block diagram.

[0086] These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce a product including an instruction device that implements the functions specified in one or more processes in the flowchart and / or one or more boxes in the block diagram.

[0087] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, so that the instructions executed on the computer or other programmable device provide steps for implementing the functions specified in one or more processes in the flowchart and / or one or more boxes in the block diagram.

[0088] The present disclosure has been described in detail. To avoid obscuring the concept of the present disclosure, some details known in the art have not been described. Based on the above description, those skilled in the art can fully understand how to implement the technical solutions disclosed herein.

[0089] The methods and apparatus of the present disclosure may be implemented in many ways. For example, the methods and apparatus of the present disclosure may be implemented by software, hardware, firmware, or any combination of software, hardware, and firmware. The above order of steps for the method is for illustration only, and the steps of the method of the present disclosure are not limited to the order specifically described above, unless otherwise specifically stated. In addition, in some embodiments, the present disclosure may also be implemented as programs recorded in a recording medium, which include machine-readable instructions for implementing the methods according to the present disclosure. Therefore, the present disclosure also covers recording media that store programs for executing the methods according to the present disclosure.

[0090] It should be noted that the terms "first", "second", etc. in the specification, claims, and drawings of the present disclosure are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that the numbers used in this way are interchangeable where appropriate so that the embodiments of the present disclosure described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product, or apparatus that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products, or apparatus.

[0091] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present disclosure and not to limit it. Although the present disclosure has been described in detail with reference to the preferred embodiments, ordinary technicians in the relevant field should understand that the specific implementation methods of the present disclosure can still be modified or some technical features can be replaced by equivalents without departing from the spirit of the technical solutions of the present disclosure, which should all be included in the scope of the technical solutions requested for protection in the present disclosure.

Claims

1. A data management method, comprising: Obtaining visual features in environmental data; Storing the visual features as a first index in an ontology database; Storing non-visual features in the environmental data as a second index in the ontology database.

2. The data management method according to claim 1, wherein, The visual features include at least one of color features and shape features.

3. The data management method according to claim 1 or 2, wherein, The obtaining visual features in environmental data includes: Performing color extraction on the acquired image data through a color space model in an image sensor for collecting environmental data to obtain color features, where the visual features include the color features.

4. The data management method according to any one of claims 1 to 3, wherein, The obtaining visual features in environmental data includes: Performing shape extraction on an object in the environmental data through a predetermined contour recognition algorithm to obtain shape features.

5. The data management method according to claim 4, wherein, The visual features include the shape features.

6. The data management method according to any one of claims 1 to 5, wherein, The obtaining visual features in environmental data includes: Extracting supplementary visual features in the environmental data according to preset types of supplementary visual features.

7. The data management method according to claim 6, wherein, The visual features include the supplementary visual features.

8. The data management method according to any one of claims 1 to 7, wherein, The storing the visual features as a first index in an ontology database includes: Respectively storing each type of the visual features as the first index in the ontology database.

9. The data management method according to claim 8, wherein, The number of types of the visual features is greater than 1.

10. The data management method according to any one of claims 1 to 9, wherein, The respectively storing each type of the visual features as the first index in the ontology database includes: Sorting the types in ascending order of feature complexity coefficient according to the feature complexity coefficient of each type of the visual features; Generating the first index according to the sorting result, where the retrieval priority of the visual features of the types ranked higher is higher than that of the visual features of the types ranked lower.

11. The data management method according to claim 10, wherein, The storing the visual features as a first index in an ontology database further includes: Updating the visual features according to the number of each type of the visual features in the visual features, where the sorting the types in ascending order of feature complexity coefficient is performed for the updated visual features.

12. The data management method according to claim 11, wherein, The updating the visual features according to the number of each type of the visual features in the visual features includes: Updating the visual features by deleting the visual features of the types with the number of visual features being 1 in the visual features.

13. The data management method according to claim 12, wherein, If the number of each type of the visual features in the visual features is 1, there is no first index, and the second index forms a single-class element index sequence for storage.

14. The data management method according to claim 8 or 9, wherein, The feature complexity coefficient of a type is determined according to the number of different visual features of the corresponding type, and the higher the number, the higher the feature complexity coefficient.

15. The data management method according to any one of claims 1 to 14, wherein The retrieval priority of the second index is lower than that of the first index.

16. The data management method according to any one of claims 1 to 15, wherein, The method conforms to at least one of the following: The first index is the main index; The second index is the secondary index.

17. A retrieval method, comprising: Obtaining feature information of a retrieval object; Querying the feature information in an ontology database to obtain a retrieval result, where the ontology database is generated according to the data management method described in any one of claims 1 to 16.

18. The retrieval method according to claim 17, wherein, Querying the feature information in the ontology database to obtain a retrieval result includes: Querying the feature information according to the index with the highest priority in the ontology database to obtain a candidate range; Within the candidate range, in the order of decreasing priority of the indexes in the ontology database, query the feature information to narrow the candidate range until the retrieval result is obtained.

19. A data management device, comprising: A visual feature acquisition unit configured to acquire visual features in environmental data; A first index determination unit configured to store the visual features as a first index in an ontology database; A second index determination unit configured to store non-visual features in the environmental data as a second index in the ontology database.

20. A retrieval device, comprising: An information acquisition unit configured to acquire feature information of a retrieval object; A query unit configured to query the feature information in an ontology database to obtain a retrieval result, wherein the ontology database is generated according to the data management method described in any one of claims 1 to 16.

21. A data processing device, comprising: A memory; And A processor coupled to the memory, the processor being configured to execute the method described in any one of claims 1 to 18 based on instructions stored in the memory.

22. A computer-readable storage medium having computer program instructions stored thereon, the instructions, when executed by a processor, implementing the steps of the method described in any one of claims 1 to 18.

23. A computer program for causing a processor to execute the method described in any one of claims 1 to 18.

24. A computer program product comprising a computer program or instructions, the computer program or instructions, when executed by a processor, implementing the method described in any one of claims 1 to 18.

Images