Cooperative data management method and related device

Abstract

The invention provides a collaborative data management method and a related device, the collaborative data management method is applied to a first host in a collaborative management system, and the collaborative management system comprises at least one terminal, the first host and an intelligent storage unit which are connected in sequence; the method comprises the following steps: in response to an access request from a first terminal, obtaining first album data and semantic tag metadata which correspond to each other; generating interface display information matched with a first terminal display rule according to the first album data, the semantic tag metadata and the interface display preference information; and sending the first photo album data, the semantic tag metadata and interface display information to the first terminal, wherein the interface display information is used for indicating the first terminal to display a photo album conforming to the display rule and a semantic tag corresponding to the photo album on a screen according to the first photo album data and the semantic tag metadata. In this way, the adaptive display interface information can be generated according to the display requirements of different terminals, and the display requirements of different devices can be flexibly met.

Description

Technical Field

[0001] This application belongs to the field of image data processing technology, specifically relating to a collaborative data management method and related apparatus. Background Technology

[0002] Currently, in the "MiniPC main computing power plus NAS storage" architecture, when a user accesses the site, the MiniPC main computing power retrieves the corresponding data from the NAS storage and then sends it directly to the terminal to generate a unified album interface. This makes the software unable to adapt to the display requirements of different terminals. Summary of the Invention

[0003] This application provides a collaborative data management method and related apparatus to adapt to the display needs of different terminals.

[0004] In a first aspect, this application provides a collaborative data management method applied to a first host in a collaborative management system, the collaborative management system comprising at least one terminal, a first host, and an intelligent storage unit connected in sequence; the method includes: In response to an access request from a first terminal, corresponding first album data and semantic tag metadata are obtained; the first terminal is any one of the at least one terminal; the first album data is obtained from the smart storage unit according to the shared directory, and the semantic tag metadata is obtained from the first host; the access request includes the interface display preference information of the first terminal; Based on the first album data, the semantic tag metadata, and the interface display preference information, interface display information adapted to the first terminal display rules is generated; The first album data, the semantic tag metadata, and the interface display information are sent to the first terminal. The interface display information is used to instruct the first terminal to display albums that conform to the display rules and semantic tags corresponding to the albums on the screen according to the first album data and the semantic tag metadata.

[0005] In conjunction with the first aspect, in one possible embodiment, before responding to the access request from the first terminal, the method further includes: storing the original album data from any of the at least one terminal and generating the shared directory; associating the shared directory with the album data access directory in the smart storage unit, the album data access directory including the storage address of the original album data in the smart storage unit, the original album data including a first image; determining the available computing resources of the first host and the smart storage unit; determining a target computing party based on the available computing resources, the target computing party including the first host and the smart storage unit; and having the target computing party perform image processing on the first image to obtain the first album data and the semantic tag metadata.

[0006] In conjunction with the first aspect, in one possible embodiment, storing the original album data from any of the at least one terminal and generating the shared directory includes: when original album data uploaded by at least one terminal is obtained, obtaining the idle state of the smart storage unit; when the idle state indicates that the free space of the smart storage unit meets a preset condition, outputting the original album data to the smart storage unit for storage; and obtaining the updated shared directory fed back by the smart storage unit.

[0007] In conjunction with the first aspect, in one possible embodiment, determining the target computing party based on the available computing resources includes: determining the available computing resource level of the first host; when the available computing resource level is a first level, determining the first host as the target computing party; when the available computing resource level is a second level, determining the first host and the intelligent storage unit as the target computing party; and when the available computing resource level is a third level, determining the intelligent storage unit as the target computing party.

[0008] In conjunction with the first aspect, in one possible embodiment, the target computing party performs image processing on the first image to obtain the first album data and the semantic tag metadata, including: when the target computing party is the first host, the first host performs image processing on the first image to obtain the first album data and the semantic tag metadata; when the target computing party is the first host and the smart storage unit, the first host and the smart storage unit jointly perform image processing on the first image to obtain the first album data and the semantic tag metadata; when the target computing party is the smart storage unit, a lightweight task is sent to the smart storage unit, and the processing result from the smart storage unit for the lightweight task is obtained.

[0009] In conjunction with the first aspect, in one possible embodiment, the process of performing image processing on the first image to obtain the first album data and the semantic tag metadata includes: preprocessing the first image to obtain the first album data; performing image recognition on the first image to determine the image category of the first image; and associating semantic tags with the corresponding first image based on the image category to obtain semantic tag metadata.

[0010] In conjunction with the first aspect, in one possible embodiment, the first album data includes a second image; determining the image category of the first image based on image recognition of the first image includes: inputting the second image into an AI image classification model, the AI ​​image classification model including multiple sub-models; extracting features from the second image by the multiple sub-models to obtain a target feature vector; and determining all image categories contained in the second image by the multiple sub-models based on the target feature vector.

[0011] Secondly, this application provides a collaborative data management device applied to a first host in a collaborative management system, the collaborative management system comprising at least one terminal, a first host, and an intelligent storage unit connected in sequence; the collaborative data management device includes: The acquisition unit is configured to respond to an access request from a first terminal and acquire corresponding first album data and semantic tag metadata; the first terminal is any one of the at least one terminal; the first album data is acquired from the smart storage unit according to the shared directory, and the semantic tag metadata is acquired from the first host; the access request includes the interface display preference information of the first terminal; The generation unit is used to generate interface display information that adapts to the display rules of the first terminal based on the first album data, the semantic tag metadata and the interface display preference information; The sending unit is used to send the first album data, the semantic tag metadata, and the interface display information to the first terminal. The interface display information is used to instruct the first terminal to display albums that conform to the display rules and semantic tags corresponding to the albums on the screen according to the first album data and the semantic tag metadata.

[0012] Thirdly, this application provides an electronic device including a processor, a memory, a communication interface, and one or more programs, said one or more programs being stored in the memory and configured to be executed by the processor, said programs including instructions for performing the steps of either the first or second aspect of this application.

[0013] Fourthly, this application provides a computer-readable storage medium storing a computer program for electronic data interchange, wherein the computer program causes a computer to perform some or all of the steps described in either the first or second aspect of this application.

[0014] Fifthly, this application provides a computer program product, wherein the computer program product includes a non-transitory computer-readable storage medium storing a computer program operable to cause a computer to perform some or all of the steps described in either the first or second aspect of this application. The computer program product may be a software installation package.

[0015] As can be seen, in this application, the system first responds to an access request from a first terminal to obtain corresponding first album data and semantic tag metadata; then, based on the first album data, the semantic tag metadata, and the interface display preference information, it generates interface display information adapted to the display rules of the first terminal; finally, it sends the first album data, the semantic tag metadata, and the interface display information to the first terminal, wherein the interface display information instructs the first terminal to display albums conforming to the display rules and corresponding semantic tags on the screen based on the first album data and the semantic tag metadata. In this way, it is possible to generate adapted display interface information according to the display needs of different terminals, flexibly adapting to the display requirements of different devices. Attached Figure Description

[0016] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1 This is a schematic diagram of the architecture of the collaborative management system provided in the embodiments of this application; Figure 2 This is a schematic diagram of the structure of the first host provided in an embodiment of this application; Figure 3 This is a schematic diagram of the structure of the intelligent storage unit provided in the embodiments of this application; Figure 4 This is a flowchart illustrating the first collaborative data management method provided in the embodiments of this application; Figure 5 This is a schematic diagram of the initial album interface provided in the embodiments of this application; Figure 6 This is the second collaborative data management method provided in the embodiments of this application; Figure 7 This is a schematic diagram of the structure of the collaborative data management device provided in the embodiments of this application; Figure 8 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0018] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present application.

[0019] The terms "first," "second," etc., in the specification, claims, and accompanying drawings of this application are used to distinguish different objects, not to describe a specific order. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or apparatus that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to these processes, methods, systems, products, or apparatuses.

[0020] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0021] The following is a brief introduction to the relevant terminology used in this application.

[0022] MiniPC (Mini Personal Computer, often simply referred to as "mini PC") is a small, low-power, highly integrated general-purpose computing device. It retains the core computing components of a traditional desktop computer (CPU / GPU / NPU, memory, basic interfaces) while eliminating the redundant design of traditional chassis. Its size is only 1 / 10 to 1 / 5 of a traditional desktop computer, and its power consumption is typically between 10-50W. It can achieve the same operating system operation, software deployment, and computing power as a traditional PC. It combines portability, low power consumption, and computing power flexibility, making it the mainstream computing device for edge computing, home office work, and embedded computing scenarios.

[0023] NAS (Network Attached Storage) is a distributed network storage device based on computer networks and accessible to multiple devices. It uses hard disk arrays / storage pools as the core storage carrier, runs a dedicated storage operating system, and supports standardized network storage protocols such as SMB / NFS / iSCSI. It can realize cross-device and cross-system file sharing, data backup, incremental synchronization, and remote access. Its core function is networked unified data storage and management. Unlike local storage devices such as USB flash drives and external hard drives, its core value is to break down data silos on a single device and provide centralized storage services for multiple terminals. Power consumption is usually between 5-20W, making it suitable for the light storage needs of home users and small and medium-sized enterprises.

[0024] Softmax (normalized exponential function) is one of the core activation functions in deep learning. It is specifically used for the output layer of multi-class classification tasks. It can transform any real-valued K-dimensional input vector into a probability distribution vector with a sum of 1 in the interval (0,1), where each element represents the predicted probability of the corresponding class.

[0025] Currently, in the "MiniPC main computing power plus NAS storage" architecture, when a user accesses the site, the MiniPC main computing power retrieves the corresponding data from the NAS storage and then sends it directly to the terminal to generate a unified album interface. This makes the software unable to adapt to the display requirements of different terminals.

[0026] To address the aforementioned issues, this application provides a collaborative data management method and related apparatus. This collaborative data management method can be applied to scenarios involving collaborative data management between main computing nodes and intelligent storage nodes. Responding to an access request from a first terminal, it acquires corresponding first album data and semantic tag metadata; based on the first album data, the semantic tag metadata, and the interface display preference information, it generates interface display information adapted to the display rules of the first terminal; it sends the first album data, the semantic tag metadata, and the interface display information to the first terminal, whereby the interface display information instructs the first terminal to display albums conforming to the display rules and their corresponding semantic tags on the screen according to the first album data and the semantic tag metadata. This allows for the generation of adapted display interface information based on the display needs of different terminals, flexibly adapting to the display requirements of different devices. This solution is applicable to various scenarios, including but not limited to the application scenarios mentioned above.

[0027] The system architecture involved in the embodiments of this application is described below.

[0028] Please see Figures 1 to 3 This application provides a collaborative management system, which includes at least one terminal (e.g., ...) connected in sequence. Figure 1 The first terminal, the second terminal... the nth terminal shown can be understood as follows: Figure 1 For illustrative purposes only, at least one terminal may include one terminal, two terminals, or more terminals (unrestricted here), a first host, and an intelligent storage unit. The first host serves as the main computing node, performing the primary computing tasks; the intelligent storage unit serves as a storage node, storing data and sharing computing tasks with the first host when needed. Figure 1 The dashed lines in the diagram represent wireless connections, and the solid lines represent wired connections. This can be understood as... Figure 1 For example only, the intelligent storage unit, the first host and at least one terminal can be connected by wired or wireless means or other means, and are not limited to one another.

[0029] Optionally, the first host can be a MiniPC or other small / micro host and small / micro computing power device, and the intelligent storage unit can be a NAS or other storage device configured with CPU / GPU / video processing chip / image processing chip, etc., which can realize computing functions; the terminal can be any electronic device, such as mobile phone, tablet, etc., or even the MiniPC itself, without being limited to a single device.

[0030] Specifically, by constructing a collaborative hardware system of terminals, a first host, and intelligent storage units, the first host can interact with the terminals and intelligent storage units respectively, thereby realizing collaborative data management between the first host and intelligent storage units.

[0031] Optional, please refer to Figure 3 The intelligent storage unit can be equipped with a CPU (such as an Intel N100 quad-core processor, or other processor models), a small memory module (such as 8GB or other specifications), and a storage sub-unit (such as a hard drive, responsible for unified storage of photo album data, etc.). Please refer to [link / reference]. Figure 2 The primary host, serving as the main computing core, is equipped with a multi-core processor and a large memory module (e.g., 16GB or other specifications). It is directly connected to the intelligent storage unit via a local area network, supporting large-scale AI model recognition and batch task processing. The terminal, as the user's direct operating device, supports a high-speed upload interface for album data. Users can interact with the primary host through the terminal to perform operations such as adding, deleting, modifying, and querying album data. They can also adjust software interface settings or other interactive functions; no single limitation is made here.

[0032] Furthermore, a dedicated data channel (e.g., a 2.5GbE LAN) is established between the primary host and the intelligent storage unit via a local area network (LAN), supporting photo data transfer speeds of 200MB / s or higher, with latency controlled to less than or equal to 2ms. In the connection link between the terminal and the primary host, the terminal accesses the software interface, which includes data stored in the intelligent storage unit and semantic tag metadata of the primary host, only through the LAN / WAN. The entire data processing is completed within the storage computing power cluster built by the primary host and the intelligent storage unit.

[0033] The specific methods will be described in detail below.

[0034] Please see Figures 4 to 6 This application also provides a collaborative data management method applied to a first host in a collaborative management system, the collaborative management system including at least one terminal, a first host, and an intelligent storage unit; the method includes: Step S201: Respond to the access request from the first terminal and obtain the corresponding first album data and semantic tag metadata.

[0035] Wherein, the first terminal is any one of the at least one terminals; the first album data is obtained from the smart storage unit according to the shared directory, and the semantic tag metadata is obtained from the first host; the access request includes the interface display preference information of the first terminal.

[0036] In specific implementation, such as Figure 5 As shown, the terminal can access the software interface and send an access request to the first host through the software interface. After the first host receives the access request from the terminal, it responds to the access request by retrieving the first album data from the intelligent storage unit, and then retrieves the corresponding voice tag metadata from its own storage space as the basic data for the initial album interface 100 in the software interface.

[0037] Optionally, the initial album interface 100 includes icons for multiple sub-albums (such as...). Figure 5 The album is divided into six sub-albums, each with corresponding category tags (e.g., first album, second album, third album, fourth album, fifth album, and sixth album). Figure 5 The first, second, third, fourth, fifth, and sixth tags in the image (the category tags can be text, images, or other identifiers, and are not required to be unique) are used to navigate to the corresponding sub-album. Users can click on the corresponding sub-album icon or category tag to access the sub-album and view the images within that sub-album. This is understandable. Figure 5 For illustrative purposes only, the number of sub-albums in the initial album interface 100 can be other than the number specified here.

[0038] Step S202: Generate interface display information adapted to the first terminal display rules based on the first album data, the semantic tag metadata, and the interface display preference information.

[0039] In practice, after obtaining the first album data and semantic tag metadata, the first host, combined with the interface display preference information carried in the access request, generates interface display information consistent with the display rules of the first terminal. These display rules include, but are not limited to, interface styles and classification patterns. Interface style refers to the style of the initial album interface 100, including but not limited to one or more of the following: interface background, icon style, category tag style, and image thumbnail style. Classification pattern refers to how the images in the first album data are categorized; classification patterns include, but are not limited to, categorization by function, by type, and by shape, and are not required to be unique.

[0040] Step S203: Send the first album data, the semantic tag metadata, and the interface display information to the first terminal.

[0041] The interface display information is used to instruct the first terminal to display albums and corresponding semantic tags on the screen according to the first album data and the semantic tag metadata, which conform to the display rules.

[0042] In the specific implementation, after obtaining the first album data, semantic tag metadata and interface display information, the initial album interface 100 is generated according to the display rules defined by the interface display information, so that users can view or operate it.

[0043] As can be seen, in this embodiment, by responding to an access request from the first terminal, corresponding first album data and semantic tag metadata are obtained; interface display information adapted to the display rules of the first terminal is generated based on the first album data, the semantic tag metadata, and the interface display preference information; the first album data, the semantic tag metadata, and the interface display information are sent to the first terminal, wherein the interface display information is used to instruct the first terminal to display albums conforming to the display rules and the corresponding semantic tags on the screen according to the first album data and the semantic tag metadata. In this way, adapted display interface information can be generated according to the display needs of different terminals, flexibly adapting to the display needs of different devices.

[0044] In one possible embodiment, please refer to the following for details. Figure 6Before responding to the access request from the first terminal, the method further includes: storing the original album data from any of the at least one terminal and generating the shared directory; associating the shared directory with the album data access directory in the smart storage unit, the album data access directory including the storage address of the original album data in the smart storage unit, the original album data including a first image; determining the available computing resources of the first host and the smart storage unit; determining a target computing party based on the available computing resources, the target computing party including the first host and the smart storage unit; and having the target computing party perform image processing on the first image to obtain the first album data and the semantic tag metadata.

[0045] Specifically, in the collaborative management system, once the connection link between the first host, the terminal, and the intelligent storage unit is established, the terminal can upload the original album data (i.e., ...) to the first host. Figure 6 (See "Upload Files from Terminal" in the document).

[0046] Once the first host obtains the raw album data uploaded by the first terminal, it begins to process as follows: Figure 6 The "intelligent storage unit" shown above. In its specific implementation, the original album data is first sent directly to the intelligent storage unit for storage, generating a corresponding shared directory. This shared directory is associated with the album data access directory in the intelligent storage unit. The intelligent storage unit can directly query the specific location (i.e., storage address) where the album data (including the original album data and the first album data) is stored through the album data access directory. Meanwhile, the first host can request the corresponding original album data from the intelligent storage unit through the shared directory.

[0047] In one possible embodiment, storing the original album data from any of the at least one terminal and generating the shared directory includes: when original album data uploaded by at least one terminal is obtained, obtaining the idle state of the smart storage unit; when the idle state indicates that the free space of the smart storage unit meets a preset condition, outputting the original album data to the smart storage unit for storage; and obtaining the updated shared directory fed back by the smart storage unit.

[0048] In specific implementation, after the first host obtains the original album data from at least one terminal, it sends a status acquisition command to the intelligent storage unit. After receiving the status acquisition command, the intelligent storage unit sends its own idle status to the first host. The idle status includes the current size of the free space of the intelligent storage unit. When the size of the free space meets the preset conditions, the original album data is sent to the intelligent storage unit for unified storage.

[0049] Optionally, the preset condition may be: the available space is greater than the original album data uploaded this time, or the available space is at least greater than twice the original album data uploaded this time. When it is determined that the smart storage unit meets the preset condition, the original album data is sent to the smart storage unit.

[0050] After receiving the original album data, the intelligent storage unit first stores all the original album data. Then, it performs incremental duplicate data verification on the original album data to delete data in the original album data that is identical to the album data already stored in the intelligent storage unit, retaining only the new data (i.e., original album data that is not duplicated). Specifically, incremental duplicate data verification includes: directly reading the globally unique hash value of the file generated when storing each picture in the original album data, comparing it with the hash values ​​of all files in the hash index library. If the hash values ​​are completely identical, it is determined to be duplicate data; otherwise, it is new data.

[0051] If the data is duplicated, it is deleted, only the metadata link (pointing to the stored original album data) is retained, and the process is terminated. If the data is not duplicated, it is marked as "valid incremental data". After the incremental duplicate data verification, the album data access directory is updated, and the shared directory is updated based on the updated album data access directory. Then, the updated directory is synchronized to the first host.

[0052] Optionally, the intelligent storage unit can also receive each image in real time during the process of receiving raw album data. After all blocks of a single image are transmitted, the original image is immediately and completely stored in the album's raw data pool (e.g., / volume1 / photo / raw), generating a unique file ID and storage path. Simultaneously, incremental duplicate data verification is triggered, the global hash value of the image is calculated, and the hash index table of the intelligent storage unit is entered. At the same time, basic integrity verification is performed (checking whether the data is damaged due to transmission). If the verification fails, the terminal is notified to re-upload; if the verification succeeds, the local backup is completed.

[0053] Specifically, the aforementioned incremental hash verification is as follows: Figure 6 "Hash verification" in the context.

[0054] As can be seen, this embodiment can deduplicate incremental data, improving data cleanliness and saving storage space. Simultaneously, by replacing file copying with shared directory mapping, the latency of album data interaction between the first host and the intelligent storage unit is reduced, and the consistency between tag metadata and NAS stored data reaches 100%.

[0055] Furthermore, after the original album data is stored, it needs to be processed to transform it into data that conforms to the storage rules of the intelligent storage unit.

[0056] First, determine the available computing resources of the first host, and then execute different data processing strategies based on the level of available computing resources.

[0057] In one example, determining the target computing party based on the available computing resources includes: determining the available computing resource level of the first host; when the available computing resource level is a first level, determining the first host as the target computing party; when the available computing resource level is a second level, determining the first host and the intelligent storage unit as the target computing party; and when the available computing resource level is a third level, determining the intelligent storage unit as the target computing party.

[0058] In the specific implementation, the first host is pre-set with different levels based on the available computing resources. For example, three levels are preset: when the available computing resources are in the range of 60%-100%, it is determined to be level one, indicating that the first host has sufficient available computing resources and can act as a target computing source independently; when the available computing resources are in the range of 20%-60%, it is determined to be level two, indicating that the first host has medium available computing resources and can act as a target computing source together with the intelligent storage unit; when the available computing resources are in the range of 0%-20%, it is determined to be level three, indicating that the first host has insufficient available computing resources and the intelligent storage unit can be used as the target computing source. It is understood that in practical applications, other levels can be set, such as increasing the number of levels for more granular management, or decreasing the number of levels; this is not a limitation on uniqueness.

[0059] In one possible embodiment, the target computing party performs image processing on the first image to obtain the first album data and the semantic tag metadata, including: when the target computing party is the first host, the first host performs image processing on the first image to obtain the first album data and the semantic tag metadata; when the target computing party is the first host and the smart storage unit, the first host and the smart storage unit jointly perform image processing on the first image to obtain the first album data and the semantic tag metadata; when the target computing party is the smart storage unit, a lightweight task is sent to the smart storage unit, and the processing result from the smart storage unit for the lightweight task is obtained.

[0060] For example, for album data processing, including but not limited to the following lightweight tasks: incremental data lightweight update, data preprocessing, feature extraction, image classification, and post-processing.

[0061] Specifically, incremental data lightweight update refers to performing a second incremental hash check on the original album data stored in the smart storage unit (the initial incremental hash check was already performed during storage).

[0062] Data preprocessing refers to cleaning the original photo album data (removing duplicates, low-quality, junk, and invalid files) and normalizing the format. Data cleaning requires incremental hash verification to confirm the existence of duplicate data, filtering low-quality images (such as images that are too small or too blurry), filtering junk images (screenshots, cached images, black screens, green screens, images or videos that are too short), and filtering non-image / non-video files to ensure that the purity of the data entering the AI ​​image classification model meets the preset requirements.

[0063] Feature extraction refers to extracting the corresponding target feature vector from a preprocessed second image.

[0064] Image classification refers to the automatic classification of images in the first album of data using an AI image classification model based on target feature vectors.

[0065] Post-processing is the final stage in the entire photo album data processing workflow. It does not perform computationally intensive AI inference / feature extraction itself, but instead performs structured aggregation, hierarchical storage, efficient index construction, and incremental maintenance on all the scattered data (hash values, metadata, feature vectors, coarse classification results, etc.) generated in the first four steps. The core goal is to avoid "repeatedly parsing the original files and repeatedly calculating features" during subsequent fine classification / retrieval, thus significantly reducing computational consumption. It reduces the time taken for photo album retrieval (such as searching for "coconuts by the sea" or "faces") from seconds of "traversing the entire file" to milliseconds of "index matching", ensuring consistency of results across multiple terminals such as NAS / MiniPC / mobile phones.

[0066] In practical implementation, when available computing resources are at the first level, the target computing entity is determined as the first host. At this point, the first host has sufficient computing resources and can execute most of the lightweight tasks. This allows the high computing performance of the first host to quickly obtain the processing results, resulting in the first album data and semantic tag data. For example, the first host sends the highest priority lightweight task to the intelligent storage unit; the intelligent storage unit only performs high-priority tasks on the images: only "extracting structural metadata" and "incremental hashing"; then the intelligent storage unit feeds back the processing results to the first host, which then continues to execute subsequent tasks such as "redundant filtering," "format normalization," "feature extraction," and "image classification." The MiniPC autonomously completes data preprocessing (i.e., using idle computing power) using these tasks. Figure 6 Data preprocessing), coarse feature extraction (i.e.) Figure 6In addition to feature extraction and image classification, the intelligent storage unit only performs storage and indexing, minimizing its own power consumption.

[0067] When available computing resources are at the second level, the target computing entities are identified as the first host and the intelligent storage unit. This means the first host has relatively sufficient computing resources, but to reduce its computational load, it can collaborate with the intelligent storage unit to process lightweight tasks. For example, the first host issues high-priority lightweight tasks to the intelligent storage unit, which processes these tasks and sends the results back to the first host. The first host then executes subsequent lightweight tasks based on the results, ultimately obtaining the first album data and semantic tag data. For instance, the intelligent storage unit performs high / medium-priority lightweight tasks on each image, such as data preprocessing (without feature extraction), marking the image as "to be extracted," and storing it in the intelligent storage unit's preprocessing data pool. Simultaneously, the processing results are fed back to the first host, which then continues with subsequent feature extraction, image classification, and post-processing tasks. The intelligent storage unit only handles high / medium priority lightweight tasks for AI photo album classification. Low priority tasks (such as image classification and post-processing) are processed in a delayed manner in the task queue (or executed when the MiniPC's computing power is temporarily idle), balancing the reduction of the MiniPC's workload with the rational utilization of NAS computing power / power consumption, and avoiding meaningless full-load operation. It is understandable that after the MiniPC completes image classification, the intelligent storage unit can also perform post-processing; this is not a limitation.

[0068] When available computing resources are at level three, the target computing entity is determined to be the intelligent storage unit. This means the primary host's computing resources are insufficient. To reduce the computational load on the primary host, the intelligent storage unit can handle all lightweight tasks. For example, the intelligent storage unit performs a full lightweight task on an image. This means the intelligent storage unit processes lightweight tasks within its processing capabilities, such as incremental data updates, data preprocessing (e.g., redundancy cleaning, format normalization (e.g., RAW to JPG preview), metadata structure extraction), and feature extraction. Immediately after processing, the feature vectors / metadata are pushed to the MiniPC to prepare for the MiniPC's core fine classification (i.e., image classification). The MiniPC does not need to participate in the entire process; the primary host only needs to perform the final image classification and post-processing. It is understood that after the MiniPC completes image classification, the intelligent storage unit can also perform post-processing; this is not a strict limitation.

[0069] As can be seen, in this embodiment, adjusting the workload of the first host and the intelligent storage unit in handling lightweight tasks according to different levels of available computing resources not only improves the utilization rate of auxiliary computing power in the intelligent storage unit, but also improves the overall work efficiency.

[0070] In one possible embodiment, the process of performing image processing on the first image to obtain the first album data and the semantic tag metadata includes: preprocessing the first image to obtain the first album data; performing image recognition on the first image to determine the image category of the first image; and associating semantic tags with the corresponding first image based on the image category to obtain semantic tag metadata.

[0071] In the specific implementation, for the raw album data that has completed initial storage, an image classification processing flow is executed. First, preprocessing is performed on each first image to determine whether the first image is valid incremental data; if the first image is valid incremental data, a second image is obtained; the first album data is generated based on the second image. Preprocessing includes, but is not limited to, filtering for duplicate data, filtering for invalid data, and filtering for non-album data.

[0072] The duplicate data filtering process includes: reading the global hash value of the first image (generated during storage) and comparing it with the hash value in the feature index library of the intelligent storage unit; if the hash values ​​match, it is directly identified as duplicate data; if the hash values ​​do not match, the basic visual feature values ​​of the file (such as the edge features and color distribution of the image) are further extracted and compared with the feature values ​​of similar files in the hash index library. If the feature similarity is greater than a first preset value, it is determined to be similar duplicate data, marked as "similar data," and moved to the "similar data archive directory" of the intelligent storage unit, where the user can choose whether to retain or delete it. If the hash values ​​do not match and the feature similarity is less than the first preset value, it is determined to be purely valid data, and the subsequent invalid data filtering and non-album data filtering steps continue.

[0073] Invalid data filtering includes filtering invalid files according to preset rules, which can be filtering rules for low-quality and junk data. For example, for images: files with resolution < 320×240, file size < 100KB, blurriness > 80% (determined by edge detection algorithm), or files that are damaged and cannot be decoded; for videos: files with duration < 3 seconds, resolution < 480×360, bitrate < 500kbps, or black / green screen ratio > 90%.

[0074] Non-album data filtering: Identify non-image / video files (such as documents and installation packages) through file header features (Magic Number), and automatically move them into the "Miscellaneous Items Catalog" of the smart storage unit, excluding them from album processing.

[0075] After preprocessing, the data for the first album can be obtained.

[0076] As can be seen, in this embodiment, by performing preprocessing after storage deduplication, double deduplication is achieved, and invalid data is filtered out, thereby improving data quality.

[0077] Furthermore, image recognition is performed on the first image in the first album data to classify these images. Specifically, image recognition is performed on the first image to determine its category, including: inputting the second image into an AI image classification model, which includes multiple sub-models; extracting features from the second image using the multiple sub-models to obtain a target feature vector; and determining all image categories contained in the second image based on the target feature vector.

[0078] In the specific implementation, the first host is equipped with a pre-trained AI image classification model, which includes multiple sub-models. Each sub-model performs a type of image classification inference, such as a sub-model specifically for scene classification, a sub-model specifically for people classification, or other classification sub-models.

[0079] The technical solution in this embodiment will be introduced using two sub-models as examples.

[0080] The second image is input into the scene classification sub-model and the person classification sub-model, respectively. When the second image is input into the scene classification sub-model, the second image is input into the backbone network of the scene classification model. The feature extraction from shallow to deep layers is completed through lightweight convolutional layers, and finally the deep feature map (3D) is output. The principle is: through lightweight convolutional operations, irrelevant pixels are filtered out step by step to extract the scene's exclusive semantic features. See the notes for the core lightweight module analysis.

[0081] The hierarchical feature extraction process (from shallow to deep, with feature map size gradually decreasing, channel number gradually increasing, and features evolving from general to specific) includes shallow feature extraction, mid-level feature extraction, and deep feature extraction. Shallow feature extraction: using convolutional and pooling layers, it extracts edge, texture, and color features of the image (e.g., the "blue texture" of a beach, the "straight edges" of a living room), outputting a shallow feature map. Mid-level feature extraction: using bottleneck layers and depthwise separable convolutions, it fuses the shallow features into shape / region features (e.g., the "regional division of the sea surface and sand" of a beach, the "shape features of the sofa and table" of a living room), outputting a mid-level feature map. Deep feature extraction: using SE attention mechanisms and depthwise separable convolutions, it focuses on the core semantic regions of the scene (e.g., the "sea surface" of a beach, the "sofa" of a living room), suppressing irrelevant regions (e.g., pebbles on the beach, small ornaments in the living room), outputting a deep semantic feature map (smallest in size, most channels, and most specific features).

[0082] The deep feature map is input into the classification head network. Through three steps, the features are mapped to multiple scene category spaces, outputting a multi-dimensional original category probability vector. This essentially converts the deep semantic features into a "matching probability" for each scene category; the higher the probability, the more closely the image matches the scene. Specifically, Global Average Pooling (GAP) is performed on the deep feature map to globally average the features, preserving the global semantic features of the scene, eliminating positional differences, and compressing them into a one-dimensional feature vector. Then, a lightweight fully connected layer maps the one-dimensional feature vector into a multi-dimensional original score vector (corresponding to multiple common scenes, such as beach, living room, park, etc.). Next, the Softmax activation function converts the multi-dimensional original score vector into a probability vector between 0 and 1, ensuring the sum of the probabilities of all categories is 1, facilitating the determination of the best-matching scene, such as [beach: 0.92, park: 0.05, living room: 0.03, ...]. Finally, a multi-dimensional quantized probability vector is output (the target feature vector, with each dimension corresponding to the matching probability of a scene category).

[0083] Understandably, the principle of the person classification sub-model is the same as that of the scene classification sub-model, only the content of the analysis is different. It extracts deep feature maps and maps them to multiple person category spaces to output multi-dimensional original category probability vectors. In other words, it converts deep semantic features into "matching probabilities" for each scene category; the higher the probability, the more the image matches the scene. Specifically, global average pooling (GAP) is performed on the deep feature maps to globally average the deep feature maps, retaining the global semantic features of the scene, eliminating positional differences, and compressing them into a one-dimensional feature vector. Then, a lightweight fully connected layer is applied to map the one-dimensional feature vector into a multi-dimensional original score vector (corresponding to multiple common object categories, such as people, food, animals, plants, etc.). The Softmax activation function then converts the multi-dimensional original score vector into a probability vector between 0 and 1, ensuring that the sum of the probabilities of all categories is 1, facilitating the determination of the best-matching scene, such as [people: 0.91, food: 0.01, animals: 0.07, plants: 0.01, ...]. Finally, a multi-dimensional quantized probability vector is output (each dimension corresponds to the matching probability of a person category).

[0084] Finally, all categories output by the two sub-models are counted. For the scene category, if a valid scene exists, it is counted as one category; otherwise, it is not counted. For the person category, if a face is detected, it is counted as one category; otherwise, it is not counted (regardless of the number of faces, it is counted as one category). For the item category, the number of unique item categories is taken. For example, [coconut, sun hat, mobile phone] is counted as three categories, while [coconut, coconut, sun hat] is counted as two categories. The final count is: scene category + person category + item category = total number of categories. Example: A beach image contains 1 person, coconut, sun hat, and mobile phone, i.e., scene category 1 + person category 1 + item category 3, for a total of 5 element categories.

[0085] The above primarily describes the solutions of the embodiments of this application from the perspective of the method execution process. It is understood that, in order to achieve the above functions, mobile electronic devices include corresponding hardware structures and / or software modules for executing each function. Those skilled in the art should readily recognize that, in conjunction with the units and algorithm steps of the various examples described in the embodiments provided herein, this application can be implemented in hardware or a combination of hardware and computer software. Whether a function is executed by hardware or by computer software driving hardware depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0086] This application embodiment can divide the electronic device into functional units according to the above method example. For example, each function can be divided into a separate functional unit, or two or more functions can be integrated into one processing unit. The integrated unit can be implemented in hardware or as a software functional unit. It should be noted that the unit division in this application embodiment is illustrative and only represents one logical functional division. In actual implementation, there may be other division methods.

[0087] Please see Figure 7 This application also provides a collaborative data management device, applied to a first host in a collaborative management system, the collaborative management system including at least one terminal, a first host, and an intelligent storage unit connected in sequence; the collaborative data management device includes: The acquisition unit is configured to respond to an access request from a first terminal and acquire corresponding first album data and semantic tag metadata; the first terminal is any one of the at least one terminal; the first album data is acquired from the smart storage unit according to the shared directory, and the semantic tag metadata is acquired from the first host; the access request includes the interface display preference information of the first terminal; The generation unit is used to generate interface display information that adapts to the display rules of the first terminal based on the first album data, the semantic tag metadata and the interface display preference information; The sending unit is used to send the first album data, the semantic tag metadata, and the interface display information to the first terminal. The interface display information is used to instruct the first terminal to display albums that conform to the display rules and semantic tags corresponding to the albums on the screen according to the first album data and the semantic tag metadata.

[0088] As can be seen, in this embodiment, by responding to an access request from the first terminal, corresponding first album data and semantic tag metadata are obtained; interface display information adapted to the display rules of the first terminal is generated based on the first album data, the semantic tag metadata, and the interface display preference information; the first album data, the semantic tag metadata, and the interface display information are sent to the first terminal, wherein the interface display information is used to instruct the first terminal to display albums conforming to the display rules and the corresponding semantic tags on the screen according to the first album data and the semantic tag metadata. In this way, adapted display interface information can be generated according to the display needs of different terminals, flexibly adapting to the display needs of different devices.

[0089] In one possible embodiment, before responding to the access request from the first terminal, the method further includes: storing the original album data from any of the at least one terminal and generating the shared directory; associating the shared directory with the album data access directory in the smart storage unit, the album data access directory including the storage address of the original album data in the smart storage unit, the original album data including a first image; determining the available computing resources of the first host and the smart storage unit; determining a target computing party based on the available computing resources, the target computing party including the first host and the smart storage unit; and having the target computing party perform image processing on the first image to obtain the first album data and the semantic tag metadata.

[0090] In one possible embodiment, storing the original album data from any of the at least one terminal and generating the shared directory includes: when original album data uploaded by at least one terminal is obtained, obtaining the idle state of the smart storage unit; when the idle state indicates that the free space of the smart storage unit meets a preset condition, outputting the original album data to the smart storage unit for storage; and obtaining the updated shared directory fed back by the smart storage unit.

[0091] In one possible embodiment, determining the target computing party based on the available computing resources includes: determining the available computing resource level of the first host; when the available computing resource level is a first level, determining the first host as the target computing party; when the available computing resource level is a second level, determining the first host and the intelligent storage unit as the target computing party; and when the available computing resource level is a third level, determining the intelligent storage unit as the target computing party.

[0092] In one possible embodiment, the target computing party performs image processing on the first image to obtain the first album data and the semantic tag metadata, including: when the target computing party is the first host, the first host performs image processing on the first image to obtain the first album data and the semantic tag metadata; when the target computing party is the first host and the smart storage unit, the first host and the smart storage unit jointly perform image processing on the first image to obtain the first album data and the semantic tag metadata; when the target computing party is the smart storage unit, a lightweight task is sent to the smart storage unit, and the processing result from the smart storage unit for the lightweight task is obtained.

[0093] In one possible embodiment, the process of performing image processing on the first image to obtain the first album data and the semantic tag metadata includes: preprocessing the first image to obtain the first album data; performing image recognition on the first image to determine the image category of the first image; and associating semantic tags with the corresponding first image based on the image category to obtain semantic tag metadata.

[0094] In one possible embodiment, the first album data includes a second image; determining the image category of the first image based on image recognition of the first image includes: inputting the second image into an AI image classification model, the AI ​​image classification model including multiple sub-models; extracting features from the second image by the multiple sub-models to obtain a target feature vector; and determining all image categories contained in the second image by the multiple sub-models based on the target feature vector.

[0095] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, as a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired or wireless means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more sets of available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium. A semiconductor medium can be a solid-state drive.

[0096] This application also provides an electronic device 80, such as... Figure 8As shown, it includes at least one processor 81; a display screen 82; and a memory 83, and may also include a communications interface 85 and a bus 84. The processor 81, display screen 82, memory 83, and communications interface 85 can communicate with each other via the bus 84. The display screen 82 is configured to display a preset user guide interface in the initial setup mode. The communications interface 85 can transmit information. The processor 81 can call logical instructions in the memory 83 to execute the methods described in the above embodiments.

[0097] Optionally, the electronic device 80 may be a mobile electronic device, an electronic device, or other devices, and is not limited to any particular type.

[0098] Furthermore, the logic instructions in the aforementioned memory 83 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium.

[0099] The memory 83, as a computer-readable storage medium, can be configured to store software programs, computer-executable programs, such as program instructions or modules corresponding to the methods in the embodiments of this disclosure. The processor 81 executes functional applications and data processing by running the software programs, instructions, or modules stored in the memory 83, thereby implementing the methods in the above embodiments.

[0100] The memory 83 may include a program storage area and a data storage area. The program storage area may store the operating system and application programs required for at least one function; the data storage area may store data created based on the use of the electronic device 80. Furthermore, the memory 83 may include high-speed random access memory (RAM) and may also include non-volatile memory. For example, various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks, may be used, or they may be transient storage media.

[0101] This application also provides a computer storage medium storing a computer program for electronic data interchange, which causes a computer to perform some or all of the steps of any of the methods described in the above method embodiments, wherein the computer includes an electronic device.

[0102] This application also provides a computer program product, which includes a non-transitory computer-readable storage medium storing a computer program operable to cause a computer to perform some or all of the steps of any of the methods described in the above method embodiments. The computer program product may be a software installation package, and the computer may include an electronic device.

[0103] It should be understood that, in the various embodiments of this application, the order of the above-mentioned processes does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0104] In the several embodiments provided in this application, it should be understood that the disclosed methods, apparatuses, and systems can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for example, the division of units is merely a logical functional division, and other division methods may exist in actual implementation; 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 interfaces, and the indirect coupling or communication connection between devices or units may be electrical, mechanical, or other forms.

[0105] 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.

[0106] Furthermore, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can be physically comprised separately, or two or more units can be integrated into one unit. The integrated unit described above can be implemented in hardware or in the form of hardware plus software functional units.

[0107] The integrated units implemented as software functional units described above can be stored in a computer-readable storage medium. These software functional units, stored in a storage medium, include several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute some steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes: a USB flash drive, a portable hard drive, a magnetic disk, an optical disk, volatile memory, or non-volatile memory. The non-volatile memory can be read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), or flash memory. The volatile memory can be random access memory (RAM), which is used as an external cache. By way of example, but not limitation, many forms of random access memory (RAM) are available, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), synchronous linked DRAM (SLDRAM), and direct rambus RAM (DR RAM), etc., which are various media capable of storing program code.

[0108] While the present invention has been disclosed above, it is not limited thereto. Any person skilled in the art can easily conceive of variations or substitutions without departing from the spirit and scope of the present invention, and various modifications and alterations can be made, including combinations of the different functions and implementation steps described above, as well as software and hardware implementation methods, all of which are within the protection scope of the present invention.

Claims

1. A collaborative data management method, characterized in that, A first host is applied in a collaborative management system, the collaborative management system comprising at least one terminal and a first host and an intelligent storage unit connected in sequence; the method includes: In response to an access request from a first terminal, corresponding first album data and semantic tag metadata are obtained; the first terminal is any one of the at least one terminal; the first album data is obtained from the smart storage unit according to the shared directory, and the semantic tag metadata is obtained from the first host; the access request includes the interface display preference information of the first terminal; Based on the first album data, the semantic tag metadata, and the interface display preference information, interface display information adapted to the first terminal display rules is generated; The first album data, the semantic tag metadata, and the interface display information are sent to the first terminal. The interface display information is used to instruct the first terminal to display albums that conform to the display rules and semantic tags corresponding to the albums on the screen according to the first album data and the semantic tag metadata.

2. The method according to claim 1, characterized in that, Before responding to the access request from the first terminal, the method further includes: The original album data from any of the at least one terminal is stored, and the shared directory is generated; the shared directory is associated with the album data access directory in the smart storage unit, the album data access directory includes the storage address of the original album data in the smart storage unit, and the original album data includes a first picture; Determine the available computing resources of the first host and the intelligent storage unit; The target computing entity is determined based on the available computing resources, and the target computing entity includes the first host and the intelligent storage unit; The target computing party performs image processing on the first image to obtain the first album data and the semantic tag metadata.

3. The method according to claim 2, characterized in that, The step of storing the original photo album data from any of the at least one terminal and generating the shared directory includes: When at least one terminal uploads raw album data, the idle state of the smart storage unit is obtained; When the idle state indicates that the free space of the smart storage unit meets the preset conditions, the original album data is output to the smart storage unit for storage. Obtain the updated shared directory fed back by the intelligent storage unit.

4. The method according to claim 2, characterized in that, The step of determining the target computing provider based on the available computing resources includes: Determine the available computing resource level of the first host; When the available computing resources are at the first level, the first host is identified as the target computing provider; When the available computing resources are at the second level, the first host and the intelligent storage unit are identified as the target computing entity. When the available computing resource level is Level 3, the intelligent storage unit is identified as the target computing unit.

5. The method according to claim 2, characterized in that, The target computing party performs image processing on the first image to obtain the first album data and the semantic tag metadata, including: When the target computing party is the first host, the first host performs image processing on the first image to obtain the first album data and the semantic tag metadata; When the target computing entity is the first host and the intelligent storage unit, the first host and the intelligent storage unit jointly perform image processing on the first image to obtain the first album data and the semantic tag metadata; When the target computation is the intelligent storage unit, a lightweight task is sent to the intelligent storage unit, and the processing result of the lightweight task from the intelligent storage unit is obtained.

6. The method according to claim 5, characterized in that, The process of performing image processing on the first image to obtain the first album data and the semantic tag metadata includes: The first image is preprocessed to obtain the first album data; Perform image recognition on the first image to determine its image category; Based on the semantic tags associated with the first image corresponding to the image category, semantic tag metadata is obtained.

7. The method according to claim 6, characterized in that, The first album data includes a second image; based on image recognition of the first image, the image category of the first image is determined, including: The second image is input into an AI image classification model, which includes multiple sub-models; the multiple sub-models extract features from the second image to obtain a target feature vector; and the multiple sub-models determine all image categories contained in the second image based on the target feature vector.

8. A collaborative data management device, characterized in that, A first host is used in a collaborative management system, the collaborative management system comprising at least one terminal, a first host, and an intelligent storage unit connected in sequence; The collaborative data management device includes: The acquisition unit is configured to respond to an access request from a first terminal and acquire corresponding first album data and semantic tag metadata; the first terminal is any one of the at least one terminal; the first album data is acquired from the smart storage unit according to the shared directory, and the semantic tag metadata is acquired from the first host; the access request includes the interface display preference information of the first terminal; The generation unit is used to generate interface display information that adapts to the display rules of the first terminal based on the first album data, the semantic tag metadata and the interface display preference information; The sending unit is used to send the first album data, the semantic tag metadata, and the interface display information to the first terminal. The interface display information is used to instruct the first terminal to display albums that conform to the display rules and semantic tags corresponding to the albums on the screen according to the first album data and the semantic tag metadata.

9. An electronic device, characterized in that, The method includes a processor, a memory, a communication interface, and one or more programs, said one or more programs being stored in the memory and configured to be executed by the processor, said programs including instructions for performing the steps of the method as described in any one of claims 1-6.

10. A computer-readable storage medium, characterized in that, A computer program for electronic data interchange is stored, wherein the computer program causes a computer to execute instructions for the steps of the method as described in any one of claims 1-6.

Images