File naming method and system based on big data processing

Abstract

The invention provides a file naming method and system based on big data processing, and relates to the technical field of file naming, and the method comprises the steps: obtaining a plurality of video files to be named; processing the plurality of to-be-named video files based on a file analysis model to determine a representative frame of each to-be-named video file; based on the representative frame of each to-be-named video file, determining a plurality of alternative names of each to-be-named video file and content similarity between different representative frames by using a convolutional neural network model; determining a file target name of each to-be-named video file based on the plurality of alternative names of each to-be-named video file and the content similarity between different representative frames; and naming each video file based on the file target name of each to-be-named video file. According to the method, a large number of video files received at a time can be accurately named.

Description

Technical Field

[0001] The present invention relates to the technical field of file naming, and in particular to a file naming method and system based on big data processing. Background Art

[0002] With the popularity of the Internet and mobile devices, users are receiving and sharing video files on social software more and more frequently. Especially on social media, instant messaging tools and file sharing platforms, users often receive a large number of video files at one time. These files usually do not have clear names, or the naming methods are not uniform, which makes it very difficult for users to manage and find these files. Traditional file naming methods mainly rely on manual operations by users, which is not only time-consuming and labor-intensive, but also prone to errors. Users may need to open video files one by one, check the content and then manually name them, which is particularly inconvenient when dealing with a large number of files. In addition, manual naming may also lead to inconsistent naming, further increasing the difficulty of file management.

[0003] Therefore, how to accurately name a large number of video files received at one time is a problem that needs to be solved urgently. Summary of the invention

[0004] The main technical problem solved by the present invention is how to accurately name a large number of video files received at one time.

[0005] According to a first aspect, the present invention provides a file naming method based on big data processing, comprising: obtaining multiple video files to be named; processing the multiple video files to be named based on a file analysis model to determine a representative frame of each video file to be named; using a convolutional neural network model based on the representative frame of each video file to be named to determine multiple alternative names for each video file to be named and content similarities between different representative frames; determining a file target name for each video file to be named based on the multiple alternative names for each video file to be named and the content similarities between different representative frames; and naming each video file based on the file target name of each video file to be named.

[0006] In a possible implementation, determining the file target name of each video file to be named based on the multiple candidate names of each video file to be named and the content similarity between different representative frames includes: Constructing a file naming graph structure, wherein the file naming graph structure includes a plurality of nodes and a plurality of edges between the plurality of nodes, wherein each node represents each video file, and the node features of each node include a representative frame of each to-be-named video file and a plurality of alternative names of each to-be-named video file, and the features of the edges between the nodes represent the content similarity between different representative frames; The file naming graph structure is processed based on a graph convolutional network to determine a file target name for each video file to be named.

[0007] In a possible implementation, the file analysis model is a gated recurrent unit, the input of the file analysis model is the multiple video files to be named, and the output of the file analysis model is a representative frame of each video file to be named.

[0008] In a possible implementation, the input of the graph convolutional network is the file naming graph structure, and the output of the graph convolutional network is the file target name of each video file to be named.

[0009] According to a second aspect, the present invention provides a file naming system based on big data processing, comprising: A first acquisition module is used to acquire multiple video files to be named; A representative frame determination module, used for processing the plurality of video files to be named based on a file analysis model to determine a representative frame of each video file to be named; An alternative naming determination module, used to determine multiple alternative names for each video file to be named and content similarities between different representative frames using a convolutional neural network model based on the representative frames of each video file to be named; A target naming determination module, used to determine a file target naming of each video file to be named based on a plurality of candidate names of each video file to be named and content similarity between different representative frames; The naming module is used to name each video file based on the file target name of each video file to be named.

[0010] In a possible implementation, the target naming determination module is further used to: Constructing a file naming graph structure, wherein the file naming graph structure includes a plurality of nodes and a plurality of edges between the plurality of nodes, wherein each node represents each video file, and the node features of each node include a representative frame of each to-be-named video file and a plurality of alternative names of each to-be-named video file, and the features of the edges between the nodes represent the content similarity between different representative frames; The file naming graph structure is processed based on a graph convolutional network to determine a file target name for each video file to be named.

[0011] In a possible implementation, the file analysis model is a gated recurrent unit, the input of the file analysis model is the multiple video files to be named, and the output of the file analysis model is a representative frame of each video file to be named.

[0012] In a possible implementation, the input of the graph convolutional network is the file naming graph structure, and the output of the graph convolutional network is the file target name of each video file to be named.

[0013] According to a third aspect, an embodiment of the present invention provides an electronic device, comprising: a processor; a memory; and a computer program; wherein the computer program is stored in the memory and configured to be executed by the processor to implement the method as described above, the method comprising: obtaining a plurality of video files to be named; processing the plurality of video files to be named based on a file analysis model to determine a representative frame of each video file to be named; using a convolutional neural network model based on the representative frame of each video file to be named to determine a plurality of alternative names for each video file to be named and content similarities between different representative frames; determining a file target name for each video file to be named based on the plurality of alternative names for each video file to be named and content similarities between different representative frames; and naming each video file based on the file target name of each video file to be named.

[0014] According to the fourth aspect, the present embodiment provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the aforementioned file naming method based on big data processing, the method comprising: obtaining a plurality of video files to be named; processing the plurality of video files to be named based on a file analysis model to determine a representative frame of each video file to be named; using a convolutional neural network model based on the representative frame of each video file to be named to determine a plurality of alternative names for each video file to be named and content similarities between different representative frames; determining a file target name for each video file to be named based on the plurality of alternative names for each video file to be named and content similarities between different representative frames; and naming each video file based on the file target name of each video file to be named.

[0015] The present invention provides a file naming method and system based on big data processing, the method comprising acquiring a plurality of video files to be named; processing the plurality of video files to be named based on a file analysis model to determine a representative frame of each video file to be named; using a convolutional neural network model based on the representative frame of each video file to be named to determine a plurality of alternative names for each video file to be named and content similarities between different representative frames; determining a file target name for each video file to be named based on the plurality of alternative names for each video file to be named and content similarities between different representative frames; naming each video file based on the file target name of each video file to be named, the method can accurately name a large number of video files received at one time. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 A schematic diagram of an application scenario of a file naming method based on big data processing provided by an embodiment of the present invention; Figure 2 A flowchart of a file naming method based on big data processing provided by an embodiment of the present invention; Figure 3 A schematic diagram of a flow chart of determining a probability value of a fuzzy test based on multiple data input by a user at different time points provided by an embodiment of the present invention; Figure 4 A schematic diagram of a file naming system based on big data processing provided by an embodiment of the present invention; Figure 5 A schematic diagram of an electronic device provided by an embodiment of the present invention; Figure 6 A schematic diagram of the structure of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0017] The present invention is further described in detail below by specific embodiments in conjunction with the accompanying drawings. Wherein similar elements in different embodiments adopt associated similar element numbers. In the following embodiments, many detailed descriptions are for making the present invention better understood. However, those skilled in the art can easily recognize that some features can be omitted in different situations, or can be replaced by other elements, materials, methods. In some cases, some operations related to the present invention are not shown or described in the specification, this is to avoid the core part of the present invention being overwhelmed by too much description, and for those skilled in the art, it is not necessary to describe these related operations in detail, and they can fully understand the related operations according to the description in the specification and the general technical knowledge in the art.

[0018] Figure 1 A schematic diagram of an application scenario of a file naming method based on big data processing provided in an embodiment of the present invention. Figure 1 The application scenario of the file naming method based on big data processing may include a server 11, a network 12, a terminal 13 and a storage device 14.

[0019] In some embodiments, the server 11 may be a single server or a server group. The server 11 may access information and / or data stored in the terminal 13 or the storage device 14 through the network 12. In some embodiments, the server 11 may be used to execute Figure 2 The file naming method based on big data processing shown in .

[0020] The network 12 may facilitate the exchange of information and / or data. In some embodiments, the network 12 may be any form of wired or wireless network, or any combination thereof.

[0021] Terminal 13 may refer to one or more terminal devices used by a user. In some embodiments, terminal 13 may include one or more combinations of a mobile device, a tablet computer, a laptop computer, and the like.

[0022] The storage device 14 may store data and / or instructions. For example, the storage device 14 may store data instructions for a file naming method based on big data processing.

[0023] In an embodiment of the present invention, there is provided Figure 2 A file naming method based on big data processing is shown, and the file naming method based on big data processing includes steps S1 to S5: Step S1, obtaining a plurality of video files to be named; The multiple video files to be named are multiple video files to be named obtained from video files received by the user. For example, the user receives 10 video files through social software, and the system automatically obtains these files.

[0024] Step S2, processing the multiple video files to be named based on the file analysis model to determine a representative frame of each video file to be named; The representative frame is to process each video file using the file analysis model and extract the representative frame that can reflect the video content.

[0025] The file analysis model is a gated recurrent unit, the input of the file analysis model is the multiple video files to be named, and the output of the file analysis model is a representative frame of each video file to be named. The gated recurrent unit (GRU) is used to process sequence data and timing information. The gated recurrent unit includes three components: a memory unit, an update gate, and a reset gate.

[0026] Video files are composed of a series of continuous frames, each of which is an image at a point in time. There is a temporal dependency between these frames, that is, the information of the previous frame may be helpful for the recognition of the next frame. The order of frames in a video file is very important, and different frame orders may lead to completely different understandings of the content. For example, the start and end frames of an action have a clear temporal order. The gated recurrent unit can effectively capture the temporal dependency between video frames, store and transmit information through the memory unit, and ensure that the model can understand the dynamic changes of the video.

[0027] In some embodiments, the file analysis model includes an alternative frame extraction layer, an alternative frame association determination layer, an alternative frame analysis layer, and a representative frame determination layer. The alternative frame extraction layer, the alternative frame association determination layer, the alternative frame analysis layer, and the representative frame determination layer all include a gated loop unit structure, the input of the alternative frame extraction layer is the multiple video files to be named, the output of the alternative frame extraction layer is multiple alternative frames of each video file to be named, the input of the alternative frame association determination layer is multiple alternative frames of each video file to be named, the output of the alternative frame association determination layer is the association between each alternative frame in the video file to be named and other alternative frames, the input of the alternative frame analysis layer is multiple alternative frames of each video file to be named, the output of the alternative frame analysis layer is the character expression, background content richness, and color vividness of each alternative frame, the input of the representative frame determination layer is the character expression, background content richness, color vividness, and association between each alternative frame in the video file to be named and other alternative frames, and the output of the representative frame determination layer is the representative frame of each video file to be named.

[0028] Through the candidate frame extraction layer, the candidate frame relevance determination layer, the candidate frame analysis layer and the representative frame determination layer, the video files are processed step by step to ensure that the representative frame of each video file can best reflect the video content. This method not only improves the accuracy of representative frame selection, but also greatly reduces the amount of data for subsequent processing and improves the overall processing efficiency.

[0029] The character expression is used to evaluate the expression of the character in each candidate frame. For example, the more expressive the character is, the greater the expression is.

[0030] The background content richness is used to evaluate the richness of the background content in each candidate frame. For example, the greater the number and types of objects detected in the background, the greater the background content richness.

[0031] Color vividness is used to evaluate the vividness of colors.

[0032] Step S3, using a convolutional neural network model to determine multiple candidate names for each video file to be named and content similarities between different representative frames based on the representative frames of each video file to be named; The input of the convolutional neural network model is the representative frame of each video file to be named, and the output of the convolutional neural network model is multiple alternative names for each video file to be named and the content similarity between different representative frames.

[0033] The multiple candidate names of each video file to be named are multiple possible file names generated by the convolutional neural network model according to the representative frame content. For example, the multiple candidate names of the video file to be named include "outdoor activities", "friends gathering", "landscape shooting", etc.

[0034] The content similarity between different representative frames is to measure the content similarity between representative frames of different video files.

[0035] The representative frame is a key frame extracted from the video file that can better reflect the content of the video. The representative frame usually contains the most typical and representative scene in the video, which can summarize the main content of the video. The convolutional neural network extracts the features of the representative frame through the convolution layer and the pooling layer. The extracted features are mapped to multiple categories through the fully connected layer to generate multiple candidate names. Based on the classification results, multiple possible file names are generated. The convolutional neural network can extract the features of each representative frame through the convolution layer and the pooling layer. The content similarity between different representative frames is evaluated by calculating the similarity of the feature vectors of different representative frames (such as Euclidean distance, cosine similarity, etc.).

[0036] Step S4, determining a file target name for each video file to be named based on a plurality of candidate names for each video file to be named and content similarities between different representative frames; In some embodiments, Figure 3 A schematic diagram of a process for determining a file target name for each to-be-named video file provided by an embodiment of the present invention, wherein the process for determining a file target name for each to-be-named video file comprises steps S21-S22: Step S21, constructing a file naming graph structure, wherein the file naming graph structure includes multiple nodes and multiple edges between the multiple nodes, each node represents each video file, and the node features of each node include a representative frame of each video file to be named and multiple alternative names for each video file to be named, and the features of the edges between the nodes represent the content similarity between different representative frames.

[0037] The file naming graph structure is a graph data structure, in which each node represents a video file, the features of each node include a representative frame and multiple alternative names, and the features of the edges between nodes represent the content similarity between different representative frames.

[0038] Step S22: Process the file naming graph structure based on a graph convolutional network to determine a file target name for each video file to be named.

[0039] A graph convolutional network (GCN) is a deep learning model for processing graph structure data. The input of the graph convolutional network is the file naming graph structure, and the output of the graph convolutional network is the file target name of each video file to be named.

[0040] By building a file naming graph structure, we can capture the relationship between video files and ensure the consistency and logic of naming. Video files with high similarity usually have similar content, which ensures the difference of naming and avoids naming confusion. The features of each node include representative frames and multiple alternative names, which provide comprehensive information and help the model generate high-quality names. The features of the edges between nodes represent the content similarity and provide the association information between video files, which helps the model make more reasonable naming decisions.

[0041] The graph convolutional network captures the relationship between nodes and aggregates neighbor information to ensure that video files with high similarity are not given the same name to avoid naming duplication. Through the processing of the graph convolutional network, the features of each node are richer and more accurate, which can better reflect the content of different representative frames and highlight the uniqueness of different video files.

[0042] By constructing a file naming graph structure, we can capture the relationship between video files, provide rich contextual information, and ensure the consistency and logic of naming. By capturing the relationship between nodes and aggregating neighbor information, the graph convolutional network can efficiently and accurately determine the target name for each video file to be named. This method can not only avoid duplication of names, but also highlight the content of different representative frames, improving the accuracy and quality of naming. Through such a graph structure and graph convolutional network, a large number of video files can be effectively named, ensuring that each video file is named uniquely and accurately.

[0043] Step S5, naming each video file based on the file target naming of each video file to be named.

[0044] When the file target name of each video file to be named is determined, each video file is named.

[0045] Based on the same inventive concept, Figure 4 A schematic diagram of a file naming system based on big data processing provided by an embodiment of the present invention, the file naming system based on big data processing includes: A first acquisition module 41 is used to acquire a plurality of video files to be named; A representative frame determination module 42, configured to process the plurality of video files to be named based on a file analysis model to determine a representative frame of each video file to be named; An alternative naming determination module 43, used to determine multiple alternative names for each video file to be named and content similarities between different representative frames using a convolutional neural network model based on the representative frames of each video file to be named; A target naming determination module 44 is used to determine a file target naming of each video file to be named based on a plurality of candidate names of each video file to be named and content similarities between different representative frames; The naming module 45 is used to name each video file based on the file target name of each video file to be named.

[0046] Based on the same inventive concept, an embodiment of the present invention provides an electronic device, such as Figure 5 As shown, including: It includes: a processor 51; a memory 52; and a computer program; wherein the computer program is stored in the memory 52 and is configured to be executed by the processor 51 to implement the file naming method based on big data processing as provided above, the method including: obtaining multiple video files to be named; processing the multiple video files to be named based on a file analysis model to determine a representative frame of each video file to be named; using a convolutional neural network model based on the representative frame of each video file to be named to determine multiple alternative names for each video file to be named and content similarities between different representative frames; determining a file target name for each video file to be named based on the multiple alternative names for each video file to be named and content similarities between different representative frames; and naming each video file based on the file target name of each video file to be named.

[0047] Based on the same inventive concept, this embodiment provides a computer-readable storage medium having a computer program stored thereon, which, when executed by the processor 51, implements the aforementioned file naming method based on big data processing, the method comprising: obtaining a plurality of video files to be named; processing the plurality of video files to be named based on a file analysis model to determine a representative frame of each video file to be named; using a convolutional neural network model based on the representative frame of each video file to be named to determine a plurality of alternative names for each video file to be named and content similarities between different representative frames; determining a file target name for each video file to be named based on the plurality of alternative names for each video file to be named and content similarities between different representative frames; and naming each video file based on the file target name of each video file to be named.

[0048] The file naming method based on big data processing provided in the embodiment of the present application can be applied to terminal devices (such as mobile phones), tablet computers, laptops, ultra-mobile personal computers (ultra-mobile personal computers, UMPCs), handheld computers, netbooks, personal digital assistants (personal digital assistants, PDAs), wearable devices (such as smart watches, smart glasses or smart helmets, etc.), augmented reality (augmented reality, AR) \ virtual reality (virtual reality, VR) devices, smart home devices, car computers and other electronic devices, and the embodiment of the present application does not impose any restrictions on this.

[0049] Taking the mobile phone 100 as an example of the electronic device, Figure 6 A schematic structural diagram of the mobile phone 100 is shown.

[0050] like Figure 6 As shown, the mobile phone 100 may include a processing module 110, an external memory interface 120, an internal memory 121, a universal serial bus (USB) interface 130, a charging management module 140, a power management module 141, a battery 142, an antenna 1, an antenna 2, a mobile communication module 150, a wireless communication module 160, an audio module 170, a speaker 170A, a receiver 170B, a microphone 170C, an earphone interface 170D, a sensor module 180, a button 190, a motor 191, an indicator 192, a camera 193, a display screen 194, and a subscriber identification module (SIM) card interface 195, etc.

[0051] The processing module 110 may include one or more processing units, for example, the processing module 110 may include an application processor (AP), a modem processor, a graphics processor (GPU), an image signal processor (ISP), a controller, a memory, a video codec, a digital signal processor (DSP), a baseband processor, and / or a neural-network processing unit (NPU), etc. Different processing units may be independent devices or integrated into one or more processors.

[0052] The processing module 110 can be used to: obtain multiple video files to be named; process the multiple video files to be named based on a file analysis model to determine a representative frame of each video file to be named; use a convolutional neural network model to determine multiple alternative names for each video file to be named and content similarities between different representative frames based on the representative frames of each video file to be named; determine a file target name for each video file to be named based on the multiple alternative names for each video file to be named and content similarities between different representative frames; name each video file based on the file target name of each video file to be named.

[0053] The basic concepts have been described above. Obviously, for those skilled in the art, the above detailed disclosure is only for example and does not constitute a limitation of this specification. Although not explicitly stated here, those skilled in the art may make various modifications, improvements and corrections to this specification. Such modifications, improvements and corrections are suggested in this specification, so such modifications, improvements and corrections still belong to the spirit and scope of the exemplary embodiments of this specification.

[0054] At the same time, this specification uses specific words to describe the embodiments of this specification. For example, "one embodiment", "an embodiment", and / or "some embodiments" refer to a certain feature, structure or characteristic related to at least one embodiment of this specification. Therefore, it should be emphasized and noted that "one embodiment" or "an embodiment" or "an alternative embodiment" mentioned twice or more in different positions in this specification does not necessarily refer to the same embodiment. In addition, certain features, structures or characteristics in one or more embodiments of this specification can be appropriately combined.

[0055] In addition, unless explicitly stated in the claims, the order of the processing elements and sequences described in this specification, the use of alphanumeric characters, or the use of other names are not intended to limit the order of the processes and methods of this specification. Although the above disclosure discusses some invention embodiments that are currently considered useful through various examples, it should be understood that such details are only for illustrative purposes, and the attached claims are not limited to the disclosed embodiments. On the contrary, the claims are intended to cover all modifications and equivalent combinations that are consistent with the essence and scope of the embodiments of this specification. For example, although the system components described above can be implemented by hardware devices, they can also be implemented only by software solutions, such as installing the described system on an existing server or mobile device.

[0056] Similarly, it should be noted that in order to simplify the description disclosed in this specification and thus help understand one or more embodiments of the invention, in the above description of the embodiments of this specification, multiple features are sometimes combined into one embodiment, figure or description thereof. However, this disclosure method does not mean that the features required by the subject matter of this specification are more than the features mentioned in the claims. In fact, the features of the embodiments are less than all the features of the single embodiment disclosed above.

[0057] Finally, it should be understood that the embodiments described in this specification are only used to illustrate the principles of the embodiments of this specification. Other variations may also fall within the scope of this specification. Therefore, as an example and not a limitation, alternative configurations of the embodiments of this specification may be considered consistent with the teachings of this specification. Accordingly, the embodiments of this specification are not limited to the embodiments explicitly introduced and described in this specification.

Claims

1. A file naming method based on big data processing, characterized in that: include: Get multiple video files to be named; Processing the multiple video files to be named based on the file analysis model to determine a representative frame of each video file to be named; Determine multiple candidate names for each video file to be named and content similarities between different representative frames using a convolutional neural network model based on the representative frames of each video file to be named; Determine a file target name for each video file to be named based on a plurality of candidate names for each video file to be named and content similarities between different representative frames; Each video file is named based on the file target name of each video file to be named.

2. The file naming method based on big data processing as claimed in claim 1, characterized in that: The determining of the file target name of each video file to be named based on the multiple candidate names of each video file to be named and the content similarity between different representative frames includes: Constructing a file naming graph structure, wherein the file naming graph structure includes a plurality of nodes and a plurality of edges between the plurality of nodes, wherein each node represents each video file, and the node features of each node include a representative frame of each to-be-named video file and a plurality of alternative names of each to-be-named video file, and the features of the edges between the nodes represent the content similarity between different representative frames; The file naming graph structure is processed based on a graph convolutional network to determine a file target name for each video file to be named.

3. The file naming method based on big data processing as claimed in claim 1, characterized in that: The file analysis model is a gated loop unit, the input of the file analysis model is the multiple video files to be named, and the output of the file analysis model is a representative frame of each video file to be named.

4. The file naming method based on big data processing as claimed in claim 2, characterized in that: The input of the graph convolutional network is the file naming graph structure, and the output of the graph convolutional network is the file target name of each video file to be named.

5. A file naming system based on big data processing, characterized in that: include: A first acquisition module is used to acquire multiple video files to be named; A representative frame determination module, used for processing the plurality of video files to be named based on a file analysis model to determine a representative frame of each video file to be named; An alternative naming determination module, used to determine multiple alternative names for each video file to be named and content similarities between different representative frames using a convolutional neural network model based on the representative frames of each video file to be named; A target naming determination module, used to determine a file target naming of each video file to be named based on a plurality of candidate names of each video file to be named and content similarity between different representative frames; The naming module is used to name each video file based on the file target name of each video file to be named.

6. The file naming system based on big data processing as claimed in claim 5, characterized in that: The target naming determination module is also used for: Constructing a file naming graph structure, wherein the file naming graph structure includes a plurality of nodes and a plurality of edges between the plurality of nodes, wherein each node represents each video file, and the node features of each node include a representative frame of each to-be-named video file and a plurality of alternative names of each to-be-named video file, and the features of the edges between the nodes represent the content similarity between different representative frames; The file naming graph structure is processed based on a graph convolutional network to determine a file target name for each video file to be named.

7. The file naming system based on big data processing as claimed in claim 5, characterized in that: The file analysis model is a gated loop unit, the input of the file analysis model is the multiple video files to be named, and the output of the file analysis model is a representative frame of each video file to be named.

8. The file naming system based on big data processing as claimed in claim 6, characterized in that: The input of the graph convolutional network is the file naming graph structure, and the output of the graph convolutional network is the file target name of each video file to be named.

9. An electronic device, characterized in that: include: processor; Memory; And a computer program; wherein, the computer program is stored in the memory and is configured to be executed by the processor to implement the file naming method based on big data processing as described in any one of claims 1 to 4.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the file naming method based on big data processing as described in any one of claims 1 to 4 is implemented.

Images