CN112243018B

CN112243018B - Content processing method, device and storage medium

Info

Publication number: CN112243018B
Application number: CN201910655699.8A
Authority: CN
Inventors: 孙子荀
Original assignee: Tencent Technology Shenzhen Co Ltd
Current assignee: Tencent Technology Shenzhen Co Ltd
Priority date: 2019-07-19
Filing date: 2019-07-19
Publication date: 2023-03-10
Anticipated expiration: 2039-07-19
Also published as: CN112243018A

Abstract

The embodiment of the application discloses a content processing method, a content processing device and a storage medium, wherein the content to be issued and the content type of the content to be issued can be obtained; selecting a target content processing path from a plurality of candidate content processing paths based on the content type, wherein the target content processing path comprises at least two content processing branches; extracting at least two categories of sub-contents from the content to be released; distributing the sub-content to a corresponding content processing branch in the target content processing path for processing to obtain a content processing result corresponding to each sub-content; and generating the processed content to be issued based on the content processing result corresponding to each sub-content. According to the content processing method and device, at least two sub-contents of the content to be issued are distributed to the corresponding content processing branches in the target content processing path to be processed, multiple kinds of processing can be carried out simultaneously, and the content processing speed is improved.

Description

Content processing method, device and storage medium

Technical Field

The present application relates to the field of computer technologies, and in particular, to a content processing method, device, and storage medium.

Background

At present, there is a content processing platform, where a user sends content to be published to a platform, and the platform may process the content and push the content to users of multiple other platforms, but an existing content processing method generally includes multiple processing nodes, but the existing content processing method only supports serial execution of multiple processing nodes, and after the execution of a current processing node is completed, a next processing node may be executed.

Disclosure of Invention

In view of this, embodiments of the present application provide a content processing method, apparatus, and storage medium, which can improve the speed of content processing.

In a first aspect, an embodiment of the present application provides a content processing method, including:

acquiring content to be issued and the content type of the content to be issued;

selecting a target content processing path from at least two candidate content processing paths based on the content type, wherein the target content processing path comprises at least two content processing branches;

extracting at least two categories of sub-contents from the content to be released;

distributing the sub-content to a corresponding content processing branch in the target content processing path for processing to obtain a content processing result corresponding to each sub-content;

and generating the processed content to be issued according to the content processing result corresponding to each sub-content.

In some embodiments, before the distributing the sub-content to the corresponding content processing branch in the target content processing path for processing to obtain the content processing result corresponding to each sub-content, the content processing method further includes:

adjusting a content processing branch in a target content processing path based on an adjustment operation of a user for the target content processing path to obtain an adjusted content processing path;

the allocating the sub-content to the corresponding content processing branch in the target content processing path for processing includes: and distributing the sub-content to a corresponding content processing branch in the adjusted content processing path for processing.

In some embodiments, the content processing branch comprises at least one processing node; the allocating the sub-content to the corresponding content processing branch in the target content processing path for processing to obtain the content processing result corresponding to each sub-content includes:

distributing the sub-content to corresponding content processing branches in the content processing path to obtain the sub-content corresponding to each content processing branch;

determining a target processing node to be used for processing the sub-content currently according to the arrangement sequence of the processing nodes in the content processing branch;

calling the target processing node to process the sub-content corresponding to the content processing branch to obtain a processing result of the target processing node;

and obtaining a content processing result corresponding to the sub-content according to the processing result corresponding to the target processing node.

In some embodiments, the obtaining, according to the processing result corresponding to the target processing node, a content processing result corresponding to the sub-content includes:

if the processing result of the target processing node comprises a processing indication mark, outputting the processing indication mark when the processing indication mark indicates to delete the content to be released, and obtaining a content processing result corresponding to the sub-content;

when the processing indication mark indicates to execute a next processing node and the next processing node of the target processing node exists in the content processing branch, updating the target processing node to the next processing node and returning to the step of executing the calling target processing node to process the sub-content;

and when the processing indication mark indicates to execute the next processing node and the next processing node of the target processing node does not exist in the content processing branch, obtaining a content processing result corresponding to the sub-content based on the processing results corresponding to all the processing nodes in the content processing branch.

In some embodiments, the content processing branch further comprises a connection node disposed between a target processing node and a processing node next to the target processing node;

when the processing indication mark indicates to delete the content to be released, the processing indication mark is output to obtain a content processing result corresponding to the sub-content, and the method comprises the following steps:

and when the processing indication mark indicates to delete the content to be released, triggering a first connecting node between a target processing node and a next processing node of the target processing node to finish the processing of a content processing branch where the connecting node is located, and outputting the processing indication mark to obtain a content processing result corresponding to the sub-content.

when the processing indication identifier indicates that the next processing node is executed and the next processing node of the target processing node exists in the content processing branch, updating the target processing node as the next processing node, and returning to execute the step of calling the target processing node to process the sub-content, including:

when the processing indication mark indicates to execute the next processing node, triggering a second connecting node between the target processing node and the next processing node of the target processing node;

the second connecting node sends the sub-content to the next processing node of the target processing node;

and updating the target processing node to be the next processing node of the target processing node, and calling the target processing node to process the sub-content.

In some embodiments, the processing nodes include a general purpose processing layer, a base processing layer, and a logical processing layer;

the calling target processing node processes the sub-content to obtain a sub-content processing result of the target processing node, and the method comprises the following steps:

calling the general processing layer, and identifying and dividing the sub-content to obtain an identification unit;

calling the basic processing layer, and calculating the identification unit according to a preset processing algorithm to obtain a basic processing result;

calling the logic processing layer, and comparing the basic processing result with a preset condition to obtain a processing indication mark;

and outputting the basic processing result and the processing indication mark as the processing result of the target processing node.

if the processing result of the target processing node does not comprise a processing indication identifier and the next processing node of the target processing node exists in the content processing branch, updating the target processing node to be the next processing node and returning to execute the step of calling the target processing node to process the sub-content;

and if the processing result of the target processing node does not comprise the processing indication identifier and the next processing node of the target processing node does not exist in the content processing branch, obtaining a content processing result corresponding to the sub-content based on the processing results corresponding to all the processing nodes in the content processing branch.

In some embodiments, the processing node comprises a general purpose processing layer, a base processing layer, and a logical processing layer;

the invoking the target processing node to process the sub-content corresponding to the content processing branch to obtain a processing result of the target processing node includes:

calling a general processing layer of a target processing node to identify and divide the sub-content to obtain an identification unit;

calling a basic processing layer of a target processing node, and calculating the identification unit according to a preset processing algorithm to obtain a basic processing result of the target processing node;

and outputting the basic processing result as the processing result of the target processing node.

In some embodiments, the content processing method further comprises:

if the content processing result corresponding to at least one sub-content is the processing indication mark, deleting the content to be issued;

and if the content processing result corresponding to all the sub-contents is not the processing indication mark, generating the processed content to be issued according to the content processing result corresponding to the sub-contents.

In a second aspect, an embodiment of the present application provides a content processing apparatus, including:

the device comprises an acquisition unit, a processing unit and a processing unit, wherein the acquisition unit is used for acquiring the content to be issued and the content type of the content to be issued;

a selecting unit, configured to select a target content processing path from at least two candidate content processing paths based on the content type, where the target content processing path includes at least two content processing branches;

the extraction unit is used for extracting at least two categories of sub-contents from the content to be issued;

the processing unit is used for distributing the sub-contents to corresponding content processing branches in the target content processing path for processing to obtain a content processing result corresponding to each sub-content;

and the fusion unit is used for acquiring the processed content to be issued based on the content processing result corresponding to each sub-content.

In some embodiments, the processing unit comprises:

the distribution subunit is configured to distribute the sub-content to a corresponding content processing branch in the content processing path to obtain a sub-content corresponding to each content processing branch;

the determining subunit is used for determining a target processing node to be used for processing the sub-content currently according to the arrangement sequence of the processing nodes in the content processing branch;

the calling subunit is used for calling the target processing node to process the sub-content corresponding to the content processing branch to obtain a processing result of the target processing node;

and the fusion subunit is used for obtaining a content processing result corresponding to the sub-content according to the processing result corresponding to the target processing node.

In a third aspect, embodiments of the present application provide a storage medium having stored thereon a plurality of instructions for storing a computer program thereon, which, when run on a computer, causes the computer to execute a content processing method provided in any of the embodiments of the present application.

The embodiment of the application discloses a content processing method, a content processing device and a storage medium, and the content to be issued and the content type of the content to be issued can be obtained; selecting a target content processing path from at least two candidate content processing paths based on the content type, wherein the target content processing path comprises at least two content processing branches; extracting at least two categories of sub-contents from the content to be released; distributing the sub-content to a corresponding content processing branch in the target content processing path for processing to obtain a content processing result corresponding to each sub-content; and acquiring the processed content to be issued based on the content processing result corresponding to each sub-content. According to the content processing method and device, the plurality of sub-contents of the content to be issued are distributed to the corresponding content processing branches in the target content processing path for processing, multiple kinds of processing can be carried out simultaneously, and the content processing speed is improved.

Drawings

In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings needed to be used in the description of the embodiments will be briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without creative efforts.

Fig. 1 is a schematic view of an application scenario of a content processing method according to an embodiment of the present application.

Fig. 2a is a schematic flow chart of a content processing method according to an embodiment of the present application.

Fig. 2b is another schematic flow chart of the content processing method according to the embodiment of the present application.

Fig. 3a is a schematic structural diagram of a first content processing method according to an embodiment of the present application.

Fig. 3b is a schematic structural diagram of a second content processing method according to an embodiment of the present application.

Fig. 4 is a schematic diagram of a terminal provided in an embodiment of the present application.

Fig. 5a is a schematic diagram of a processing node hierarchy according to an embodiment of the present disclosure.

Fig. 5b is a schematic diagram of a video processing flow provided by the embodiment of the present application.

Fig. 5c is a schematic view of a processing flow of a video passing through different content processing paths according to an embodiment of the present application.

Fig. 5d is another schematic diagram of a processing node hierarchy according to an embodiment of the present disclosure.

Fig. 5e is a schematic structural diagram of a content processing system according to an embodiment of the present application.

Fig. 5f is a schematic diagram of a re-arranged processing node hierarchy according to an embodiment of the present application.

Detailed Description

The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.

The embodiment of the invention provides a content processing method, a content processing device and a storage medium.

An embodiment of the present invention provides a content processing system, which includes any one of the content processing apparatuses provided in the embodiments of the present invention, and the content processing apparatus may be specifically integrated in a server. The server may be represented as a single server, or may be represented as a server cluster composed of a plurality of servers.

In addition, the content processing system may further include other devices, such as a terminal, and the terminal may be: a mobile phone, a tablet Computer, a notebook Computer, or a Personal Computer (PC).

Terminals can be divided into transmitting terminals and receiving terminals. The sending terminal is used for sending the content to be issued, the receiving terminal is used for receiving the processed content to be issued, the sending terminal and the receiving terminal are expression forms of different functions of the terminals, and the same terminal can be used for expressing the function of the receiving terminal and can also be expressed as the function of the sending terminal.

For example, referring to fig. 1, in an example in which the content processing apparatus is integrated on a single server, the content processing system includes a terminal and a server, and the terminal and the server are linked via a network. The network includes network entities such as routers and gateways. A server integrated with the content processing apparatus may be used as a scheduling server, which may be in contact with a plurality of terminals in the system through network links.

For example, the user may operate on a sending terminal, and the sending terminal may obtain the content to be published based on the user behavior, and analyze the content to be published to obtain the content type of the content to be published.

The scheduling server can also select a target content processing path from a plurality of candidate content processing paths based on the content type, and extract a plurality of categories of sub-content from the content to be issued; distributing the sub-content to a corresponding content processing branch in the target content processing path for processing to obtain a content processing result corresponding to each sub-content; and generating the processed content to be issued according to the content processing result corresponding to each sub-content.

In an embodiment, the scheduling server may further send the processed content to be published to other terminals through a network link, for example, push the processed content to be published to users of the receiving terminals.

The above example of fig. 1 is only an example of a system architecture for implementing the embodiment of the present invention, and the embodiment of the present invention is not limited to the system architecture shown in fig. 1, and various embodiments of the present invention are proposed based on the system architecture.

The following are detailed below. The numbers in the following examples are not intended to limit the order of preference of the examples.

The present embodiment will be described from the perspective of a content processing apparatus, which may be integrated in a server in particular.

As shown in fig. 2a, a content processing method is provided, which can be executed by a processor of a server, and will be described below by taking an example that a content processing apparatus is integrated in a single server, and the specific flow of the content processing method is as follows:

101. and acquiring the content to be issued and the content type of the content to be issued.

The content to be issued is information which needs to be processed and identified. The content type may reflect the attribute of the content to be published, and the content type may be divided according to actual requirements, for example, the content to be published may be divided into a video, a text, a sound, and a combination thereof based on the data format of the content to be published.

In an embodiment, the content to be issued may be acquired based on the content acquisition instruction. For example, the content obtaining instruction may carry the content to be published, and the scheduling server may parse the content obtaining instruction to obtain the content to be published.

In one embodiment, an instruction interface may be provided for facilitating triggering of the content acquisition instruction. The representation form of the command interface is various, such as an input box, a selection box, an icon, a button and the like.

In an embodiment, the content obtaining instruction may be triggered based on a user behavior, for example, a sending button may be set on a terminal interface, the content obtaining instruction may be triggered by clicking the sending button after the user edits the content to be sent, and then the server is scheduled, that is, the content to be issued carried by the instruction may be obtained based on the content obtaining instruction.

In one embodiment, referring to fig. 5e, the content to be published may be stored in the storage server through the instruction interface.

102. Based on the content type, a target content processing path is selected from a plurality of candidate content processing paths, wherein the target content processing path comprises a plurality of content processing branches.

The content processing path is a flow for processing the content to be issued.

The candidate content processing path may be from a database storing the content processing path, or from a real-time record of local memory aggregation.

For example, the target processing path may be selected according to a correspondence between the sub-content of the content to be distributed and the content processing branch. For example, the data format of the content to be published is a video format, the video is generally composed of multiple frames of pictures, audio and characters, the content type of the content to be published is the video format, sub-content such as picture frames, audio, text, characteristic fingerprints and the like can be extracted, and a content processing path with a corresponding content processing branch is selected as a target content processing path. The corresponding content processing branch means that the content processing branch can process sub-content of the corresponding data format.

In an embodiment, a target content processing path may also be selected according to a requirement of a receiving terminal that receives the processed content to be distributed, referring to fig. 5c, because different receiving terminals have different requirements for the processed content to be distributed, for example, some receiving terminals may only receive videos in a conventional format, and some receiving terminals may only receive emerging formats with higher bandwidth efficiency, so that a target content processing path including or not including a video transcoding step needs to be selected according to the requirement of the receiving terminal, in addition, different receiving terminals have different processing requirements for the processed content to be distributed, referring to fig. 5c, for example, different terminals have different conditions for yellow gambling poison passing, some auditing receiving terminals do not need to perform deduplication processing, and some receiving terminals do not need to add labels or covers. In addition, after the same content to be published is processed through different processing paths, the obtained processed content to be published is different.

103. At least two categories of sub-content are extracted from the content to be published.

The sub-content is a content in a single data format that constitutes the content to be published, for example, the content to be published is an article, and then text information and a feature fingerprint in the content to be published can be generally extracted as the sub-content.

Accordingly, the category of the sub-content is the data format attribute of the sub-content.

The characteristic fingerprint is a parameter for distinguishing the content to be published from other content, and can be understood as vector representation of a line of text. The more similar the characteristic fingerprints, the more similar the corresponding content.

In another embodiment, the data format of the content to be distributed may be analyzed, and then the content data in a single format in the content to be distributed may be copied as the sub-content.

In one embodiment, video or audio may be transcoded, and content to be published input by a user is decoded and a reconstructed video frame or audio frame is generated as one of the sub-content of the video or audio.

104. And distributing the sub-content to the corresponding content processing branch in the target content processing path for processing to obtain a content processing result corresponding to each sub-content.

The type of the sub-content corresponds to the type of the content processing branch, for example, the data type of the sub-content is a text type, and the corresponding content processing path is a path for processing a text.

And the content processing result corresponding to each sub-content is a result obtained after the content processing branch corresponding to the sub-content is executed.

In an embodiment, before allocating the sub-content to a corresponding content processing branch in the target content processing path for processing to obtain a content processing result corresponding to each sub-content, the method further includes:

and adjusting the content processing branch in the target content processing path based on the adjustment operation of the user for the target content processing path to obtain the adjusted content processing path.

In an embodiment, a content processing branch in the target content processing path may be adjusted by obtaining a processing node movement instruction. The representation form of the trigger interface of the instruction is various, for example, an input box, a selection box, an icon, a button and the like.

In an embodiment, in order to facilitate the user to observe and analyze the content processing path, after the target content processing path is obtained, the scheduling server may send the target content processing path to the adjustment terminal through the network link, and display the target content processing path on an operation interface of the adjustment terminal.

In an embodiment, in order to improve the accuracy of content processing, a user of the terminal may drag a processing node in a target content processing path on an interface to generate a processing node moving instruction, and the scheduling server may change the order of invoking the processing nodes according to the processing node moving instruction. The operation of the user can be displayed on the interface of the adjusting terminal, so that the user can conveniently adjust the processing flow.

In an embodiment, the sub-content is assigned to the corresponding content processing branch according to the data type, for example, the picture frame sub-content is assigned to the content processing branch for processing pictures, and the text sub-content is assigned to the content processing branch for processing texts.

In an embodiment, the content processing branch comprises at least one processing node; the allocating the sub-content to the corresponding content processing branch in the target content processing path to process, so as to obtain a content processing result corresponding to each sub-content, which may specifically be implemented by the following steps:

The processing node is a node for processing and identifying the sub-content, and may be a processor or a processing algorithm. Different processing nodes can respectively realize different processing and identification of sub-content, such as identifying heavy content, identifying watermarks, label association, extracting cover drawings and the like.

Referring to fig. 5e, for example, the processing node may perform security quality processing on the sub-content, (which may specifically include quality evaluation, violence, vulgar, title party, wrongly written characters, etc.), perform feature recognition processing (which may specifically include classification, theme, label, etc.) on the sub-content, and perform understanding and generating processing (which may specifically include cover map, abstract, structuring, clipping, etc.) on the sub-content.

There are many processing results of the target processing node, which can be mainly divided into two types, one type is processed sub-content, for example, a picture frame after removing a watermark; the other is a judgment result as to whether the processing result meets the requirement, for example, after recognizing that the repetition rate is lower than a preset value, the processing result of the target processing node is the requirement, and the next processing node is executed.

In an embodiment, the processing result of the target processing node may be stored in the server, and when the target processing node does not exist in the content processing branch, the processing result fusion instruction is triggered, and the scheduling server may obtain the processing results corresponding to all the processing nodes stored in the server according to the processing result fusion instruction, and fuse the processing results to serve as the content processing result corresponding to the sub-content. For example, the image frames with the watermarks removed, the black edges removed, and the image quality improved may be fused and used as the content processing result of the content processing branch corresponding to the image frame.

In an embodiment, the obtaining a content processing result corresponding to the sub-content according to the processing result corresponding to the target processing node may specifically include the following steps:

The processing indication flag may be a determination result of whether the sub-content meets a preset condition, and the determination result may indicate a next operation of the scheduling server. For example, if the processing instruction identifier is an ineligible condition, the scheduling server stops executing other processing nodes in the content processing branch where the target processing node is located, and outputs the ineligible condition (i.e., the processing instruction identifier) as a content processing result of the sub-content corresponding to the content processing branch. On the contrary, if the processing indication flag is in accordance with the condition, the scheduling server calls the next processing node of the target processing node according to the arrangement sequence of the processing nodes in the content processing branch to continue processing the sub-content. If the next processing node of the target processing node does not exist, the scheduling server obtains the processing results corresponding to the sub-content from the processing results of all the processing nodes in the merged content processing branch, for example, the sub-content processed by the processing nodes is merged as the processing result corresponding to the sub-content.

In an embodiment, the content processing branch further comprises a connection node, the connection node being disposed between a target processing node and a processing node next to the target processing node;

when the processing indicator indicates to delete the content to be published, the processing indicator is output to obtain a content processing result corresponding to the sub-content, and the method specifically includes the following steps:

The connection node is a node for indicating a call sequence of the processing units in the content processing branch, and specifically, the connection node may be a processor, or may be an algorithm, and the algorithm may be stored in the scheduling server. For example, the first connection node may generate an instruction to stop executing the processing node, and the scheduling server obtains the instruction, then stops executing other processing nodes in the content processing branch where the target processing node is located, and outputs a processing indication identifier (that is, an ineligible condition) as a content processing result corresponding to the sub-content corresponding to the content processing branch.

In an embodiment, when the processing indication identifier indicates that a next processing node is executed and the next processing node of the target processing node exists in the content processing branch, the step of updating the target processing node to the next processing node and returning to execute the step of invoking the target processing node to process the sub-content includes:

when the processing indication mark indicates to execute a next processing node, triggering a second connection node between a target processing node and the next processing node of the target processing node;

The second connection node may generate an instruction for invoking a next processing unit, and the scheduling server may obtain the instruction, that is, may invoke the next processing unit of the target processing unit according to the instruction.

When the processing indication identifier indicates that the next processing node is executed and the next processing node of the target processing node does not exist in the content processing branch, the connection node does not exist behind the target processing node, and the scheduling server fuses the processing results of all the processing nodes in the content processing branch to obtain the processing result corresponding to the sub-content, for example, the sub-content processed by the processing nodes is fused to be used as the processing result corresponding to the sub-content.

In one embodiment, the processing nodes include a general purpose processing layer, a base processing layer, and a logical processing layer;

Referring to fig. 5a, a processing node includes 3 processing layers, where a processing layer is a step for implementing processing of the processing node on sub-content, for example, a general processing layer is a first step of processing the sub-content, the first step of processing generally identifies and divides the sub-content to obtain an identification unit, a basic processing layer is a second step of processing the sub-content, the second step of processing generally refers to calculation performed on the identification unit, a logic processing layer is a third step of processing the sub-content, and the third step of processing may be a step for determining whether a calculation result satisfies a preset condition. As shown in fig. 5d, taking text analysis as an example, the general processing layer may perform syntactic analysis, sentence segmentation, and lexical division on the obtained text data to obtain words and sentences as the recognition units, the basic processing layer may perform shallow semantic understanding according to the recognition units, classify the recognition units according to semantic understanding results, generate word vectors and sentence vectors according to the recognition units, calculate similarity, context distribution probability, and the like of the word vectors and the sentence vectors, further obtain historical release contents, perform heat analysis according to the historical release contents, and the logic processing layer may perform quality discrimination, abstract generation, heat discrimination, and the like through wrongly-recognized words and keywords according to the processing results of the basic processing layer.

For example, referring to fig. 5f, taking the deduplication process as an example, the general processing layer may be configured to obtain a picture fingerprint, an audio fingerprint, a semantic feature, or a text feature, specifically, the text feature may be represented by a word vector, the basic processing layer may obtain historical release content in a local memory or other servers through a network link, calculate similarity between the historical release content and content to be released according to the picture fingerprint, the audio fingerprint, the semantic feature, or the text feature, perform video duplication recognition, title duplication recognition, or audio duplication recognition, obtain a duplication rate of the sub-content, and the logic processing layer may determine whether the duplication rate satisfies a preset condition. .

For example, taking label extraction as an example, the general processing layer may be configured to divide words and a map, identify parts of speech of the words or identify characters in the map, calculate the importance degree of the words or the characters in the article or the video according to the words or the characters, specifically, calculate the word frequency and the inverse document frequency of the words or the characters by using a tf-idf (term frequency-inverse document frequency) algorithm, then calculate the importance degree parameter of the words or the characters in the article or the video according to a product of the word frequency and the inverse document frequency, and select a target label of the article or the video from the words or the characters according to a calculation result, and the scheduling server may obtain the target label according to a label interface. The label interface is an external protocol and interface.

In an embodiment, the logic processing layer may adjust based on a requirement of a terminal that receives the processed content to be distributed, for example, the preset condition may be changed, so as to delete the content to be distributed that does not meet the requirement of the receiving terminal.

In one embodiment, the user may edit the preset conditions within the logical processing layer on the page of the terminal, thereby adjusting the target content processing path.

In an embodiment, the scheduling server may invoke the processing result of the logical processing layer through an external protocol and an interface.

In one embodiment, if the generic processing layer and the basic processing layer are stored locally, then the generic processing layer and the basic processing layer need to be encapsulated by a generic Remote Procedure Call (RPC) request and connected to the logical processing layer, so that the logical processing layer can Call the processing result of the basic processing layer.

In an embodiment, the basic processing layer may be connected to the local storage or another server connected to the network through the external protocol and the interface, so that the basic processing layer may obtain the historical release content through the external protocol and the interface, and perform basic processing on the sub-content according to the historical release content. For example, the base processing layer may identify the repeated portions of the historical published content and the to-be-published content by comparing the historical published content and the to-be-published content. The historical release content can be from a database for storing the historical release content, and can also be from a real-time record summarized by a local memory.

In an embodiment, the obtaining, according to the processing result corresponding to the target processing node, a content processing result corresponding to the sub-content includes:

In an embodiment, the basic processing layer may obtain the historical release content through the external protocol and the interface, and perform the basic processing on the sub-content according to the historical release content, which is not described herein again.

In one embodiment, the processing node comprises a general processing layer, a base processing layer, and a logical processing layer;

the calling the target processing node to process the sub-content corresponding to the content processing branch to obtain a processing result of the target processing node, including:

In one embodiment, the dispatch center server may invoke the processing results of the base processing layer by using the external protocol and the interface.

In an embodiment, if the generic processing layer and the basic processing layer are stored locally, the generic processing layer and the basic processing layer need to be encapsulated by a generic Remote Procedure Call (RPC) request mode and connected to an external protocol and an interface, so that the scheduling center server can Call the processing result of the basic processing layer.

In an embodiment, the generating the processed content to be published according to the content processing result corresponding to each sub-content includes:

and if the content processing results corresponding to all the sub-contents do not comprise the processing indication marks, generating the processed content to be released according to the content processing results corresponding to the sub-contents.

If the content processing result corresponding to at least one sub-content is the processing indication mark, a command for deleting the content to be issued is generated, and the scheduling server can delete the corresponding content to be issued according to the command.

If the content processing results corresponding to all the sub-contents do not contain the processing indication identifiers, a to-be-issued content command after the generation and the processing is generated, the scheduling server can analyze the command and fuse the content processing results corresponding to each sub-content to generate the to-be-issued content after the processing. For example, the content processing result corresponding to the sub-content is the processed sub-content, and the scheduling server may fuse all the processed sub-contents according to the original structure of the content to be released, so as to obtain the processed content to be released. Specifically, the processed picture frame, the audio and the text may be combined according to the original structure to generate the processed content to be published.

In an embodiment, the scheduling server may send the processed content to be published to a user of a corresponding receiving terminal through a network link.

In another embodiment, the processed content to be published is stored in a storage server, the storage server may generate a publication content message corresponding to the processed content to be published, and send the publication content message to a user of a corresponding receiving terminal through a network link, and the receiving terminal user may obtain the processed content to be published in the storage server through the publication content message. The content distribution message may be distributed to a single receiving terminal or may be distributed to a plurality of receiving terminals in a group. The release content message may include the title, the category label, the cover page, the source, the id, and other contents of the processed content to be released.

In an embodiment, referring to fig. 5e, the scheduling server may, in addition to playing a role of invoking a processing node and adjusting a path, adjust a processing rate of a target content processing path, and relieve a pressure of heavy usage, for example, when a plurality of contents to be published are processed simultaneously, if the same processing node needs to be invoked, the scheduling server may end a time sequence of a processing node on the processing node, and schedule a time for invoking the processing node in different content processing paths, so as to avoid a process flow disorder.

In addition, referring to fig. 5e, the scheduling server may also monitor the call sequence of the processing nodes to play a role in monitoring the processing flow, and in addition, the scheduling server may also dynamically adjust the number of requests for concurrently calling different processing nodes in combination with the capacity model to play a role in controlling the speed of processing the published content and controlling the publishing flow. The capacity model can be obtained by off-line training in other servers through a Machine Learning (ML) database, and the scheduling server can download the trained capacity model through a network. Machine Learning (ML) is a multi-domain cross subject, and relates to multiple subjects such as probability theory, statistics, approximation theory, convex analysis and algorithm complexity theory. The method specially studies how a computer simulates or realizes the learning behavior of human beings so as to acquire new knowledge or skills and reorganize the existing knowledge structure to continuously improve the performance of the computer.

In an embodiment, referring to fig. 5b, before extracting the sub-content of the content to be published, a general content processing path may be selected from the candidate content processing paths according to the content type, for example, the general content processing path may include a plurality of general processing branches, and each general processing branch has a general processing node thereon, for example, detail page generation, content classification, title processing, hotspot judgment, video decoding, and the like. These general processing branches are basically independent of the single sub-content of the content to be published and are generally only used to process the whole content to be published. In some embodiments, after the video is decoded and the sub-content is extracted, the target content processing path may be run to obtain a content processing result corresponding to the sub-content, and finally, the content to be released after processing may be generated according to the content processing result corresponding to the sub-content and the general processing branch.

As can be seen from the above, the content to be published and the content type of the content to be published can be obtained in the embodiment of the present application; selecting a target content processing path from a plurality of candidate content processing paths based on the content type, wherein the target content processing path comprises a plurality of content processing branches; extracting sub-contents of a plurality of categories from the content to be published; distributing the sub-content to a corresponding content processing branch in the target content processing path for processing to obtain a content processing result corresponding to each sub-content; and generating the processed content to be issued according to the content processing result corresponding to each sub-content. Therefore, the content processing speed is improved by using the plurality of content processing branches to process the sub-content of the content to be issued at the same time.

The method described in the above embodiments is further described in detail below.

In this embodiment, a method according to an embodiment of the present invention will be described in detail by taking an example in which a content processing apparatus is integrated in a single server to perform video processing.

As shown in fig. 2b, the content processing method of the present application will be described in the embodiment by taking an example that the content processing apparatus is integrated in a server cluster, where the server cluster is composed of a plurality of servers, including a scheduling center server for scheduling other servers and terminals, a content analysis server for analyzing content data of content to be distributed, a processing node server for processing and identifying sub-content, a storage server for storing historical distribution content and candidate content processing paths, and the like.

The content processing method of the application comprises the following specific processes:

201. the server obtains the content to be released and the content type of the content to be released.

In this embodiment, an input frame and a sending button are arranged on an interface of a sending terminal, a user sending terminal inputs content to be sent in the input frame, and then clicks the sending button, a content obtaining instruction can be triggered, the content obtaining instruction carries the content to be published, a scheduling center server can obtain the content to be published based on the content obtaining instruction, then the scheduling center server can send the content to be published to a content analysis server, the content analysis server can analyze a data format of the content to be published, a content type of the content to be published is determined according to the data format of the content to be published, and then the scheduling center server can obtain a corresponding content type from the content analysis server through a network link.

For example, the content to be published is an article with a video inserted therein, the user inputs the content to be published on an input interface of the terminal and clicks a sending button, the scheduling center server may obtain the article with the video inserted therein and send the article to the content analysis server, and the content analysis server may determine that the content type of the content to be published is a combination of the video and a text, and then send the content type to the scheduling center server.

202. The server selects a target content processing path from the plurality of candidate content processing paths based on the content type.

In an embodiment, the scheduling center server stores a mapping relationship between a content type and a content processing path, where the mapping relationship may be stored in a form of a linked list, a table, a hash table (hash), an array, a tree (tree), and the like, and the scheduling center server may obtain a target content processing path from the storage server according to the mapping relationship.

The dispatching center server can select a target processing path according to the corresponding relation between the sub-content of the content to be issued and the content processing branch. For the specific steps, reference is made to the foregoing embodiments, which are not described herein again.

203. The server extracts at least two categories of sub-content from the content to be published.

In an embodiment, the server may obtain the reconstructed video frame and audio frame through video decoding, then extract each frame of picture and audio according to the video frame and audio frame, and calculate the feature fingerprint thereof as the sub-content of the content to be distributed.

The pictures can contain characters, the characters in the pictures can be extracted to serve as text sub-contents, and the characters in the pictures need to be identified when advertisement filtering and yellow betting virus auditing are carried out.

204. And the server distributes the sub-content to the corresponding content processing branch in the target content processing path for processing to obtain a content processing result corresponding to each sub-content.

The type of the sub-content corresponds to the type of the content processing branch, and the content processing result corresponding to each sub-content is a result obtained after the content processing branch corresponding to the sub-content is executed.

The scheduling center server may send the sub-content to the corresponding processing node server according to the arrangement order of the processing nodes in the target processing path, where the processing node server stores the corresponding processing algorithm, and may obtain the processing result of the processing node according to the running algorithm.

If the processing result of the processing node does not include the processing indication identifier, the scheduling center server may obtain the processing result of the processing node in the processing node server according to the instruction, send the processing result of the processing node to the server for fusing the processing results, obtain the content processing result corresponding to the sub-content, and send the content processing result corresponding to the sub-content to the scheduling center server according to the instruction. For example, for a watermark identification process of a picture frame in a video, the picture frame with the determined watermark position may be sent to a dispatch center server. The dispatch center server may then send the sub-content to the next processing node server according to the order of the processing nodes in the content processing branch. For example, after determining the watermark position, the dispatch center server may send the picture frame of the watermark position to the processing node server where the watermark is removed.

In one embodiment, the current processing node server must process the sub-content based on the processing results of the previous processing node server. The base processing layer of the previous processing node may be connected to the generic processing layer of the current processing node via an external protocol and interface. The dispatching center server can send the processed sub-content to the current processing node server through an external protocol and an interface.

For example, for the identification and removal of picture frames in a video, for example, after the previous processing node server sends the picture frame at the determined watermark position to the scheduling center server, the scheduling center server may send the picture frame at the watermark position to the processing node server for removing the watermark.

In an embodiment, the algorithm of the connection node is stored in the dispatch center server, and if the processing result of the processing node includes a processing indication identifier, after the dispatch center server obtains the processing indication identifier, the dispatch center server may trigger a corresponding instruction for executing a next operation according to the processing indication identifier, for example, if the processing indication identifier is not compliant, the dispatch center server stops executing the content processing branch where the processing node is located, and takes the processing indication identifier as the content processing result corresponding to the sub-content. Accordingly, if the processing indication is identified as eligible, the dispatch center server may send the sub-content to the next processing node server according to the order of the processing nodes in the content processing branch.

For example, for the recognition filtering processing of the video, when the processing node server determines that the repetition rate of the video is higher than the preset value, the processing indication mark is an ineligible condition, the scheduling center server stops executing the content processing branch where the processing node is located, and the ineligible condition is taken as the content processing result corresponding to the sub-content. When the processing node server determines that the repetition of the video is lower than the preset value, the processing indication mark is in accordance with the condition, and the scheduling center server can send the sub-content to the next processing node server according to the sequence of the processing nodes in the content processing branch.

In some embodiments, referring to fig. 5e, the content processing system further includes an auditing terminal for auditing management, and the server may generate an audio or video auditing report according to the sub-content, and send the report to the auditing terminal for manual auditing. For example, an operator may classify the sub-content according to the audio or video audit report, add a classification flag, and trigger the processed content to be distributed to the receiving terminal.

In an embodiment, the auditing terminal may further obtain an account of the terminal user who issued the content to be issued, and start a green channel mode or a fast channel mode for some accounts with higher quality scores. The green channel mode and the fast channel mode may be simplified candidate content processing paths with fewer processing nodes. In addition, the scheduling server may also select a target content processing path, specifically, the scheduling server may send the selected target content processing path to the auditing terminal and display the target content processing path on the auditing terminal, and an operator may adjust a processing node in the target content processing path on a page of the auditing terminal to obtain an adjusted content processing path.

Referring to fig. 5e, the content processing system further includes an operation terminal for operation management, where the operation terminal may obtain feedback information of the receiving terminal, where the feedback information may include a click rate and an evaluation, obtain recent hotspot information according to the click rate of the receiving terminal user on the published content, score the published content according to the evaluation of the receiving terminal user, and delete the processed content to be published with a lower score in the storage server, for example, may perform off-shelf processing on the processed content to be published with a lower score.

The processing node comprises a general processing layer, a basic processing layer and a logic processing layer, wherein the processing layer is an implementation step for implementing the processing of the processing node on the sub-content, the basic processing layer is a second step of processing on the sub-content, and the logic processing layer is a third step of processing on the sub-content.

In some embodiments, when the sub-content is subjected to basic processing, the historical published content in the storage server needs to be called, for example, when the characters of the video are identified, the sub-content needs to be compared with the content data of the historical published content in the storage server. The general processing layer can be packaged in a universal Remote Procedure Call (RPC) request mode and connected with the basic processing layer, the sentence cutting, part-of-speech tagging and syntax analysis results of the text are sent to the basic processing layer, and the scheduling center server sends the historical release content in the storage server to the basic processing layer through an external protocol and an interface so as to compare the sub-content of the basic processing layer with the content data of the historical release content. And then sending the comparison result to a logic processing layer, judging whether the repetition rate meets the condition or not according to the preset condition in the logic processing layer, and sending a processing indication mark to a scheduling center server.

205. And the server generates the processed content to be issued according to the content processing result corresponding to each sub-content.

If the content processing result corresponding to at least one sub-content is the processing indication mark, deleting the content to be issued; and if the content processing results corresponding to all the sub-contents do not comprise the processing indication marks, generating the processed content to be released according to the content processing results corresponding to the sub-contents.

If the content processing result corresponding to at least one sub-content is the processing indication mark, the dispatching center server generates a command for deleting the content to be issued according to the processing indication mark (namely, the processing indication mark does not meet the condition), and deletes the content to be issued in the memory according to the command.

If the content processing results corresponding to all the sub-contents do not comprise the processing indication marks, the scheduling center server can send the content processing results corresponding to the sub-contents to the server for fusing the processing results, and the server for fusing the processing results can combine the processed sub-contents according to the original structure of the content to be released to generate the processed content to be released. And according to the instruction, the processed content to be issued is sent to the dispatching center server.

For the specific steps, reference is made to the above embodiments, which are not described herein again.

206. And the server sends the processed content to be issued to the corresponding receiving terminal.

The receiving terminal may include a plurality of terminals, and the receiving terminal may include a transmitting terminal that transmits the content to be transmitted.

In some embodiments, referring to fig. 5e, the dispatch center server may push the processed content to be published to the receiving terminal through the push interface. For example, an article published by a public account is pushed to a subscriber account. The push interface is an interface used for triggering a push instruction, the push instruction can carry the processed content to be issued and the identification information of the receiving terminal, and the content to be issued is sent to the receiving terminal according to the identification information of the receiving terminal.

As can be seen from the above, the content to be published and the content type of the content to be published can be obtained in the embodiment of the present application; selecting a target content processing path from a plurality of candidate content processing paths based on the content type, wherein the target content processing path comprises a plurality of content processing branches; extracting sub-contents of a plurality of categories from the content to be published; distributing the sub-content to a corresponding content processing branch in the target content processing path for processing to obtain a content processing result corresponding to each sub-content; and generating the processed content to be issued according to the content processing result corresponding to each sub-content. Therefore, the content processing speed is improved by using the plurality of content processing branches to simultaneously process the sub-content of the content to be issued.

In order to better implement the above method, an embodiment of the present invention further provides a content processing apparatus, where the content processing apparatus may be specifically integrated in an electronic device, and the electronic device may be a terminal, a server, a personal computer, or another device.

For example, in this embodiment, the method according to the embodiment of the present invention will be described in detail by taking an example in which the content processing apparatus is integrated in a server.

For example, as shown in fig. 3a, the content processing apparatus may include an acquisition unit 301, a selection unit 302, an extraction unit 303, a processing unit 304, and a fusion unit 305. The following were used:

the first acquiring unit 301 is configured to acquire content to be published and a content type of the content to be published.

In some embodiments, the obtaining unit 301 may specifically be configured to:

the content to be published may be obtained based on the content obtaining instruction. For example, the content obtaining instruction may carry the content to be published, and the server may parse the content obtaining instruction to obtain the content to be published.

A second selecting unit 302, configured to select a target content processing path from a plurality of candidate content processing paths based on the content type, where the target content processing path includes a plurality of content processing branches.

In some embodiments, after the selecting unit 302 is executed, an adjusting unit is further executed, where the adjusting unit is specifically configured to: adjusting a content processing branch in a target content processing path based on an adjustment operation of a user for the target content processing path to obtain an adjusted content processing path;

wherein, the allocating the sub-content to the corresponding content processing branch in the target content processing path for processing includes: and distributing the sub-content to a corresponding content processing branch in the adjusted content processing path for processing.

And a (third) extracting unit 303, configured to extract sub-content of multiple categories from the content to be distributed.

The sub-content is a component of the content to be released in a single data format.

A processing unit 304, configured to allocate the sub-content to a corresponding content processing branch in the target content processing path for processing, so as to obtain a content processing result corresponding to each sub-content.

Referring to fig. 3b, in an embodiment, the processing unit 304 may include an allocating subunit 3041, a determining subunit 3042, a calling subunit 3043, and a fusing subunit 3044, as follows:

an allocating subunit 3041, configured to allocate the sub-content to a corresponding content processing branch in the content processing path, to obtain a sub-content corresponding to each content processing branch;

a determining subunit 3042, configured to determine, according to an arrangement order of the processing nodes in the content processing branch, a target processing node currently to be used for processing the sub-content;

a calling subunit 3043, configured to call the target processing node to process the sub-content corresponding to the content processing branch, so as to obtain a processing result of the target processing node;

and a fusion subunit 3044, configured to obtain a content processing result corresponding to the sub-content according to the processing result corresponding to the target processing node.

Wherein the fusogenic subunit 3044 is specifically for:

If the processing result of the target processing node does not include the processing indication identifier and the next processing node of the target processing node exists in the content processing branch, updating the target processing node to the next processing node and returning to the step of calling the target processing node to process the sub-content;

and if the processing result of the target processing node does not include the processing indication identifier and the next processing node of the target processing node does not exist in the content processing branch, obtaining a content processing result corresponding to the sub-content based on the processing results corresponding to all the processing nodes in the content processing branch.

In an embodiment, the calling subunit 3043 is specifically configured to:

calling the general processing layer to carry out general processing on the sub-content to obtain a general processing result;

calling the basic processing layer according to the general processing result, and performing basic processing on the sub-content to obtain a basic processing result;

calling the logic processing layer according to the basic processing result to perform logic processing on the sub-content to obtain a processing indication mark;

In an embodiment, the call subunit 3043 may also be configured to:

calling a general processing layer of the target processing node to process the sub-content to obtain a general processing result;

calling a basic processing layer of a target processing node according to the general processing result, and performing basic processing on the sub-content to obtain a basic processing result of the target processing node;

And (v) a merging unit 305, configured to generate the processed content to be published according to the content processing result corresponding to each sub-content.

In one embodiment, the fusion unit 305 may include a delete subunit and a generate subunit, as follows:

the deleting subunit is configured to delete the content to be published if a content processing result corresponding to at least one sub-content is the processing indication identifier;

and the generating subunit is configured to generate the processed content to be released according to the content processing result corresponding to the sub-content if the content processing result corresponding to all the sub-contents does not include the blowing indication identifier.

In a specific implementation, the above units may be implemented as independent entities, or may be combined arbitrarily to be implemented as the same or several entities, and the specific implementation of the above units may refer to the foregoing method embodiments, which are not described herein again.

As can be seen from the above, in the content processing apparatus according to the embodiment of the present invention, the obtaining unit obtains the content to be distributed and the content type of the content to be distributed; selecting, by a selection unit, a target content processing path from a plurality of candidate content processing paths based on the content type, wherein the target content processing path comprises a plurality of content processing branches; extracting sub-contents of a plurality of categories from the content to be published by an extraction unit; distributing the sub-content to a corresponding content processing branch in the target content processing path by a processing unit for processing to obtain a content processing result corresponding to each sub-content; and generating the processed content to be issued according to the content processing result corresponding to each sub-content by the fusion unit.

According to the content processing method and device, the sub-content of the content to be issued is processed simultaneously by using the plurality of content processing branches, so that the content processing speed is improved.

The embodiment of the application also provides a server, and the server can be a mobile phone, a tablet computer, a micro processing box, an unmanned aerial vehicle or an image acquisition device and the like. As shown in fig. 4, it shows a schematic structural diagram of a server according to an embodiment of the present application, specifically:

the server may include components such as a processor 401 of one or more processing cores, memory 402 of one or more computer-readable storage media, a power supply 403, an input module 404, and a communication module 405. Those skilled in the art will appreciate that the server architecture shown in FIG. 4 is not meant to be limiting, and may include more or fewer components than shown, or some components may be combined, or a different arrangement of components. Wherein:

the processor 401 is a control center of the server, connects various parts of the entire server using various interfaces and lines, and performs various functions of the server and processes data by running or executing software programs and/or modules stored in the memory 402 and calling data stored in the memory 402, thereby performing overall monitoring of the server. In some embodiments, processor 401 may include one or more processing cores; in some embodiments, processor 401 may integrate an application processor, which primarily handles operating systems, user interfaces, applications, etc., and a modem processor, which primarily handles wireless communications. It will be appreciated that the modem processor described above may not be integrated into the processor 401.

The memory 402 may be used to store software programs and modules, and the processor 401 executes various functional applications and data processing by operating the software programs and modules stored in the memory 402. The memory 402 may mainly include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application program required by at least one function (such as a sound playing function, an image playing function, etc.), and the like; the storage data area may store data created according to the use of the server, and the like. Further, the memory 402 may include high speed random access memory, and may also include non-volatile memory, such as at least one magnetic disk storage device, flash memory device, or other volatile solid state storage device. Accordingly, the memory 402 may also include a memory controller to provide the processor 401 access to the memory 402.

The server also includes a power supply 403 for supplying power to the various components, and in some embodiments, the power supply 403 may be logically connected to the processor 401 via a power management system, so that the functions of managing charging, discharging, and power consumption are implemented via the power management system. The power supply 403 may also include any component of one or more dc or ac power sources, recharging systems, power failure detection circuitry, power converters or inverters, power status indicators, and the like.

The server may also include an input module 404, the input module 404 operable to receive entered numeric or character information and generate keyboard, mouse, joystick, optical or trackball signal inputs related to user settings and function control.

The server may also include a communication module 405, and in some embodiments, the communication module 405 may include a wireless sub-module through which the server may wirelessly transmit over short distances to provide wireless broadband internet access. For example, the communication module 405 may be used to assist a user in sending and receiving e-mails, browsing web pages, accessing streaming media, and the like.

Although not shown, the server may further include a display unit and the like, which will not be described in detail herein. Specifically, in this embodiment, the processor 401 in the server loads the executable file corresponding to the process of one or more application programs into the memory 402 according to the following instructions, and the processor 401 runs the application program stored in the memory 402, thereby implementing various functions as follows:

acquiring content to be published and the content type of the content to be published;

selecting a target content processing path from a plurality of candidate content processing paths based on the content type, wherein the target content processing path comprises a plurality of content processing branches;

extracting sub-contents of a plurality of categories from the content to be published;

The above operations can be implemented in the foregoing embodiments, and are not described in detail herein.

As can be seen from the above, the present embodiment can obtain the content to be published and the content type of the content to be published; selecting a target content processing path from a plurality of candidate content processing paths based on the content type, wherein the target content processing path comprises a plurality of content processing branches; extracting sub-contents of a plurality of categories from the content to be published; distributing the sub-content to a corresponding content processing branch in the target content processing path for processing to obtain a content processing result corresponding to each sub-content; and generating the processed content to be issued according to the content processing result corresponding to each sub-content. Therefore, the content processing speed is improved by using the plurality of content processing branches to simultaneously process the sub-content of the content to be issued.

It will be understood by those skilled in the art that all or part of the steps of the methods of the above embodiments may be performed by instructions, or by instructions controlling associated hardware, which may be stored in a computer-readable storage medium and loaded and executed by a processor.

To this end, the present application provides a storage medium, in which a plurality of instructions are stored, and the instructions can be loaded by a processor to execute the steps in any one of the content processing methods provided by the embodiments of the present application. For example, the instructions may perform the steps of:

Wherein the storage medium may include: read Only Memory (ROM), random Access Memory (RAM), magnetic or optical disks, and the like.

Since the instructions stored in the storage medium may execute the steps in any content processing method provided in the embodiments of the present application, beneficial effects that can be achieved by any content processing method provided in the embodiments of the present application may be achieved, for details, see the foregoing embodiments, and are not described herein again.

The content processing method, device and storage medium provided in the embodiments of the present application are described in detail above, and specific examples are applied herein to explain the principles and embodiments of the present application, and the description of the embodiments is only used to help understand the method and its core idea of the present application; meanwhile, for those skilled in the art, according to the idea of the present application, there may be variations in the specific embodiments and the application scope, and in summary, the content of the present specification should not be construed as a limitation to the present application.

Claims

1. A method for processing content, comprising:

acquiring content to be published and a content type of the content to be published, wherein the content type to be published comprises a video format;

based on the content type, selecting a target content processing path from at least two candidate content processing paths, wherein the target content processing path is a process for processing the content to be issued, the target content processing path comprises at least two content processing branches, the content processing branches comprise at least one processing node, and the at least two content processing branches comprise a text content processing branch, an audio feature processing branch, a key frame processing branch and a video feature processing branch which are parallel to each other;

extracting sub-content of at least two categories from content to be released, wherein the categories are attributes of the sub-content about data formats, and the sub-content of the at least two categories comprises text content, audio features, key frames and video features; the category of the sub-content corresponds to the type of the content processing branch;

determining a target processing node to be used for processing the sub-content currently according to the arrangement sequence of the processing nodes in the content processing branch; processing nodes in the text content processing branch line sequentially comprise a text extraction node, a text de-duplication node and a text title de-duplication node; the processing nodes in the audio characteristic processing branch comprise audio characteristic extraction nodes; processing nodes in the key frame processing branch comprise an image watermark identification node, a watermark removal node, a video deduplication node, an intelligent screenshot node and a video cover processing node, wherein the image watermark identification node, the video deduplication node and the intelligent screenshot node are parallel to each other, the watermark removal node is positioned behind the image watermark identification node, and the video cover processing node is positioned behind the intelligent screenshot node; the processing nodes of the video feature processing branch comprise video feature extraction nodes;

distributing the sub-content to a target processing node of a corresponding content processing branch in the target content processing path for processing according to the data format of the sub-content to obtain a content processing result corresponding to each sub-content;

and according to the original structure of the content to be released, fusing the content processing results corresponding to each sub-content to generate the processed content to be released.

2. The content processing method according to claim 1, wherein before the allocating the sub-content to the corresponding content processing branch in the content processing path for processing to obtain the content processing result corresponding to each sub-content, the method further comprises:

the allocating the sub-content to the corresponding content processing branch in the target content processing path for processing includes: and distributing the sub-content to the corresponding content processing branch in the adjusted content processing path for processing.

3. The content processing method of claim 1, wherein the content processing branch comprises at least one processing node; the allocating the sub-content to the corresponding content processing branch in the target content processing path for processing to obtain the content processing result corresponding to each sub-content includes:

distributing the sub-content to a corresponding content processing branch in the content processing path to obtain the sub-content corresponding to each content processing branch;

4. The content processing method according to claim 3, wherein obtaining the content processing result corresponding to the sub-content according to the processing result corresponding to the target processing node comprises:

when the processing indication mark indicates to execute a next processing node and the next processing node of the target processing node exists in the content processing branch, updating the target processing node to be the next processing node, and returning to execute the step of calling the target processing node to process the sub-content;

and when the processing indication mark indicates that the next processing node is executed and the next processing node of the target processing node does not exist in the content processing branch, obtaining a content processing result corresponding to the sub-content based on the processing results corresponding to all the processing nodes in the content processing branch.

5. The content processing method according to claim 4, wherein the content processing branch further comprises a connection node provided between a target processing node and a processing node next to the target processing node;

when the processing indication mark indicates to delete the content to be released, triggering a first connecting node between a target processing node and a processing node next to the target processing node to finish the processing of the content processing branch where the connecting node is located, and outputting the processing indication mark to obtain a content processing result corresponding to the sub-content.

6. The content processing method according to claim 4, wherein the content processing branch further comprises a connection node provided between a target processing node and a processing node next to the target processing node;

and updating the target processing node as the next processing node of the target processing node, and calling the target processing node to process the sub-content.

7. The content processing method of claim 4, wherein the processing nodes include a general processing layer, a base processing layer, and a logical processing layer;

calling the general processing layer to identify and divide the subcontent to obtain an identification unit;

8. The content processing method according to claim 3, wherein obtaining the content processing result corresponding to the sub-content according to the processing result corresponding to the target processing node comprises:

9. The content processing method of claim 8, wherein the processing nodes include a general processing layer, a base processing layer, and a logical processing layer;

10. The content processing method according to any one of claims 4 to 9, wherein the generating the processed content to be published according to the content processing result corresponding to each sub-content comprises:

11. A content processing apparatus, characterized by comprising:

the device comprises an acquisition unit, a processing unit and a processing unit, wherein the acquisition unit is used for acquiring content to be issued and content types of the content to be issued, and the content types to be issued comprise video formats;

a selecting unit, configured to select a target content processing path from at least two candidate content processing paths based on the content type, where the target content processing path is a process for processing the content to be published, the target content processing path includes at least two content processing branches, each content processing branch includes at least one processing node, and the at least two content processing branches include a text content processing branch, an audio feature processing branch, a key frame processing branch, and a video feature processing branch that are parallel to each other;

the content publishing device comprises an extracting unit and a publishing unit, wherein the extracting unit is used for extracting at least two categories of sub-content from content to be published, the categories are attributes of the sub-content about data formats, and the at least two categories of sub-content comprise text content, audio features, key frames and video features; the category of the sub-content corresponds to the type of the content processing branch;

the processing unit is used for determining a target processing node to be used for processing the subcontent currently according to the arrangement sequence of the processing nodes in the content processing branch; processing nodes in the text content processing branch line sequentially comprise a text extraction node, a text de-duplication node and a text title de-duplication node; the processing nodes in the audio characteristic processing branch comprise audio characteristic extraction nodes; processing nodes in the key frame processing branch comprise a picture watermark identification node, a watermark removal node, a video duplication removal node, an intelligent screenshot node and a video cover processing node, wherein the picture watermark identification node, the video duplication removal node and the intelligent screenshot node are parallel to each other, the watermark removal node is positioned behind the picture watermark identification node, and the video cover processing node is positioned behind the intelligent screenshot node; the processing nodes of the video feature processing branch comprise video feature extraction nodes; distributing the sub-content to a target processing node of a corresponding content processing branch in the target content processing path for processing according to the data format of the sub-content to obtain a content processing result corresponding to each sub-content;

and the fusion unit is used for fusing the content processing result corresponding to each sub-content according to the original structure of the content to be issued to generate the processed content to be issued.

12. A storage medium storing a plurality of instructions for storing a computer program thereon, characterized in that when the computer program is run on a computer, it causes the computer to execute the content processing method according to any one of claims 1 to 10.