WO2019061285A1

WO2019061285A1 - Video recording method and video recording system of intelligent terminal

Info

Publication number: WO2019061285A1
Application number: PCT/CN2017/104354
Authority: WO
Inventors: 金晓兰
Original assignee: 深圳传音通讯有限公司
Priority date: 2017-09-29
Filing date: 2017-09-29
Publication date: 2019-04-04

Abstract

Provided in the present invention are a video recording method and a video recording system of an intelligent terminal, comprising the following steps: starting up a video recording system of an intelligent terminal; capturing a target image comprising a target person; identifying face information of a person in the target image; matching the face information with preset standard face information; when the face information matches the standard person information, locking the target face in the target image according to the face information; anchoring the overall contour of the target person in the target image according to the target face information; and blurring images of people other than the target person in the target image. According to the present invention, the intelligent terminal may autonomously track the image or audio of a target person during the recording process, and thus eliminate the interference of the image or audio of a non-target person.

Description

Video recording method and video recording system of intelligent terminal

Technical field

The present invention relates to the field of video recording technology for intelligent terminals, and in particular, to a video recording method and a video recording system for an intelligent terminal.

Background technique

With the development and advancement of technology, the development of intelligent terminals is also changing with each passing day. Intelligent terminals Currently, intelligent terminals have gradually penetrated into all aspects of people's lives because of their convenience and functionality, and have become an indispensable auxiliary equipment in people's lives. At the same time, with the diversification of the functionality of smart terminals, the convenience of the functions of smart terminals has become one of the user's choice factors.

In particular, due to the development of smart terminal photography and video technology, smart terminals have gradually replaced other photography and video equipment, and have become a common photography or video recording device in people's lives. However, for non-professional intelligent terminal users, how to capture the image of the target person in time during the recording process and discharge the interference of the non-target person, using the photography or video technology of the existing smart terminal, is still difficult to achieve.

Summary of the invention

To solve the above problems, the present invention provides a video recording method and a video recording system for an intelligent terminal. The invention utilizes a face recognition technology and a face tracking technology to preset a face image and audio information of a target person in the smart terminal, so that the intelligent terminal autonomously tracks the image or audio of the target person during the recording process, and Discharges the image or audio of a non-target person.

Specifically, the present invention provides a method for recording a smart terminal, comprising the steps of: starting a recording system of the smart terminal, capturing a target image including a target person, and identifying face information of a person in the target image, Matching the face information with a preset standard face information, and when the face information matches the standard person information, locking the target face in the target image according to the face information, according to The target face information anchors an overall outline of the target person in the target image, and blurs a person image other than the target person in the target image.

Preferably, locking the overall outline of the target person in the target image according to the target face information The step of: capturing the target face in the target image, taking the body skin information in the target face as a reference, acquiring body information of the target person in the target image, and performing a preset character action The data table is matched, the action state of the target person is calculated, and the target person in the target image is anchored.

Preferably, the recording method further includes: after the target image is recorded, capturing an operation instruction, performing frame processing on the target image, performing super-resolution processing on a frame-by-frame basis, and performing each frame image in the target image. The blurred image in the picture is clear.

Preferably, the recording method further includes: reading audio information in the target image, matching with a preset standard audio information, extracting target audio information that matches the standard audio information, and deleting the target image. Audio information other than the target audio information.

Preferably, the step of extracting target audio information matching the standard audio information comprises: extracting an audio file in the target image, framing the audio file, and extracting an audio frame matching the standard audio information .

Another aspect of the present invention provides a recording system of an intelligent terminal, which includes the following module, a system control module, a recording system of the smart terminal, a camera module, and a target image including a target person, a face recognition module Identifying face information of a person in the target image, and a face verification module matching the face information with a preset standard face information, the face positioning module, when the face information and the face information When the standard person information is matched, the target face in the target image is locked according to the face information, and the human body positioning module anchors the entire target person in the target image according to the target face information. The contour, the first image processing module blurs the image of the person other than the target person in the target image.

Preferably, the human body positioning module includes a human body recognition unit that acquires human body information of the target person in the target image based on human body skin information in the target human face, and a human body positioning unit that will The information is matched with a preset character motion data table, the action state of the target person is calculated, and the target person in the target image is anchored.

Preferably, the recording system further includes: an instruction module, capturing an operation instruction, an image framing module, performing framing processing on the target image, and a second image processing module performing super-resolution processing on a frame-by-frame basis to The blurred image in each frame of the image is sharpened.

Preferably, the video recording system further includes: a first information matching module, reading audio information in the target image, matching with a preset standard audio information, and extracting, by the target audio extraction module, the standard audio information The matched target audio information, the audio filtering module, deletes audio information other than the target audio information in the target image.

Preferably, the target audio extraction module includes an audio framing unit that extracts audio in the target image a file, the audio file is framed, and the target audio extraction unit extracts an audio frame that matches the standard audio information.

Compared with the prior art, the technical advantage of the present invention is that the present invention provides a video recording method and a video recording system for an intelligent terminal, which combines face recognition technology and face tracking technology, and further utilizes the particularity of human skin. A preset character motion data table is used to calculate the overall contour of the target person, thereby realizing the technical effect of tracking the target person during the image recording process; at the same time, combined with the fuzzy calculation method in the image processing technology, the discharged recorded image is realized. The non-personal influence interference; and the invention is based on the recorded image file, and also provides a method for image file sharpness recovery and audio file filtering, and the user can implement the recorded image file based on the above technical means of the present invention. Further optimization processing.

DRAWINGS

1 is a schematic flow chart of a video recording method of a smart terminal according to a preferred embodiment of the present invention;

2 is a schematic flow chart of a method for clearing an image file according to a preferred embodiment of the present invention;

3 is a schematic flow chart of a method for filtering an audio file according to a preferred embodiment of the present invention;

4 is a structural diagram of a smart terminal recording system in accordance with a preferred embodiment of the present invention;

FIG. 5 is a structural diagram of a video recording system of a smart terminal according to another preferred embodiment of the present invention; FIG.

FIG. 6 is a structural diagram of a video recording method of a smart terminal according to another preferred embodiment of the present invention.

Detailed ways

The advantages of the present invention are explained in detail below with reference to the accompanying drawings and specific embodiments.

The following description refers to the same or similar elements in the different figures unless otherwise indicated. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present disclosure. Instead, they are merely examples of devices and methods consistent with aspects of the present disclosure as detailed in the appended claims.

First, it should be understood that although the terms first, second, third, etc. may be employed in the present disclosure to describe various information, such information should not be limited to these terms. These terms are only used to distinguish the same type of information from each other. For example, the first intelligent terminal may also be referred to as a second smart terminal without departing from the scope of the present disclosure. Similarly, the second smart terminal may also be referred to as a first smart terminal.

In the practical application of the video recording technology of the smart terminal, the user often suffers from the inability to capture the image or audio of the target person or other object in time, or the surrounding noise environment disturbs the effect of the image recording, especially for the active In children's video recording process, it is often difficult for parents to capture a clean and complete video of children. In order to solve the above problems, an aspect of the present invention provides a video recording method of an intelligent terminal.

Specifically, referring to FIG. 1 , which is a schematic flowchart of a video recording method of a smart terminal according to a preferred embodiment of the present invention. It can be seen from the figure that the recording method of the intelligent terminal provided in this embodiment mainly includes the following steps:

- starting the recording system of the intelligent terminal

In this embodiment, the smart terminal includes a system control module. When the user wants to record a video file, a startup command may be sent to the smart terminal by touching a shortcut icon of the smart terminal display interface or a text name in the shortcut menu. The system control module of the intelligent terminal reads an operation instruction issued by the user, and after the analysis, starts the recording system of the intelligent terminal according to the startup instruction.

- Capture a target image including the target person

In this embodiment, the smart terminal is provided with a camera module, and the camera module includes a front and/or a rear camera disposed on the smart terminal, and an image sensor and digital signal processing disposed inside the smart terminal. And so on. The camera module can capture an image containing the target person through the camera of the smart terminal, and generate an image of the smart terminal readable by the image sensor and the digital information number processor, etc., and cache the smart signal. In the terminal.

Identifying face information of a person in the target image

In this embodiment, the video recording system of the smart terminal includes a face recognition module, and the face recognition module can read the target in time sequence after the video recording process is completed or the video file is obtained to obtain an image file. Each frame of the image is imaged and the face information in each frame of the image is identified.

Preferably, the face recognition module can accurately calibrate the position and size of the face in the image. The pattern features contained in the face image are very rich, such as histogram features, color features, template features, structural features, and Haar features. In this embodiment, the face recognition module may identify the face information from the frame-by-frame picture.

Preferably, in the embodiment, the face recognition module uses the Adaboost algorithm to select some rectangular features (weak classifiers) that best represent the face, and constructs the weak classifier into a strong classifier according to the weighted voting method. Then, several strong classifiers obtained by training are connected in series to form a cascaded classifier of cascade structure, which effectively improves the detection speed of the classifier.

Preferably, in this embodiment, the face recognition module may further perform image preprocessing to reduce various conditions and random interference in the face recognition process. Among them, the preprocessing process mainly includes ray compensation, gradation transformation, histogram equalization, normalization, geometric correction, filtering and sharpening of the face image.

- matching the face information with a preset standard face information,

In this embodiment, the smart terminal further includes a face verification module, and the face verification module may extract the face information and perform search matching with a preset standard face information stored in the database. Preferably, in this embodiment, a threshold is set in advance, and when the similarity between the face information and the preset face information exceeds the threshold, the result of the matching is output; otherwise, no deal with.

Preferably, in this embodiment, the following algorithm can be used to achieve matching of facial information: a method based on geometric features; a method of local feature analysis (Local Face Analysis); a method of feature face (Eigenface or PCA); Model method; Neural Networks method; Hidden Markov Model; Gabor wavelet transform + pattern matching.

- when the face information matches the standard person information, the target face in the target image is locked according to the face information

In this embodiment, when the face information matches the standard face information, the face positioning module in the smart terminal receives the verification result sent by the face verification module, and then records according to the video. Time sequence, locking the target face in each frame of the target image; or sequentially reading the target face in each frame of the picture recorded by the camera module and locking.

- anchoring the overall outline of the target person in the target image according to the target face information

In this embodiment, the recording system of the smart terminal further includes a person positioning module, and the person positioning module can read the target face locked by the face positioning module, and according to the target face information An overall contour of the target person is anchored in the target image.

Preferably, in this embodiment, the specific step of anchoring the overall contour of the target person from the frame image of the target image includes:

- after locking the target face in the target image, acquiring body information of the target person in the target image based on the human skin information in the target face

Human skin is one of the distinguishing features of the human body. Like the information such as fingerprints and irises, human skin is unique. Different human skins have different textures, colors, and brightness. The same person's skin has a certain similarity. In this embodiment, the human body recognizing unit in the human body positioning module of the smart terminal can use the human skin information as a reference to identify the human body with the highest similarity to the skin feature of the target human face in the frame image of the target image. Other body parts such as hands, feet, arms, legs, etc., and obtain information such as the position, shape, and the like of other parts of the human body relative to the target face.

- matching with a preset character motion data table, calculating an action state of the target person, and anchoring an overall outline of the target person in the target image

In this embodiment, the human body recognizing unit further has a human body positioning unit, and the human body positioning unit can read information about the position, shape, and the like of the other parts of the human body relative to the target human face, and the The human body information is matched with a preset character motion data table, and the action state of the target person is calculated, thereby anchoring the target person in the frame image of the target image.

Preferably, the preset character motion data table is established by counting the positional relationship of the human body's face, hands, feet, arms, legs, etc. in various actions such as running, jumping, walking, and the like.

- Blurring a person image other than the target person in the target image.

With reference to FIG. 1 , in the embodiment, a first image processing module is further disposed in the smart terminal, and the image processing module can read an overall outline of the target person anchored by the human body positioning module, and Reversely selecting a frame image area that is not anchored, and using a blurring algorithm to blur an area other than the target person in the frame image in the target image, thereby blurring the image of the non-target person in the target image Chemical.

In summary, the present invention utilizes a combination of face recognition technology and face tracking technology, and further utilizes the particularity of the human skin to calculate the overall contour of the target person through a preset character motion data table, thereby realizing the image. The technical effect of tracking the target person during the recording process; at the same time, combined with the fuzzification calculation method in the image processing technology, the non-personal influence interference in the recorded image is discharged.

In addition, in order to prevent the user from being dissatisfied with the blurred image due to errors or the like in the target recognition and tracking process, the present invention also provides a method for clearing the blurred image file. 2 is a schematic flow chart of a method for clearing an image file according to a preferred embodiment of the present invention. It can be seen from the figure that, according to the recording method of the smart terminal in the above embodiment, the video recording method provided in this embodiment further includes:

- capturing an operation command when the target image is recorded

When the user is dissatisfied with the image file recorded by the above method, the sharpness in the recorded image can be restored, so that the object to be blurred and the like can be newly selected from the clear image file. Specifically, the smart terminal of the embodiment includes an instruction module, where the instruction module can capture a trigger operation of the display interface of the smart terminal, and parse the trigger pressure, location, and the like of the trigger operation, thereby generating a running instruction. Send out.

- framing the target image

In this embodiment, the smart terminal further includes an image framing module, and the image framing module can receive the running instruction, and according to the running instruction, read a cache and a target of a storage space of the smart terminal. And performing image processing on the target image to decompose the target image into at least one single frame image.

- performing super-resolution processing on a frame-by-frame basis to sharpen blurred images in each frame of the target image

Referring to FIG. 2, in the embodiment, a second image processing module is further disposed in the smart terminal, and the second image processing module may perform super-resolution processing on a frame-by-frame basis, and each frame image in the target image is The blurred image is clear. That is, the time resolution (acquiring a multi-frame image sequence of the same scene) is used in exchange for spatial resolution, and the conversion from temporal resolution to spatial resolution is realized. Preferably, the super-resolution method used in the embodiment includes: a super-resolution reconstruction method: a regularization reconstruction method; a non-uniform spatial sample interpolation method; an iterative back projection method (IBP); and a set theory reconstruction Method (convex set projection POCS); statistical reconstruction method (maximum a posteriori probability MAP and maximum likelihood estimation ML); hybrid ML/MAP/POCS method; adaptive filtering/Wiener filtering/Kalman filtering method; deterministic reconstruction method ; methods based on learning and pattern recognition, etc.

In addition, based on the recording method of the smart terminal provided in the above embodiment of the present invention, in order to further filter out the noise in the recorded image file, clear audio of the target person is proposed, and the present invention also provides a method for filtering the audio file. Referring to FIG. 3, it is a schematic flowchart of a method for filtering an audio file according to a preferred embodiment of the present invention. It can be seen from the figure that the filtering method of the audio file provided by this embodiment mainly includes the following steps:

- reading audio information in the target image to match a preset standard audio information

After the user obtains an image file by using the video recording method provided in the above embodiment of the present invention, audio information other than the target person in the image file may be further removed. Specifically, in the embodiment, the smart terminal further includes a first information matching module, where the first information matching module can read the audio information in the target image and associate it with a preset standard. The audio information is matched such that the target audio of the target person is marked in the audio information.

Extracting target audio information that matches the standard audio information

With reference to FIG. 3, it can be seen from the figure that in the embodiment, the smart terminal further has a target audio extraction module, and the target audio extraction module can read the marking information of the first information matching module. And extracting the labeled sub audio file in the audio file from the audio file of the target image.

Preferably, in this embodiment, the specific step of extracting the marked audio file includes:

Extracting an audio file in the target image, framing the audio file

In this embodiment, to analyze the audio stream, the audio stream needs to be framed, that is, the audio stream is cut into a short segment, and each segment is called a frame unit audio stream. Therefore, in the embodiment, the target audio extraction module is provided with an audio framing unit, and the audio framing unit can use the moving window function to implement the framing operation on the audio file. After framing, there is generally overlap between the unit audio stream and the unit audio stream. For example, the length of the unit audio stream per frame is 25 milliseconds, and there is 25-10=15 milliseconds between each two frame unit audio streams. The stack is called a frame length of 25 ms and a frame shift of 10 ms.

- extracting an audio frame that matches the standard audio information.

In this embodiment, the target audio extraction module further includes a target audio extraction unit, and the target audio extraction unit may read the tag information of the first information matching module, and extract the audio file from the target image. A sub-audio file that is marked in the audio file.

- deleting audio information other than the target audio information in the target image

In this embodiment, the smart terminal further has an audio filtering module, and the audio filtering module can also read the marking information of the first information matching module, and delete the audio file of the target image. A sub-audio file that is not marked in the audio file.

Preferably, in this embodiment, the marking information of the first information matching module is read by an audio adjustment module, and the marked audio file is enhanced, and the unmarked audio file portion is correspondingly weakened. , thereby achieving the technical effect of highlighting the audio of the target person in the image file.

In summary, the present invention is based on a recorded image file, and further provides a method for image file resolution recovery and audio file filtering. The user can further optimize the recorded image file based on the above technical means of the present invention. deal with.

In addition, in order to implement the recording method of the smart terminal, another aspect of the present invention provides a recording system of the smart terminal. Referring to FIG. 4, it is a structural diagram of a smart terminal recording system in accordance with a preferred embodiment of the present invention. As can be seen from the figure, the recording system of the intelligent terminal provided by this embodiment includes the following modules:

-System Control Module

A recording system for starting the smart terminal. Specifically, when the user wants to record a video file, a startup command may be sent to the smart terminal by touching a shortcut icon of the smart terminal display interface or a text name in the shortcut menu. The system control module of the smart terminal reads an operation instruction issued by the user, and starts the recording system of the intelligent terminal according to the startup instruction after parsing.

- Camera module

Used to capture a target image that includes the target person. The camera module includes a front and/or rear camera disposed on the smart terminal, and an image sensor, a digital signal processor, and the like disposed inside the smart terminal. The camera module can capture an image containing the target person through the camera of the smart terminal, and generate an image of the smart terminal readable by the image sensor and the digital information number processor, etc., and cache the smart signal. In the terminal.

-Face recognition module

A face information for identifying a person in the target image. The face recognition module can be in chronological order Reading each frame of the target image and identifying face information in each frame of the image.

- Face verification module

For matching the face information with a preset standard face information. The face verification module may extract the face information and perform search matching with a preset standard face information stored in a database. Preferably, in this embodiment, a threshold is set in advance, and when the similarity between the face information and the preset face information exceeds the threshold, the result of the matching is output; otherwise, no deal with.

-Face positioning module

When the face information matches the standard person information, the face positioning module locks the target face in the target image according to the face information. Specifically, after receiving the verification result sent by the face verification module, the face positioning module locks the target face in each frame of the target image according to the time sequence of the video recording. Or sequentially read the target face in each frame of the picture recorded by the camera module and lock it.

- Human positioning module

An anchoring for anchoring an overall outline of the target person in the target image according to the target face information. The person positioning module can read the target face locked by the face positioning module, and anchor the overall contour of the target person in the target image according to the target face information.

Wherein, preferably, the human body positioning module has the following units:

- human body recognition unit

And configured to acquire body information of the target person in the target image based on human skin information in the target face. The human body recognizing unit in the human body positioning module of the smart terminal can use the human skin information as a reference to identify a human hand, a foot, and an arm having the highest similarity to the skin feature of the target human face in the frame image of the target image. And other body parts such as legs, and obtain information such as the position and shape of the other parts of the human body relative to the target face.

- Human positioning unit

And configured to match the human body information with a preset character motion data table, calculate an action state of the target person, and anchor a target person in the target image. The human body positioning unit can read information about the position, shape, and the like of the other parts of the human body relative to the target human face, and match the human body information with a preset character motion data table to calculate the target. The action state of the character, thereby anchoring the target person in the frame picture of the target image.

- first image processing module

And configured to perform blur processing on a character image other than the target person in the target image. Specifically, the The image processing module can read the overall contour of the target person anchored by the human body positioning module, and select a frame image region that is not anchored in the reverse direction, and adopt a blur algorithm to frame the frame image in the target image. The area other than the target person is blurred, thereby blurring the image of the non-target person in the target image.

In addition, in order to prevent the user from being dissatisfied with the blurred image due to errors or the like in the target recognition and tracking process, the present invention also provides a system capable of clearing the blurred image file. Referring to FIG. 5, it is a structural diagram of a video recording system of a smart terminal according to another preferred embodiment of the present invention. As can be seen from the figure, the video recording system provided in this embodiment further includes:

- instruction module

For capturing an operation instruction issued by the user, the instruction module may capture a trigger operation of the display interface of the smart terminal, and parse the trigger pressure, location, and the like of the trigger operation, thereby generating a running instruction to send out.

-Image framing module

And used to perform framing processing on the target image. The image framing module can receive the running instruction, and according to the running instruction, read a target image buffered with a storage space of the smart terminal, and perform framing processing on the target image to The image is decomposed frame by frame into at least one single frame picture.

- second image processing module

For super-resolution processing on a frame-by-frame basis, the blurred image in each frame of the target image is sharpened. The second image processing module may perform super-resolution processing on a frame-by-frame basis to sharpen a blurred image in each frame of the target image. That is, the time resolution (acquiring a multi-frame image sequence of the same scene) is used in exchange for spatial resolution, and the conversion from temporal resolution to spatial resolution is realized.

In addition, based on the recording system of the smart terminal provided in the above embodiment of the present invention, in order to further filter out the noise in the recorded image file, clear audio of the target person is proposed, and the present invention also provides a filtering system for the audio file. Referring to FIG. 6, which is a structural diagram of a video recording method of a smart terminal according to another preferred embodiment of the present invention. As can be seen from the figure, the video recording system provided in this embodiment further includes the following modules:

- first information matching module

For reading audio information in the target image, and matching with a preset standard audio information. The first information matching module may read the audio information in the target image and match it with a preset standard audio information, thereby marking the target audio of the target person in the audio information.

- Target audio extraction module

For extracting target audio information that matches the standard audio information. The target audio extraction module is readable Taking the tag information of the first information matching module, and extracting the labeled sub audio file in the audio file from the audio file of the target image.

Specifically, the audio extraction module includes the following units:

-Audio framing unit

And used to extract an audio file in the target image, and framing the audio file. Specifically, an audio framing unit is disposed in the target audio extraction module, and the audio framing unit may use a moving window function to implement a framing operation on the audio file.

- target audio extraction unit

For extracting an audio frame that matches the standard audio information. Specifically, the target audio extraction unit may read the tag information of the first information matching module, and extract the labeled sub audio file in the audio file from the audio file of the target image.

-audio filter module

And for deleting audio information other than the target audio information in the target image. The audio filtering module may also read the tag information of the first information matching module, and delete the unlabeled sub audio file in the audio file from the audio file of the target image.

In addition, in this embodiment, an audio adjustment module may be disposed in the smart terminal, and the marking information of the first information matching module is read by the audio adjustment module, and is marked. The audio file is enhanced and the unmarked portion of the audio file is weakened accordingly, thereby achieving the technical effect of highlighting the audio of the target person in the image file.

In summary, the present invention provides a video recording system for a smart terminal, which combines face recognition technology and face tracking technology, and further utilizes the particularity of human skin to calculate a target person through a preset character motion data table. The overall contour, thus, the technical effect of tracking the target person during the image recording process; at the same time, combined with the fuzzy calculation method in the image processing technology, the non-personal influence interference in the recorded image is discharged.

It should be noted that the embodiments of the present invention are preferred embodiments, and are not intended to limit the scope of the present invention. Any one skilled in the art may use the above-disclosed technical contents to change or modify the equivalent embodiments. Any modification or equivalent changes and modifications of the above embodiments in accordance with the technical spirit of the present invention are still within the scope of the technical solutions of the present invention.

Claims

A video recording method for an intelligent terminal, comprising the steps of:

Starting the recording system of the smart terminal,

Capture a target image that includes the target person,

Identifying face information of a person in the target image,

Matching the face information with a preset standard face information,

When the face information matches the standard person information, the target face in the target image is locked according to the face information,

Anchoring an overall outline of the target person in the target image according to the target face information,

Blur processing of a person image other than the target person in the target image.
A video recording method according to claim 1, wherein

The step of locking the overall outline of the target person in the target image according to the target face information includes,

After locking the target face in the target image,

Obtaining body information of the target person in the target image based on human skin information in the target face,

Matching with a preset character motion data table, calculating an action state of the target person, and anchoring a target person in the target image.
A video recording method according to claim 1, wherein

Also includes,

After the target image is recorded,

Capture an operation command,

Performing frame processing on the target image,

The super-resolution processing is performed frame by frame, and the blurred image in each frame of the target image is sharpened.
A video recording method according to claim 1, wherein

Also includes,

Reading audio information in the target image to match a preset standard audio information,

Extracting target audio information that matches the standard audio information,

Deleting audio information other than the target audio information in the target image.
A video recording method according to claim 4, wherein

The step of extracting target audio information that matches the standard audio information includes,

Extracting an audio file in the target image, and framing the audio file,

An audio frame that matches the standard audio information is extracted.
A video recording system for an intelligent terminal, comprising the following modules,

a system control module that starts a recording system of the smart terminal,

a camera module that captures a target image including a target person,

a face recognition module that recognizes face information of a person in the target image,

a face verification module that matches the face information with a preset standard face information,

a face positioning module, when the face information matches the standard person information, locking the target face in the target image according to the face information,

a human body positioning module that anchors an overall contour of the target person in the target image according to the target face information,

The first image processing module blurs the image of the person other than the target person in the target image.
A video recording system according to claim 6 wherein:

The human body positioning module includes

a human body recognition unit that acquires human body information of the target person in the target image based on human skin information in the target human face,

The human body positioning unit matches the human body information with a preset character motion data table, calculates an action state of the target person, and anchors a target person in the target image.
A video recording system according to claim 6 wherein:

Also includes,

An instruction module that captures an operation instruction,

The image framing module performs framing processing on the target image,

The second image processing module performs super-resolution processing on a frame-by-frame basis to sharpen the blurred image in each frame of the target image.
A video recording system according to claim 6 wherein:

Also includes,

The first information matching module reads the audio information in the target image and matches a preset standard audio information.

a target audio extraction module that extracts target audio information that matches the standard audio information,

The audio filtering module deletes audio information other than the target audio information in the target image.
The video recording system of claim 9 wherein:

The target audio extraction module includes

An audio framing unit, extracting an audio file in the target image, and framing the audio file,

The target audio extracting unit extracts an audio frame that matches the standard audio information.