WO2021169501A1

WO2021169501A1 - Living body video picture processing method and apparatus, computer device, and storage medium

Info

Publication number: WO2021169501A1
Application number: PCT/CN2020/136434
Authority: WO
Inventors: 张国辉; 马爱国
Original assignee: 平安科技（深圳）有限公司
Priority date: 2020-02-26
Filing date: 2020-12-15
Publication date: 2021-09-02
Also published as: CN111428568A; CN111428568B

Abstract

A living body video picture processing method and apparatus, a computer device, and a storage medium, relating to living body detection of a biometric identification technology in the technical field of artificial intelligence. The method comprises: receiving a detection request of a living body video and storing the living body video into a memory (S302); obtaining an original living body video data packet, and calculating the number of pictures in the original living body video data packet (S304); performing frame extraction on the pictures in the original living body video data packet according to the number of the pictures in the original living body video data packet, so as to obtain a target picture (S306); decoding the target picture to obtain the decoded target picture (S308); and feeding back the decoded target picture to an underlying interface (S310). By directly operating in the memory, the intermediate processing links are reduced, the living body video arrangement processing performance is improved, the processing speeds of living body video detection and decoding are increased, and the performance loss of a GPU is reduced.

Description

Living body video picture processing method, device, computer equipment and storage medium

This application claims the priority of a Chinese patent application filed with the Chinese Patent Office on February 26, 2020, the application number is 202010119603.9, and the invention title is "live video image processing methods, devices, computer equipment, and storage media". The entire content is approved The reference is incorporated in this application.

Technical field

This application relates to the field of biometrics technology, and in particular to methods, devices and readable storage media for processing video images in living body detection.

Background technique

As a mature biometric technology, portrait recognition has been widely used in finance, education, public security, social services, e-commerce and other fields. However, human faces can easily be deceptively recognized after being copied from videos or photos, which makes security problems. Video live detection is currently the main solution to this security problem.

In the prior art video liveness detection, the video needs to be decoded into pictures and sounds, while the liveness detection only needs to detect the decoded pictures, and does not detect the decoded sounds. When detecting live body confidence in pictures, generally not all video pictures are tested, but only about 3~6 seconds of video. Calculated at a video bit rate of 25 frames per second, the total number of video frames for video live body detection It is about 100 frames. Generally, it takes about 20-40 milliseconds to decode and format a picture using the CPU, and about 2000-4000 milliseconds for a 100-frame picture, which takes a long time and wastes operating system resources. The inventor realizes that if the video is downloaded to a local disk for processing, it will inevitably increase the amount of IO access between the disk and the memory and reduce the processing speed.

In addition, the inventor realizes that conventional video live detection decoders mostly process video in file formats, and cannot directly detect video data in the memory. In addition, part of the live body detection videos come from mobile terminal devices such as mobile phones and tablet computers, and the shooting angles are different. The video live body detection decoding needs to rotate the image, and also consumes CPU resources and processing time. The above situations will lead to a reduction in the efficiency of video live detection and a poor user experience.

Therefore, it is necessary to improve the existing technology in order to obtain a better user experience.

technical problem

Based on this, it is necessary to provide a method, device, and readable storage medium for processing video pictures in living body detection to improve the efficiency of existing video living body detection in response to the existing problems.

A method for processing live video pictures, the method comprising:

Receiving the detection request of the live video and saving the live video to the memory;

Acquiring an original live video data packet, and calculating the number of pictures in the original live video data packet;

Extracting frames from pictures in the original live video data packet according to the number of pictures in the original live video data packet to obtain a target picture;

Decode the target picture to obtain a decoded target picture;

Feed back the decoded target picture to the bottom layer interface.

A living body video picture processing device, the living body video picture processing device includes:

The receiving unit is configured to receive the detection request of the live video and save the live video to the memory;

The original live video data packet calculation unit is used to calculate the number of pictures in the original live video data packet;

The picture frame extraction unit is configured to extract frames from the pictures in the original live video data packet according to the number of pictures calculated by the original live video data packet calculation unit to obtain a target picture;

The picture decoding unit is configured to decode and restore the target picture extracted by the picture frame extraction unit to obtain the decoded target picture, and feed back the decoded target picture to the underlying interface.

A computer device includes a memory and a processor. The memory stores computer-readable instructions, and when the computer-readable instructions are executed by the processor, the processor executes the following steps:

Decode the target picture to obtain a decoded target picture;

Feed back the decoded target picture to the bottom layer interface.

A storage medium storing computer-readable instructions. When the computer-readable instructions are executed by one or more processors, the one or more processors execute the following steps:

Decode the target picture to obtain a decoded target picture;

Feed back the decoded target picture to the bottom layer interface.

The above-mentioned live video picture processing method, device, computer equipment and storage medium receive live video detection requests and save the live video to the memory; calculate the number of pictures contained in the live video to be detected, and remove the live video The frame pictures at the beginning and the end of the picture sequence are extracted at equal distances to obtain the target picture, and then the target picture is detected according to the reference frame in the video compression technology to ensure that the obtained target picture contains the complete information in the original live video. Finally, decode the target picture after frame decimation detection to finally obtain the decoded target picture and output it to the underlying interface. By directly performing picture frame extraction and decoding operations in the memory, the operation of downloading live videos to the disk is avoided, the IO (Input and Output) operations of the disk are reduced, and the operation directly in the memory reduces the number of intermediate operations. The processing link improves the processing performance of live video sorting, thereby increasing the processing speed of live video detection and decoding video, and saves GPU performance loss. In addition, before outputting the decoded target picture to the bottom interface, the decoded target picture can also be processed with the angle information of the picture, and the picture with the abnormal view angle can be rotated to obtain the picture with the normal view and output to the bottom interface, which further reduces The processing pressure of the underlying SDK (Soft Development Kit) improves the accuracy of live video detection.

Technical solutions

Type a paragraph describing the technical solution here.

Beneficial effect

Type a paragraph describing the beneficial effect here.

Description of the drawings

FIG. 1 is an implementation environment diagram of a method for processing live video images provided in an embodiment;

Figure 2 is a block diagram of the internal structure of a computer device in an embodiment;

Fig. 3 is a flowchart of a method for processing live video images in an embodiment;

FIG. 4 is a flowchart of a method for processing live video images in another embodiment;

FIG. 5 is a flowchart of decimating a picture to obtain a target picture according to the number of pictures in the original live video data packet in an embodiment;

Fig. 6 is a flowchart of decoding a target picture to obtain a decoded target picture in an embodiment;

FIG. 7 is a structural block diagram of a living body video image processing device in an embodiment;

FIG. 8 is a structural block diagram of a living body video image processing device in another embodiment;

FIG. 9 is a structural block diagram of a picture frame extraction unit in an embodiment;

Fig. 10 is a structural block diagram of a picture decoding unit in an embodiment.

The best mode of the present invention

Type here a paragraph describing the best mode of the present invention.

Embodiments of the present invention

In order to make the purpose, technical solutions, and advantages of this application clearer and clearer, the following further describes the application in detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described here are only used to explain the present application, and are not used to limit the present application.

FIG. 1 is an implementation environment diagram of a method for processing live video images provided in an embodiment. As shown in FIG. 1, the implementation environment includes a computer device 110 and a terminal 120.

The computer device 110 is a living body video image processing device, for example, a computer device used by a tester, and a living body video image processing tool is installed on the computer device 110. The terminal 120 is installed with an application that needs to process live video pictures. When the live video picture needs to be processed, the tester can send a live video picture processing request at the terminal 120, and the live video picture processing request carries a live video picture processing identifier. , The computer device 110 receives the live video image processing request, obtains the test script corresponding to the live video image processing identifier in the computer device 110 according to the live video image processing identifier, and then executes the test script using the live video image processing tool to perform a test on the terminal 120 The application is tested, and the live video image processing result corresponding to the test script is obtained.

It should be noted that the terminal 120 and the computer device 110 may be a smart phone, a tablet computer, a notebook computer, a desktop computer, a server, etc., but are not limited thereto. The computer equipment 110 and the terminal 120 can be connected via Bluetooth, USB (Universal Serial Bus, Universal Serial Bus) or other communication connection methods for connection, this application is not limited here.

Figure 2 is a schematic diagram of the internal structure of a computer device in an embodiment. As shown in Figure 2, the computer device includes a processor, a non-volatile storage medium, a memory, and a network interface connected through a system bus. Among them, the non-volatile storage medium of the computer device stores an operating system, a database, and a computer program. The database can store control information sequences. When the computer program is executed by the processor, the processor can realize a live video image. Approach. The processor of the computer equipment is used to provide calculation and control capabilities, and supports the operation of the entire computer equipment. Computer readable instructions may be stored in the memory of the computer device, and when the computer readable instructions are executed by the processor, the processor can make the processor execute a method for processing live video images. The network interface of the computer device is used to connect and communicate with the terminal. Those skilled in the art can understand that the structure shown in FIG. 2 is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may Including more or fewer parts than shown in the figure, or combining some parts, or having a different arrangement of parts.

As shown in FIG. 3, in an embodiment, a method is proposed, which can be applied to the above-mentioned computer device 110, and specifically may include the following steps S302 to S310:

Step S302, receiving the detection request of the live video and saving the live video to the memory;

In this embodiment, the detection request for receiving the live video is generally completed by a computer device. In this embodiment, a background server is used to complete. Of course, the use of a background server here is not limited to the server. As mentioned above Other computer equipment can also be undertaken. In the video in vivo detection technology, the background server is responsible for the video decoding operation work, and the video in vivo detector is set on the back-end server. After the video in vivo detector receives the detection request of the in vivo video, the video in vivo detector will obtain the original living body Video data packet, save the original live video data packet to the memory. Saving the original live video data packet to the memory for operation can improve the processing speed of the original live video data, because the speed of direct manipulation of the data in the memory is much higher than that of downloading the original live video data to the disk for operation. It is to improve the processing speed of the original live video data.

In some embodiments, the original live video data packet may also be saved in a non-volatile storage medium for processing.

Step S304: Obtain the original live video data packet, and calculate the number of pictures in the original live video data packet;

In the embodiment of the present application, after the video decoder obtains the original live video data packet, it first needs to calculate the number of frame pictures included in the original live video data packet. In the video live detection technology, the amount of video data that needs to be detected is about 3 to 6 seconds, which can be determined according to different video bit rates. For example, a video of about 3 to 6 seconds contains 90 frame pictures. In this embodiment, the frame pictures included in the video are referred to as pictures for short, and the video decoder obtains 90 pictures in the original live video data packet.

Step S306: decimating the pictures in the original live video data packet according to the number of pictures in the original live video data packet to obtain a target picture;

In the method of this application, frame extraction of pictures in the original live video data packet is the key point. Frame extraction is to extract the key frame pictures in the original live video data packet to reduce the number of pictures processed by the video decoder and make the video decoder The necessary pictures are processed, and the non-essential pictures are not processed, thereby saving system resources and speeding up processing.

FIG. 5 shows a flowchart of extracting frames to obtain a target picture according to the number of pictures in the original live video data packet in an embodiment, which may specifically include the following steps S502 to S506:

Step S502, obtaining and numbering the number of pictures in the original live video data packet;

After the video decoder obtains the number of pictures in the original live video data packet, it also needs to sequentially number the pictures in the original live video data packet. For example, the original live video data packet obtained by the video decoder contains 90 pictures. Frame pictures, that is, the 90 pictures are numbered sequentially from 0 to 89.

Step S504: Delete the preset number of pictures numbered at the beginning and end;

For the pictures in the original live video data package, there must be key frames and non-key frames. In the field of existing video compression technology, each frame represents a still image. In the actual compression process, various algorithms will be adopted. To reduce the capacity of data, IPB is the most common method. I frame is also called intra picture. I frame represents key frame and belongs to intra-frame compression. Only the picture data of this frame is needed for decoding. P frame represents this frame and the previous key frame (P frame) When decoding, the previously buffered picture needs to be superimposed on the difference defined in this frame to generate the final picture. The B frame is a two-way difference frame, that is, the B frame records the difference between the current frame and the previous and next frames. To decode the B frame, the previous cached picture must be obtained, and the picture after decoding must be superimposed to generate the final picture. The B frame has a high compression rate and a heavy burden on the CPU during decoding. Therefore, this application uses a key frame I frame and a non-key frame P frame for calculation.

In this embodiment, the pictures preset at the beginning and the end of the picture group in the original live video data packet need to be deleted. Specifically, for 90 original live video data packets, 15 pictures at the beginning and the end need to be deleted. Here, the beginning and the end Each of the 15 deleted pictures is to prevent the original live video data received by the video decoder from being fake pictures. The preset number of deleted pictures is also worthy of experience based on experience. Those skilled in the art can know that the original live video data received The frame picture in the video data packet is adjusted appropriately. For the original live video data packet of 90 pictures, after deleting the first and last 15 pictures, 60 pictures are left.

Step S506: Perform equidistant extraction of the remaining pictures from the preset number of pictures deleted from the beginning and the end to obtain the target picture.

After deleting the preset number of pictures at the beginning and end of the original live video data packet, it is necessary to perform an equidistant picture extraction operation on the remaining pictures. The equidistant extraction operation is also to prevent picture fraud, for example, the above 90 pictures After deleting the preset pictures at the beginning and the end of the original live video data packet, there are 60 frames of pictures left, and the equidistant extraction of the 60 frames of pictures equidistant n=(90-15×2)÷(6+1)= 8. According to the distance n=8, it is calculated that 6 pictures need to be extracted, that is, the calculation starts from the 15th frame of the picture group in the original live video data packet, and the 24th frame (15+8) and the 32nd frame ( 24+8), the 40th frame (32+8), the 48th frame (40+8), the 56th frame (48+8), the 64th frame (56+8) and other 6 frames of pictures, the original living body is extracted at an equal distance After the picture in the video data packet, the target picture is obtained.

In some embodiments, the target picture obtained by extracting frames is detected according to the key frame I frame and the non-key frame P frame in the above video compression technology to ensure that the extracted target picture is correct and can accurately represent the original live video. , It can avoid malicious attacks with a high probability, and improve the detection performance and response effect.

Step S308, decode the target picture to obtain a decoded target picture;

In this embodiment, after the appropriate target picture is obtained, the picture can be decoded according to the video compression technology of the original live video data packet to obtain the decoded target picture.

FIG. 6 shows a flowchart of decoding a target picture to obtain a decoded target picture in an embodiment, which may specifically include the following steps S602 to 606:

Step S602: Obtain a key frame I frame and a non-key frame P frame in the original live video data packet;

Step S604: Perform frame extraction detection on the target picture according to the information of the key frame I frame and the non-key frame P frame to ensure that the target picture contains complete picture information in the original live video;

Step S606: Decode the target picture after frame extraction detection to obtain a decoded target picture.

Step S310: Feed back the decoded target picture to the bottom layer interface.

In this embodiment, the video decoder outputs the finally obtained decoded target picture in the memory to the underlying interface, for example, the underlying SDK (Software Development Kit) performs liveness detection to complete the liveness video detection from the SDK ask.

FIG. 4 shows that in another embodiment, a method for processing live video images is proposed, which can be applied to the above-mentioned computer device 110, and is a further optimized technical solution based on the above-mentioned live video image processing method. Specifically, it can include the following steps:

Step S402, receiving the detection request of the live video and saving the live video to the memory;

Step S404: Obtain the original live video data packet, and calculate the number of pictures in the original live video data packet;

Step S406, extract frames from the pictures in the original live video data packet according to the number of pictures in the original live video data packet to obtain a target picture;

FIG. 5 shows a flowchart of extracting frames from pictures to obtain a target picture according to the number of pictures in the original live video data packet in an embodiment, which may specifically include the following steps S502 to S506:

Step S408, decode the target picture to obtain a decoded target picture;

Step S410: Obtain picture angle information of the decoded target picture, and perform a picture rotation operation according to the picture angle information to adjust the decoded target picture to a picture with a normal angle;

In this embodiment, for the original live video, some of the original live video comes from a mobile terminal, such as a mobile phone. When the rotation angle of the mobile phone when shooting the video is different, the shooting angle is also different, that is, the video is an abnormal viewing angle. At this time, it is necessary to further rotate the decoded target picture, that is, transpose the decoded target picture by 90 degrees or 270 degrees, or mirror 180 degrees, rotate it to the normal viewing angle, and then feed it back to the underlying interface. Of course, the rotation operation is not arbitrary rotation, and needs to be performed based on the angle information of the original live video. Specifically, operations such as matrix inversion or mirroring can be used to obtain the decoded target picture of the normal view. For example, the picture is a two-dimensional array, the RGB format picture of width W and height H can be expressed as array[W][H], the element value of the array contains three values (R, G, B), and the RGB format image is one 2 For a three-dimensional matrix, the matrix inversion is to invert this array[W][H] matrix to the matrix of array[H][W], and perform the same matrix inversion operation on the 3-layer image.

Step S412: Feed back the decoded target picture to the bottom layer interface.

As shown in FIG. 7, in one embodiment, a living body video image processing apparatus is provided. The living body video image processing apparatus may be integrated into the above-mentioned computer device 110, and may specifically include a receiving unit 702 and an original living body video data packet. The calculation unit 704, the picture frame extraction unit 706, and the picture decoding unit 708.

The receiving unit 702 is configured to receive a detection request of a live video and save the live video to the memory;

The original live video data packet calculation unit 704 is configured to calculate the number of pictures in the original live video data packet;

The picture frame extraction unit 706 is configured to extract frames from the pictures in the original live video data packet according to the number of pictures calculated by the original live video data packet calculation unit to obtain a target picture;

The picture decoding unit 708 is configured to decode and restore the target picture extracted by the picture frame extraction unit to obtain the decoded target picture, and feed back the decoded target picture to the underlying interface.

As shown in FIG. 9, in one embodiment, the picture frame extraction unit 706 further includes a picture quantity numbering module 706A and a picture frame extraction processing module 706B.

A picture quantity numbering module 706A, configured to obtain the number of pictures in the original live video data packet, and number the pictures in the original live video data packet;

The picture frame extraction processing module 706B is used to delete a preset number of pictures with picture numbers at the beginning and the end of the original live video data packet numbered by the picture number numbering module, and extract pictures at equal distances from the remaining pictures to obtain the target picture.

As shown in FIG. 10, in one embodiment, the picture decoding unit 708 further includes a reference frame acquisition module 708A, a picture frame detection module 708B, and a picture decoding module 708C.

The reference frame obtaining module 708A is configured to obtain the key frame I frame and the non-key frame P frame in the original live video data packet;

The picture frame extraction detection module 708B is configured to perform frame extraction detection on the target picture according to the information of the key frame I frame and the non-key frame P frame acquired by the reference frame acquisition module;

The picture decoding module 708C is configured to decode and restore the target picture output by the picture frame detection module, and feed back the decoded target picture to the underlying interface.

FIG. 8 shows a live video image processing device proposed in another embodiment, which can be integrated into the above-mentioned computer equipment 110, and is a further optimized technical solution based on the above live video image processing device. Specifically, it may include a receiving unit 802, an original live video data packet calculation unit 804, a picture frame extraction unit 806, a picture angle processing unit 808, and a picture decoding unit 810.

The receiving unit 802 is configured to receive the detection request of the live video and save the live video to the memory;

The original live video data packet calculation unit 804 is configured to calculate the number of pictures in the original live video data packet;

The picture frame extraction unit 806 is configured to extract frames from the pictures in the original live video data packet according to the number of pictures calculated by the original live video data packet calculation unit to obtain a target picture;

The picture angle processing unit 808 is configured to obtain picture angle information of the decoded target picture output by the picture decoding unit, and perform a picture rotation operation according to the picture angle information to adjust the decoded target picture to a picture with a normal angle;

The picture decoding unit 810 is configured to decode and restore the target picture extracted by the picture frame extraction unit to obtain the decoded target picture, and feed back the decoded target picture to the underlying interface.

In the embodiment of the living video picture processing device shown in FIG. 8, the picture frame extraction unit 806 is the same as the above-mentioned living video picture processing device, and also includes a picture number numbering module and a picture frame extraction processing module. The picture frame extraction unit 706 of is the same, and it is not shown in the figure. Please refer to FIG. 9.

Wherein, the picture quantity numbering module is used to obtain the number of pictures in the original live video data packet, and number the pictures in the original live video data packet;

A picture frame extraction processing module for deleting a preset number of pictures with picture numbers at the beginning and end of the original live video data packet numbered by the picture number numbering module, and extracting pictures at equal distances from the remaining pictures to obtain a target picture .

In the same way, the picture decoding unit 810 is the same as the above-mentioned living video picture processing device, and also includes a reference frame acquisition module, a picture frame detection module, and a picture decoding module. Since it is the same as the picture decoding unit 708 shown in FIG. For drawing schematic, please refer to Figure 10.

Wherein, the reference frame obtaining module is used to obtain the key frame I frame and the non-key frame P frame in the original live video data packet;

The picture frame extraction detection module is configured to perform frame extraction detection on the target picture according to the information of the key frame I frame and the non-key frame P frame acquired by the reference frame acquisition module;

The picture decoding module is configured to decode and restore the target picture output by the picture frame detection module, and feed back the decoded target picture to the bottom layer interface.

In one embodiment, a computer device is proposed. The computer device includes a memory, a processor, and a computer program that is stored on the memory and can run on the processor, and the processor executes the computer The program implements the following steps: receiving a live video detection request and saving the live video to the memory; obtaining an original live video data packet, and calculating the number of pictures in the original live video data packet; according to the original live video data The number of pictures in the packet extracts frames from the pictures in the original live video data packet to obtain a target picture; decodes the target picture to obtain a decoded target picture; and feeds back the decoded target picture to an underlying interface.

In one embodiment, when the processor executes the computer-readable instructions, the following steps are also performed: said extracting frames from the pictures in the original live video data packet according to the number of pictures in the original live video data packet to obtain The steps for the target image include:

Acquiring and numbering the number of pictures in the original live video data packet;

Delete the preset number of pictures numbered at the beginning and end;

Perform equidistant extraction of the remaining pictures from the preset number of pictures deleted at the beginning and end to obtain the target picture.

In an embodiment, the step of decoding the target picture to obtain a decoded target picture includes:

Acquiring a key frame I frame and a non-key frame P frame in the original live video data packet;

Performing frame extraction detection on the target picture according to the information of the key frame I frame and the non-key frame P frame to ensure that the target picture contains complete picture information in the original live video;

Decoding the target picture after frame decimation detection to obtain a decoded target picture.

In another embodiment, a computer device is also provided. The computer device includes a memory, a processor, and a computer program that is stored on the memory and can run on the processor, and the processor executes all The computer program implements the following steps: receiving a live video detection request and saving the live video to the memory; acquiring an original live video data packet, and calculating the number of pictures in the original live video data packet; according to the original live video The number of pictures in the video data packet extracts frames from the pictures in the original live video data packet to obtain a target picture; decodes the target picture to obtain a decoded target picture; obtains picture angle information of the decoded target picture, Perform a picture rotation operation according to the picture angle information to adjust the decoded target picture to a picture with a normal angle; and feed back the decoded target picture to an underlying interface.

Delete the preset number of pictures numbered at the beginning and end;

In one embodiment, a storage medium storing computer-readable instructions is provided. When the computer-readable instructions are executed by one or more processors, the one or more processors perform the following steps: Detect the request and save the live video to the memory; obtain the original live video data packet, and calculate the number of pictures in the original live video data packet; compare the original live video data packet according to the number of pictures in the original live video data packet The picture in the video data packet is framed to obtain a target picture; the target picture is decoded to obtain a decoded target picture; the decoded target picture is fed back to the bottom layer interface.

Delete the preset number of pictures numbered at the beginning and end;

In another embodiment, a storage medium storing computer-readable instructions is also provided. When the computer-readable instructions are executed by one or more processors, the one or more processors perform the following steps: Video detection request and save the live video to the memory; obtain the original live video data packet, and calculate the number of pictures in the original live video data packet; compare the number of pictures in the original live video data packet The picture in the original live video data packet is framed to obtain the target picture; the target picture is decoded to obtain the decoded target picture; the picture angle information of the decoded target picture is obtained, and the picture rotation operation is performed according to the picture angle information Adjusting the decoded target picture to a picture with a normal angle; feeding back the decoded target picture to the bottom layer interface.

Delete the preset number of pictures numbered at the beginning and end;

A person of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be implemented by instructing relevant hardware through a computer program. The computer program can be stored in a computer readable storage medium. When executed, it may include the procedures of the above-mentioned method embodiments. Wherein, the computer-readable storage medium may be non-volatile or volatile. The aforementioned storage media can be magnetic disks, optical disks, read-only storage memory (Read-Only Memory, ROM) and other non-volatile storage media, or random storage memory (Random Access Memory, RAM) etc.

The technical features of the above-mentioned embodiments can be combined arbitrarily. In order to make the description concise, all possible combinations of the various technical features in the above-mentioned embodiments are not described. However, as long as there is no contradiction in the combination of these technical features, All should be considered as the scope of this specification.

The above-mentioned embodiments only express several implementation manners of the present application, and their description is relatively specific and detailed, but they should not be understood as a limitation to the patent scope of the present application. It should be pointed out that for those of ordinary skill in the art, without departing from the concept of this application, several modifications and improvements can be made, and these all fall within the protection scope of this application. Therefore, the scope of protection of the patent of this application shall be subject to the appended claims.

Industrial applicability

Type a paragraph describing industrial applicability here.

Sequence Listing Free Content

Type here the free content description paragraph of the sequence listing.

Claims

A method for processing live video pictures, wherein the method includes:

Receiving the detection request of the live video and saving the live video to the memory;

Acquiring an original live video data packet, and calculating the number of pictures in the original live video data packet;

Extracting frames from pictures in the original live video data packet according to the number of pictures in the original live video data packet to obtain a target picture;

Decode the target picture to obtain a decoded target picture;

Feed back the decoded target picture to the bottom layer interface.
3. The living video picture processing method of claim 1, wherein after the step of decoding the target picture to obtain a decoded target picture, the method further comprises:

Acquire the picture angle information of the decoding target picture, and perform a picture rotation operation according to the picture angle information to adjust the decoding target picture to a picture with a normal angle.
3. The live video picture processing method of claim 1, wherein the step of extracting frames from the pictures in the original live video data packet according to the number of pictures in the original live video data packet to obtain a target picture include:

Acquiring and numbering the number of pictures in the original live video data packet;

Delete the preset number of pictures numbered at the beginning and end;

Perform equidistant extraction of the remaining pictures from the preset number of pictures deleted at the beginning and end to obtain the target picture.
3. The living video picture processing method of claim 1, wherein the step of decoding the target picture to obtain a decoding target picture comprises:

Acquiring a key frame I frame and a non-key frame P frame in the original live video data packet;

Performing frame extraction detection on the target picture according to the information of the key frame I frame and the non-key frame P frame to ensure that the target picture contains complete picture information in the original live video;

Decoding the target picture after frame decimation detection to obtain a decoded target picture.
3. The living video picture processing method of claim 1, wherein the step of decoding the target picture after frame extraction detection to obtain a decoded target picture comprises:

The target picture is decoded according to the video compression method of the original live video data packet to obtain the decoded target picture.
3. The live video picture processing method of claim 2, wherein the step of performing a picture rotation operation according to the picture angle information to adjust the decoding target picture to a picture with a normal angle comprises:

Adopt matrix inversion or mirror image processing to adjust the decoding target picture to a picture with a normal angle.
8. The method for processing live video pictures according to claim 1, wherein the step of obtaining the original live video data packet comprises:

Determine the detection quantity of video data according to the video bit rate.
A living body video picture processing device, wherein the living body video picture processing device includes:

The receiving unit is configured to receive the detection request of the live video and save the live video to the memory;

The original live video data packet calculation unit is used to calculate the number of pictures in the original live video data packet;

The picture frame extraction unit is configured to extract frames from the pictures in the original live video data packet according to the number of pictures calculated by the original live video data packet calculation unit to obtain a target picture;

The picture decoding unit is configured to decode and restore the target picture extracted by the picture frame extraction unit to obtain the decoded target picture, and feed back the decoded target picture to the underlying interface.
A computer device includes a memory and a processor. The memory stores computer-readable instructions, and when the computer-readable instructions are executed by the processor, the processor executes the following steps:

Receiving the detection request of the live video and saving the live video to the memory;

Acquiring an original live video data packet, and calculating the number of pictures in the original live video data packet;

Extracting frames from pictures in the original live video data packet according to the number of pictures in the original live video data packet to obtain a target picture;

Decode the target picture to obtain a decoded target picture;

Feed back the decoded target picture to the bottom layer interface.
The electronic device according to claim 9, wherein after the step of decoding the target picture to obtain a decoded target picture, the method further comprises:

Acquire the picture angle information of the decoding target picture, and perform a picture rotation operation according to the picture angle information to adjust the decoding target picture to a picture with a normal angle.
9. The electronic device according to claim 9, wherein the step of extracting frames from pictures in the original live video data packet according to the number of pictures in the original live video data packet to obtain a target picture comprises:

Acquiring and numbering the number of pictures in the original live video data packet;

Delete the preset number of pictures numbered at the beginning and end;

Perform equidistant extraction of the remaining pictures from the preset number of pictures deleted at the beginning and end to obtain the target picture.
The electronic device according to claim 9, wherein the step of decoding the target picture to obtain a decoded target picture comprises:

Acquiring a key frame I frame and a non-key frame P frame in the original live video data packet;

Performing frame extraction detection on the target picture according to the information of the key frame I frame and the non-key frame P frame to ensure that the target picture contains complete picture information in the original live video;

Decoding the target picture after frame decimation detection to obtain a decoded target picture.
9. The electronic device according to claim 9, wherein the step of decoding the target picture after frame extraction detection to obtain a decoded target picture comprises:

The target picture is decoded according to the video compression method of the original live video data packet to obtain the decoded target picture.
The electronic device according to claim 10, wherein the step of performing a picture rotation operation according to the picture angle information to adjust the decoding target picture to a picture with a normal angle comprises:

Adopt matrix inversion or mirror image processing to adjust the decoding target picture to a picture with a normal angle.
9. The electronic device according to claim 9, wherein before the step of obtaining the original live video data packet comprises: determining the detection quantity of the video data according to the video bit rate.
A storage medium storing computer-readable instructions. When the computer-readable instructions are executed by one or more processors, the one or more processors execute the following steps:

Receiving the detection request of the live video and saving the live video to the memory;

Acquiring an original live video data packet, and calculating the number of pictures in the original live video data packet;

Extracting frames from pictures in the original live video data packet according to the number of pictures in the original live video data packet to obtain a target picture;

Decode the target picture to obtain a decoded target picture;

Feed back the decoded target picture to the bottom layer interface.
The computer-readable storage medium according to claim 16, wherein after the step of decoding the target picture to obtain a decoded target picture, the method further comprises:

Acquire the picture angle information of the decoding target picture, and perform a picture rotation operation according to the picture angle information to adjust the decoding target picture to a picture with a normal angle.
16. The computer-readable storage medium according to claim 16, wherein the step of extracting frames from pictures in the original live video data packet according to the number of pictures in the original live video data packet to obtain a target picture include:

Acquiring and numbering the number of pictures in the original live video data packet;

Delete the preset number of pictures numbered at the beginning and end;

Perform equidistant extraction of the remaining pictures from the preset number of pictures deleted at the beginning and end to obtain the target picture.
The computer-readable storage medium according to claim 16, wherein the step of decoding the target picture to obtain a decoded target picture comprises:

Acquiring a key frame I frame and a non-key frame P frame in the original live video data packet;

Performing frame extraction detection on the target picture according to the information of the key frame I frame and the non-key frame P frame to ensure that the target picture contains complete picture information in the original live video;

Decoding the target picture after frame decimation detection to obtain a decoded target picture.
16. The computer-readable storage medium according to claim 16, wherein the step of decoding the target picture after frame extraction detection to obtain the decoded target picture comprises: a video compression method based on the original live video data packet The target picture is decoded to obtain the decoded target picture.