CN110418150B

CN110418150B - Information prompting method, equipment, system and computer readable storage medium

Info

Publication number: CN110418150B
Application number: CN201910639977.0A
Authority: CN
Inventors: 钟宜峰; 张健; 莫东松; 张进; 赵璐; 马丹; 马晓琳; 王科
Original assignee: MIGU Culture Technology Co Ltd
Current assignee: MIGU Culture Technology Co Ltd
Priority date: 2019-07-16
Filing date: 2019-07-16
Publication date: 2022-07-01
Anticipated expiration: 2039-07-16
Also published as: CN110418150A

Abstract

The invention discloses an information prompting method, equipment, a system and a computer readable storage medium, relates to the technical field of live broadcast, and aims to solve the problem that information cannot be timely prompted to a user in the existing live broadcast scene. The method comprises the following steps: collecting a live broadcast image of a live broadcast scene; encoding the live broadcast image to obtain a live broadcast stream; under the condition that a target event occurs in the live broadcast scene, adding indication information of the target event in the live broadcast stream to obtain a target live broadcast stream; and sending the target live stream to a client, or sending the target live stream to a server. The embodiment of the invention can improve the timeliness of information prompt in a live broadcast scene.

Description

Information prompting method, equipment, system and computer readable storage medium

Technical Field

The present invention relates to the field of live broadcast technologies, and in particular, to an information prompting method, device, system, and computer-readable storage medium.

Background

With the popularization of intelligent terminals, more and more users like watching live programs such as sports games through the terminals. In the live broadcasting process, the live broadcasting platform plays a live broadcasting signal. Taking the football game as an example, the football game is long in time, and most of the time is non-wonderful time. Then, during this period, the user may be distracted or otherwise treated, resulting in missing the game's highlight.

According to the analysis, the user cannot know the content information of the live program in time by using the mode of the prior art.

Disclosure of Invention

The embodiment of the invention provides an information prompting method, equipment and a system and a computer readable storage medium, which aim to solve the problem that information cannot be timely prompted to a user in the existing live broadcast scene.

In a first aspect, an embodiment of the present invention provides an information prompting method, applied to an image acquisition end, including:

collecting a live broadcast image of a live broadcast scene;

encoding the live broadcast image to obtain a live broadcast stream;

under the condition that a target event occurs in the live broadcast scene, adding indication information of the target event in the live broadcast stream to obtain a target live broadcast stream;

and sending the target live stream to a client, or sending the target live stream to a server, so that the server sends the target live stream to the client, and the client prompts the target event according to the indication information when performing live broadcasting according to the target live stream.

Wherein the target event comprises: a score change event; adding the indication information of the target event in the live stream to obtain a target live stream, wherein the step of adding the indication information of the target event in the live stream comprises the following steps:

collecting images of the score cards in the live broadcast scene;

analyzing the image of the score card;

determining that the score change event has occurred in a case where it is determined that the score has changed according to the analysis result;

adding a first identifier at a first preset position of the live stream to obtain the target live stream; the first identifier is used for indicating that the score change event occurs.

Wherein the target event comprises: a penalty event; adding the indication information of the target event in the live stream to obtain a target live stream, wherein the step of adding the indication information of the target event in the live stream comprises the following steps:

acquiring a panoramic image of the live broadcast scene;

detecting a position of a game ball in the panoramic image;

determining that the penalty ball event occurred if it is determined that the position of the game ball has not changed within a predetermined time;

adding a second identifier at a first preset position of the live stream to obtain the target live stream; the second identifier is used for indicating that the penalty event occurs.

Wherein the target event comprises: penalty events; adding the indication information of the target event in the live stream to obtain a target live stream, wherein the step of adding the indication information of the target event in the live stream comprises the following steps:

collecting images of referees in the live scenes;

detecting whether an image of a penalty tool appears in the image of the judge;

determining that the penalty event has occurred in a case where the image of the penalty appliance is determined to be present;

adding a third identifier at a first preset position of the live stream to obtain the target live stream; the third indication is for indicating that the penalty event occurred.

Wherein the target event comprises: the appearance of a target character; adding the indication information of the target event in the live stream to obtain a target live stream, wherein the method comprises the following steps:

carrying out face recognition on the live broadcast image;

comparing the recognized face image with a preset face image;

determining that the target person appears under the condition that the comparison result meets a preset condition;

and adding a fourth identifier at a first preset position of the live stream to obtain the target live stream, wherein the fourth identifier is used for indicating that the target person appears.

Wherein, after determining that the target person appears in the case that the comparison result satisfies a predetermined condition, the method further comprises:

determining the normalized coordinates of the central point of the human face outline in the human body image frame where the target person is located;

determining a first parameter according to the normalized coordinates;

and adding the first parameter at a second preset position of the live stream, wherein the first parameter represents the relative position of the center point of the face contour in a live picture.

In a second aspect, an embodiment of the present invention provides an information prompting method, applied to a server, including:

receiving a target live broadcast stream sent by an image acquisition end, wherein the target live broadcast stream comprises indication information of a target event;

and sending the target live stream to a client so that the client prompts the target event according to the indication information when the client carries out live broadcasting according to the target live stream.

Wherein, in a case that the target event does not include the occurrence of a target person, before the transmitting the target live stream to the client, the method further comprises:

decoding the target live broadcast stream to obtain a live broadcast image;

carrying out face recognition on the live broadcast image;

determining whether a target person appears according to the result of face recognition;

and in the case that the target person is determined to appear, adding a fourth identification to the first preset position of the live stream, wherein the fourth identification is used for indicating that the target person appears.

Wherein after determining whether a target person is present according to the result of the face recognition, the method further comprises:

determining a first parameter according to the normalized coordinates;

In a third aspect, an embodiment of the present invention provides an information prompting method, applied to a client, including:

receiving a target live broadcast stream sent by an image acquisition end or a server end, wherein the target live broadcast stream comprises indication information of a target event;

and prompting the target event according to the indication information when the live broadcast is carried out according to the target live broadcast stream.

Wherein, the prompting the target event according to the indication information comprises:

acquiring prompt selection information of a user;

under the condition that the prompt selection information is matched with the indication information, acquiring a prompt mode pre-selected by a user;

and prompting the target event by utilizing the prompting mode.

Wherein, when the indication information indicates that a target person appears and the prompt mode includes identifying the target person, the prompting the target event by using the prompt mode includes:

determining a face area of the target person;

and identifying the face contour of the target person in the face area.

Wherein the determining the face area of the target person comprises:

acquiring a first parameter included in a second preset position of the target live stream;

acquiring a normalization coordinate of the central point of the face contour in a human body image frame where the target person is located according to the first parameter;

acquiring a second parameter according to the normalized coordinate;

dividing the human body image frame into at least one region, wherein each region is provided with a corresponding matrix;

and traversing the matrix, and if the element value corresponding to the second parameter is the target value in the target matrix, determining that the target area corresponding to the target matrix is a human face area.

Wherein, in the face area, identifying the face contour of the target person includes:

determining pixel points corresponding to the face contour in the face region;

and highlighting the target pixel point.

In a fourth aspect, an embodiment of the present invention provides an information prompting apparatus, including: a transceiver, a memory, a processor, and a computer program stored on the memory and executable on the processor;

the processor configured to read a program in the memory to implement the steps in the method according to the first aspect; or implementing a step in a method according to the second aspect; or implementing steps in a method as described in the third aspect.

In a fifth aspect, an embodiment of the present invention provides a computer-readable storage medium for storing a computer program, which when executed by a processor implements the steps in the method according to the first aspect; or implementing a step in a method according to the second aspect; or implementing steps in a method as described in the third aspect.

In a sixth aspect, an embodiment of the present invention provides an information prompting system, including: the system comprises an image acquisition end and at least one client end;

the image acquisition terminal is used for acquiring a live broadcast image of a live broadcast scene, encoding the live broadcast image and acquiring a live broadcast stream; under the condition that a target event occurs in the live broadcast scene, adding indication information of the target event in the live broadcast stream to obtain a target live broadcast stream, and sending the target live broadcast stream to the client;

and the client is used for receiving the target live stream sent by the image acquisition end and prompting the target event according to the indication information when the target live stream is live broadcasted.

In a seventh aspect, an embodiment of the present invention provides an information prompting system, including: the system comprises an image acquisition end, a server end and at least one client end;

the image acquisition terminal is used for acquiring a live broadcast image of a live broadcast scene, encoding the live broadcast image and acquiring a live broadcast stream; under the condition that a target event occurs in the live broadcast scene, adding indication information of the target event in the live broadcast stream to obtain a target live broadcast stream, and sending the target live broadcast stream to the server;

the server is used for sending the target live stream to the client;

and the client is used for receiving the target live stream sent by the server and prompting the target event according to the indication information when the target live stream is subjected to live broadcasting.

In the embodiment of the invention, the target event in the live broadcast scene is identified by using the image acquisition end. Therefore, after receiving the live stream sent by the server, the client can prompt the target event according to the indication information when performing live broadcasting according to the target live stream. Therefore, by using the scheme of the embodiment of the invention, the user can know the events occurring in the live broadcast scene in time, thereby improving the timeliness of information prompt.

Drawings

In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings needed to be used in the description of the embodiments of the present invention 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 that other drawings can be obtained according to these drawings without inventive exercise.

FIGS. 1(a) and 1(b) are schematic diagrams of an information presentation system according to an embodiment of the present invention;

fig. 2 is a flowchart of an information prompting method according to an embodiment of the present invention;

fig. 3(a) and fig. 3(b) are schematic diagrams of an image capturing end in the system provided by the embodiment of the invention, respectively;

fig. 4 is a second flowchart of an information prompting method according to an embodiment of the present invention;

FIG. 5 is a diagram illustrating a server in the system according to an embodiment of the present invention;

fig. 6 is a third flowchart of an information prompting method according to an embodiment of the present invention;

FIG. 7 is a diagram showing one of the structures of an information presentation apparatus according to an embodiment of the present invention;

fig. 8 is one of the structural diagrams of a second processing module in the information presentation apparatus according to the embodiment of the present invention;

fig. 9 is a second structural diagram of a second processing module in the information presentation apparatus according to the embodiment of the present invention;

fig. 10 is a third structural diagram of a second processing module in the information prompt apparatus according to the embodiment of the present invention;

fig. 11 is a second structural diagram of an information presentation device according to an embodiment of the present invention;

FIG. 12 is a third block diagram of an information presentation device according to an embodiment of the present invention;

FIG. 13 is a fourth block diagram of an information presentation apparatus according to an embodiment of the present invention;

fig. 14 is a structural diagram of a prompt module in the information prompt apparatus according to the embodiment of the present invention;

fig. 15 is a structural diagram of a prompt sub-module in the information prompt apparatus according to the embodiment of the present invention;

FIG. 16 is a block diagram of an information presentation device according to an embodiment of the present invention;

FIG. 17 is a second block diagram of an information presentation device according to an embodiment of the present invention;

fig. 18 is a third structural diagram of an information presentation apparatus according to an embodiment of the present invention.

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 some, not all, embodiments of the present invention. All other embodiments, which can be obtained by a person skilled in the art without inventive step based on the embodiments of the present invention, are within the scope of protection of the present invention.

Referring to fig. 1(a), fig. 1(a) is a schematic diagram of an information prompting system according to an embodiment of the present invention. As shown in fig. 1(a), the system includes: the system comprises an image acquisition terminal 101, a service terminal 102 and at least one client (such as a mobile phone and the like) 103. The image acquisition terminal 101 is configured to acquire a live broadcast image of a live broadcast scene, encode the live broadcast image, and acquire a live broadcast stream; under the condition that a target event occurs in the live broadcast scene, adding indication information of the target event in the live broadcast stream to obtain a target live broadcast stream, and sending the target live broadcast stream to the server; the server 102 is configured to send the target live stream to the client; the client 103 is configured to receive a target live stream sent by the server, and prompt the target event according to the indication information when performing live broadcast according to the target live stream.

Live broadcast signals are acquired and encoded by equipment of an image acquisition end and then pushed to an RTMP (Real Time Messaging Protocol) stream to a server end. The server performs necessary processing (e.g., broadcast control and transcoding) on the video stream, and transmits the live stream to the client through a Content Delivery Network (CDN). The client decodes and plays the direct broadcast stream by using the player.

It should be noted that another information prompting system according to the embodiment of the present invention may not include the server shown in fig. 1 (a). Then, in this case, the data of the image capturing end will be directly sent to the client. As shown in fig. 1(b), the information presentation system includes: an image acquisition end 104 and at least one client 105. The image acquisition terminal 104 is configured to acquire a live broadcast image of a live broadcast scene, encode the live broadcast image, and acquire a live broadcast stream; under the condition that a target event occurs in the live broadcast scene, adding indication information of the target event in the live broadcast stream to obtain a target live broadcast stream, and sending the target live broadcast stream to the client; the client 105 is configured to receive a target live stream sent by the image acquisition end, and prompt the target event according to the indication information when the target live stream is live broadcast.

In the system shown in fig. 1(a) or fig. 1(b), the image capturing end, the server end, and the client end may be respectively used to perform corresponding processes performed by the image capturing end, the server end, and the client end in the following method embodiments. It should be noted that, due to differences in system composition, functions implemented by the image acquisition end, the server end, and the client end are slightly different in different system architectures. Hereinafter, specific implementation processes of the embodiments of the present invention under the above system architecture are described in detail with reference to different embodiments.

Referring to fig. 2, fig. 2 is a flowchart of an information prompting method provided by an embodiment of the present invention, and is applied to an image acquisition end, as shown in fig. 2, including the following steps:

step 201, collecting live broadcast images of live broadcast scenes.

In the embodiment of the invention, live broadcast images, such as a camera, can be acquired by utilizing image acquisition equipment in a live broadcast site.

Step 202, encoding the live broadcast image to obtain a live broadcast stream.

In this step, the live image may be encoded by any encoding method to form a live stream.

And 203, adding the indication information of the target event into the live broadcast stream to obtain the target live broadcast stream under the condition that the target event is determined to occur in the live broadcast scene.

In the embodiment of the invention, in order to enable a user to know information in a live scene in time, a plurality of auxiliary image acquisition devices can be deployed in a live scene for acquiring event information. For example, an image pickup apparatus for picking up score plate information, an image pickup apparatus for picking up referee images, an image pickup apparatus for picking up game balls, and the like may be separately disposed.

The target events include, but are not limited to: score change events, penalty ball events, penalty events, presence of a target character, and the like. Wherein, the score change event refers to that the score of the score board is changed; the penalty event refers to a penalty ball, such as a corner ball, an arbitrary ball, etc., according to the judgment of the referee; the penalty event refers to the judgment of the game by a judge by using a penalty tool, such as a red event and a yellow event; the occurrence of the target task refers to the occurrence of a preset character, such as a football star, in the live broadcast picture. Through the detection to different events and the corresponding instruction information of setting up, the variety of suggestion can be richened to more be favorable to the timely information of understanding in the live scene of user.

Different target events may be determined in different ways. As shown in fig. 3(a), different modules may be provided at the image capturing end to determine whether a target event occurs. For example, a score change detection module may be provided for detecting whether a score change event has occurred. For example, a penalty ball event detection module may be provided for detecting whether a penalty ball event has occurred. In practical applications, taking a football game as an example, the penalty event detection module may be a corner kick detection module. For example, a penalty event detection module may be provided for detecting whether a penalty event occurs. In practical applications, taking a football game as an example, the penalty event detection module may be a red and yellow card detection module. For example, a target person detection module may be provided for detecting whether a target person is present.

If the target event comprises a score change event, the step is specifically to collect images of score cards in the live broadcast scene and analyze the images of the score cards. And determining that the score change event occurs in a case where it is determined from the analysis result that the score has changed. Then, adding a first identifier at a first preset position of the live stream to obtain the target live stream; the first identifier is used for indicating that the score change event occurs. In this embodiment of the present invention, the first predetermined location refers to a location where a CSID (Chunk Stream ID) is located in header information of the RTMP protocol block. Wherein, the CSID supports the user-defined numerical range of [3,65599 ].

As shown in fig. 3(a) or 3(b), the score change detection means may include a score change image pickup apparatus, an encoding apparatus, and the like. The encoding apparatus may include an OCR (Optical Character Recognition) Recognition algorithm sub-module, an encoding service sub-module, and the like. The camera device captures image information (1 frame per second) of a live broadcast score board in real time and transmits the image information to the encoding device. And the OCR algorithm sub-module identifies the score information by using a Tesseract-OCR algorithm. When the algorithm identifies a score change, the encoding service submodule identifies the event "score change" by setting the CSID to 101 in the header information of the RTMP protocol block.

If the target event comprises a penalty ball event, the step is specifically to collect a panoramic image of the live broadcast scene and detect the position of the game ball in the panoramic image. In a case where it is determined that the position of the game ball has not changed within the predetermined time, it is determined that the penalty ball event has occurred. Then, adding a second identifier at a first preset position of the live stream to obtain the target live stream; the second identifier is used for indicating that the penalty ball event occurs.

In the embodiment of the invention, different conditions of the penalty event can be identified by using different identifiers, and events such as the penalty event can be identified uniformly. If different ones of the penalty ball events are identified, then in the event of a first penalty ball event, adding an identifier to a first predetermined location of the live stream to obtain the target live stream, the identifier indicating that the first penalty ball event occurred; and in the case that a second penalty ball event occurs, adding a mark to a first preset position of the live stream to obtain the target live stream, wherein the mark is used for indicating that the second penalty ball event occurs. The above labels used to indicate different penalty events differ. Wherein, the first penalty ball event and the second penalty ball event can be arbitrarily set according to different live scenes. For example, in the case of a soccer game, the first penalty event may be a kicking event, the second penalty event may be a corner event, and the two different penalty events are indicated by different indicia.

As shown in fig. 3(a) or 3(b), the penalty ball event detection module may include a penalty ball event camera, a coding device, and the like. The encoding apparatus may include an object detection algorithm sub-module, an encoding service sub-module, and the like. The camera device captures full-field image information (1 frame per second) in real time and transmits the image information to the encoding device. The target detection algorithm sub-module detects position coordinates of a game ball (e.g., a soccer ball) in an image using an SSD (Single Shot multi box Detector) target detection algorithm. Taking a football game as an example, if the position of a football is not changed for more than 3 seconds continuously, the coding service sub-module identifies the event of 'free kick' by setting the CSID to 102 in the header information of the RTMP protocol block.

If the target event comprises a penalty event, the step is specifically to collect images of the referees in the live broadcast scene and detect whether images of penalty appliances appear in the images of the referees. In a case where the image of the penalty appliance is determined to appear, it is determined that the penalty event has occurred. Then, adding a third identifier at a first preset position of the live stream to obtain the target live stream; the third indication is for indicating that the penalty event occurred.

In the embodiment of the invention, different conditions of the penalty events can be identified by different identifiers, and the penalty events can be identified uniformly. If different conditions in the penalty events are identified, adding an identifier at a first preset position of the live stream to obtain the target live stream when a first penalty event occurs, wherein the identifier is used for indicating that the first penalty event occurs; and in the case of occurrence of a second penalty event, adding an identifier to a first predetermined position of the live stream to obtain the target live stream, wherein the identifier is used for indicating that the second penalty event occurs. The above labels for indicating different penalty events differ. The first penalty event and the second penalty event can be set arbitrarily according to different live scenes. For example, in the case of a football game, the first penalty event may be a red event, the second penalty event may be a yellow event, and the two different penalty events are indicated by different indicia.

As shown in fig. 3(a) or fig. 3(b), the penalty event detection module may include a penalty event imaging apparatus, an encoding apparatus, and the like. The encoding apparatus may include an object detection algorithm sub-module, an encoding service sub-module, and the like. The image pickup apparatus captures referee image information (1 frame per second) in real time and transmits the image information to the encoding apparatus. Taking a football game as an example, the target detection algorithm sub-module detects the red/yellow cards in the image by using the SSD target detection algorithm. If a red/yellow card is detected in the image, the encode services sub-module identifies two events, "red" and "yellow" in the header information of the RTMP protocol block by setting the CSID to 103 and 104, respectively.

If the target event comprises the appearance of a target person, the step is specifically to perform face recognition on the live broadcast image and compare the recognized face image with a preset face image. And determining that the target person appears under the condition that the comparison result meets a preset condition. Then, adding a fourth identifier at a first preset position of the live stream to obtain the target live stream, wherein the fourth identifier is used for indicating that the target person appears.

In order to enable the client to accurately prompt the target person, normalized coordinates of the central point of the face contour in the human body image frame where the target person is located in the human body image frame where the target person appears can be determined, and then the first parameter is determined according to the normalized coordinates. And then, adding the first parameter at a second preset position of the live stream, wherein the first parameter represents the relative position of the center point of the face contour in a live picture. By the method, the client can quickly and accurately determine the face contour and prompt.

Wherein the second predetermined location may also be a CSID in header information of the RTMP protocol block, but is different from the first predetermined location. The first predetermined location and the second predetermined location may be adjacent CSIDs in the RTMP protocol block, and may also have a gap.

The absolute coordinates of the face frame detected by the face detection algorithm at the upper left corner of the live broadcast picture are assumed as follows: (x)₀，y₀) The absolute coordinates of the lower right corner are: (x)₁，y₁) The live broadcast picture size is: w h, w represents the length of the live broadcast picture, h represents the width of the live broadcast picture, and w and h are positive integers. Here, the following operations are performed:

w_p∈[0，100]，h_p∈[0，100]。

wherein (w)_p，h_p) And representing the coordinates of the center point of the normalized face frame.

Such as w_pOr h_pIs an integer of one bit, then is respectively at w_pOr h_pAnd 0 is supplemented before. Then, w is added_pAnd h_pThe concatenation is a four-bit integer. The first bit of the four-bit integer is complemented by 1 to obtain a five-bit integer, and the CSID field is set to the five-bit integer. The five-bit integer is the first parameter.

As shown in fig. 3(a), the target person detection module may include a coding device and the like. The coding device can comprise a face detection algorithm sub-module, a face recognition algorithm sub-module, a coding service sub-module and the like. And transmitting the image information of the camera for acquiring the live broadcast image to the encoding equipment. The face detection algorithm submodule and the face recognition algorithm submodule need to perform face detection and face recognition on each frame of image. Specifically, each frame of image data is input into a face detection algorithm sub-module to obtain a face frame. For example, the face detection algorithm may be an MTCNN (Multi-task convolutional neural network) face detection algorithm. And intercepting the image information in the face frame and inputting the image information into a face recognition algorithm submodule. For example, the face recognition algorithm may be facenet algorithm or the like. And when the Euclidean distance between the detected face features and the star face library features is lower than a set threshold value, the star appears in the frame image. The encode services sub-module sets the CSID in the header information of the RTMP protocol block to 111 (assuming 111 represents the star above) to identify the event "target person X is present".

It should be noted that the above-mentioned CSID values can be arbitrarily set, and the above is only an example.

And 204, sending the target live stream to a client, or sending the target live stream to a server, so that the server sends the target live stream to the client, and the client prompts the target event according to the indication information when performing live broadcasting according to the target live stream.

As can be seen from the above, in the embodiment of the present invention, the target event in the live scene is identified by using the image capturing end. Therefore, after receiving the live stream sent by the server, the client can prompt the target event according to the indication information when performing live broadcasting according to the target live stream. Therefore, by using the scheme of the embodiment of the invention, the user can know the events occurring in the live broadcast scene in time, thereby improving the timeliness of information prompt.

In practical application, in order to save computing resources and improve processing speed, some modules with large computing amount are not arranged at an image acquisition end. For example, the target person detection module may not be disposed at the image capturing end, and may be disposed at a server end, for example. At this time, the structure of the image capturing end may be as shown in fig. 3 (b).

Referring to fig. 4, fig. 4 is a flowchart of an information prompting method provided by the embodiment of the present invention, and is applied to a server, as shown in fig. 4, including the following steps:

step 401, receiving a target live broadcast stream sent by an image acquisition end, wherein the target live broadcast stream includes indication information of a target event.

The meaning of the target event may refer to the description of the foregoing method embodiment, and the specific indication manner of the indication information may also refer to the description of the foregoing embodiment.

Step 402, sending the target live stream to a client, so that the client prompts the target event according to the indication information when performing live broadcast according to the target live stream.

And at the server, under the condition that the target event does not comprise the occurrence of the target person, whether the target person occurs or not can be detected, so that the reminding of the event in the live scene is enriched. Specifically, the server decodes the target live broadcast stream, acquires a live broadcast image, and performs face recognition on the live broadcast image. And determining whether the target person appears according to the face recognition result. Then, in a case where it is determined that the target person appears, a fourth identifier indicating that the target person appears is added to the first predetermined position of the live stream.

Wherein the first location comprises the CSID in the header information of the RTMP protocol block. The fourth identification is, for example, 111 (assuming 111 represents the star mentioned above) to identify the event "target person X appears". Alternatively, other arbitrarily set values may also be used.

Also, in this embodiment, in order to enable the client to accurately prompt the target person, normalized coordinates of the center point of the face contour in the human image frame where the target person is located may be determined in the human image frame where the target person appears, and then the first parameter may be determined according to the normalized coordinates. And then, adding the first parameter at a second preset position of the live stream, wherein the first parameter represents the relative position of the center point of the face contour in a live picture.

The manner of determining the first parameter may refer to the description of the foregoing embodiments. The second predetermined location may also be the CSID in the header information of the RTMP protocol block, but is different from the first predetermined location described above. The first predetermined location and the second predetermined location may be adjacent CSIDs in the RTMP protocol block, and may also have a gap.

In conjunction with fig. 5, the server may include: a decoding device/decoding service module, a face detection module, a face recognition module, a coding device/coding service module, etc.

The server receives the RTMP data stream, and the RTMP data stream is decoded by the decoding equipment/decoding service module to obtain each frame of image data. And carrying out face detection and face recognition on each frame of image. And inputting each frame of image data into a face detection module to obtain a face frame. For example, the face detection module may perform face detection using the MTCNN face detection algorithm. And intercepting image information in the face frame and inputting the image information into the face recognition module. The face recognition module may perform face recognition using facenet algorithm or the like. And when the Euclidean distance between the detected face features and the star face library features is lower than a set threshold value, the star appears in the frame image. The encoding device/encoding service module sets the CSID in the header information of the RTMP protocol block to 111 (assuming 111 represents the above star) to identify the event "target person X appears".

Referring to fig. 6, fig. 6 is a flowchart of an information prompting method provided in the embodiment of the present invention, which is applied to a client, and as shown in fig. 6, the method includes the following steps:

step 601, receiving a target live broadcast stream sent by an image acquisition end or a server end, wherein the target live broadcast stream comprises indication information of a target event.

The meaning of the target event can refer to the description of the foregoing method embodiment, and the specific indication manner of the indication information can also refer to the description of the foregoing embodiment.

Step 602, when the live broadcast is performed according to the target live broadcast stream, prompting the target event according to the indication information.

In this step, in order to make the prompted information more conform to the requirements of the user, the prompted selection information of the user can be acquired. And acquiring a prompting mode pre-selected by a user under the condition that the prompting selection information is matched with the indication information. And then, prompting the target event by utilizing the prompting mode.

In practical application, a user can set an event reminder before the live broadcast starts and during the live broadcast. At the client, all events supporting the reminding service of the system can be prompted to the user, including the above-mentioned events of 'score change', 'free sphere of corner ball', 'red and yellow cards', 'star X appears', 'star Y appears', and the like. The user can then set which events to remind when they occur by checking. Accordingly, the client can obtain prompt selection information of the user.

Meanwhile, a prompt mode can be set according to the input of the user. The prompting mode comprises any one or more of the following modes:

marking the concerned star with a red contour line; playing a prompt tone; triggering the mobile phone to vibrate; the method comprises the following steps of (1) flashing words related to event types for three times at a specific position of a screen; intermittently increasing the brightness of the picture within a fixed time (such as within 5 seconds) (such as highlighting every 1 second for 1 second), and realizing the flickering effect of the whole picture; shaking a player window; other actions with a reminder function, etc.

And the client player receives and decodes the live stream. And reading the CSID field and prompt selection information set by a user. For example, if the CSID is 101 and the user has selected to remind a "score change" event, a "score change" event reminder is triggered. The reminding mode can comprise one or more of the combination of the above modes, or the user is reminded by using the reminding mode selected by the user according to the selection of the user.

When the indication information indicates that a target person appears and the prompting mode includes identifying the target person, in this step, a face area of the target person may be determined first. Then, in the face region, a face contour of the target person is identified.

When a face area is determined, a first parameter included in a second preset position of the target live broadcast stream is obtained, and a normalization coordinate of a central point of the face contour in a human body image frame where the target person is located is obtained according to the first parameter. And acquiring a second parameter according to the normalized coordinate. Then, the human body image frame is divided into at least one region, and each region has a corresponding matrix. And traversing the matrix, wherein in the target matrix, if the element value corresponding to the second parameter is the target value, the target area corresponding to the target matrix is the face area.

And for the determined face area, determining pixel points corresponding to the face contour in the face area, and highlighting the target pixel points. For example, using a certain color to identify the pixel points.

When the client detects that the CSID field indicates that star X is identified, the client continues to monitor the CSID field and reads the first parameter, i.e., the five-bit integer. Then, the last four digits of the five-digit integer are split into two integers w_pAnd h_p，w_p100 and h_pAnd 100 respectively represent the normalized coordinates of the center point of the face frame of the star X in the frame image.

Then, using Mask-RCNN (Region convolutional neural network) algorithm to perform human body segmentation on the frame image, obtaining one or more Mask matrices with the size w × h, where each Mask matrix represents a Region of a person in the image, and is shaped as:

the matrix parameter corresponding to the human body area is 1, and the other areas are 0. Go through all mask matrices, if an element in a certain matrix:

it means that the matrix corresponds to star X.

Then, the colors of the pixels whose coordinates are the following values are each set to have a certain color, such as red:

that is, in the matrix, each row is traversed to find x is satisfied_c，rThe smallest C larger than 0, and the element corresponds to the leftmost edge of the face contour; find out to satisfy x_c，rThe largest C, greater than 0, corresponds to the rightmost edge of the face contour. And determining pixel points of the human face outline, and marking the pixel points by red.

In conjunction with fig. 3(b) and fig. 5, it is assumed that the target face detection module is disposed at the server. Taking football live broadcast as an example, the information prompting method of the embodiment of the invention can comprise the following steps:

and step S1, the image acquisition terminal acquires an image.

The image acquisition end is provided with a plurality of camera devices, and the main camera device is used for capturing a live main picture of the event, namely a picture finally watched by a user; the score change detection camera shooting equipment captures a score board picture, and one frame is extracted every second for score identification; the corner ball free-form ball detection camera device captures the whole court picture from the upper part of the court, and takes one frame per second for football detection and positioning; the red and yellow card detection camera device captures referee pictures, and takes one frame every second for red and yellow card detection.

And step S2, detecting an event by the image acquisition terminal.

After the event starts, the main camera device starts to collect real-time live broadcast pictures, and the real-time live broadcast pictures are transmitted to the collection end coding device to be subjected to RTMP coding, so that RTMP live broadcast streams are formed. The score change detection camera device, the corner ball free-form ball detection camera device and the red and yellow card detection camera device can transmit one frame of image to the coding device every second. The images are respectively analyzed by a Tesseract-OCR recognition algorithm, a football target detection algorithm (SSD) and a red and yellow board target detection algorithm (SSD) deployed on the coding equipment, and a score value, the position of a football on a court and whether a referee lifts a red and yellow board or not can be obtained every second.

And step S3, setting indication information by the encoding equipment of the image acquisition terminal.

The CSID field in the header of the RTMP protocol block supports user-customization and may take on a range of values [3,65599 ]. In the embodiment of the invention, different events are identified by setting the CSID field.

When an event detection algorithm on coding equipment of an image acquisition end detects score change, the coding service sets a CSID field to be 101; when the football position is detected to be continuously unchanged for 3 seconds, namely, the football free kick event is judged to occur, and the coding service sets the CSID field to be 102; when it is detected that the referee raised the red or yellow cards, the coding service sets the fields to 103 and 104, respectively.

And step S4, the server detects whether the target person appears, and if so, indication information is set.

And transmitting the RTMP live broadcast stream of the live broadcast main picture to the server. And decoding the RTMP stream by using decoding equipment at the server side to obtain image information, transmitting the image information to an MTCNN face detection algorithm and a Facenet face recognition algorithm, and performing star face recognition. When a face in the star repository is recognized, the encoding service sets the CSID field to the star number (e.g., 111 for star X).

The manner of identification may refer to the description of the foregoing method embodiments.

And step S5, the client acquires prompt selection information and a prompting mode set by the user.

The user can set the reminding of the event before the live broadcast starts and during the live broadcast. Then, accordingly, the client can acquire the setting information of the user. The user can see all events supporting the reminding service, including the above-mentioned "score change", "free orb", "red and yellow cards", "star X appears", "star Y appears", and the like, and the user can remind by checking which events occur. The client-side reminder action can be as described with reference to the previous embodiments, and the user can select which reminder action or actions to trigger upon each event.

And step S6, transmitting the RTMP live stream to the client through the CDN. And an event analysis module in the client player analyzes the CSID field of the RTMP stream, and if the number of the field is judged to correspond to an event preselected by a user, a reminding action set by the user is triggered. The method of marking the stars with red outlines may refer to the description of the previous embodiments.

Referring to fig. 7, fig. 7 is a structural diagram of an information presentation device according to an embodiment of the present invention. The information prompting device is applied to an image acquisition end. As shown in fig. 7, the information presentation apparatus 700 includes:

an acquisition module 701, configured to acquire a live broadcast image of a live broadcast scene; a first processing module 702, configured to encode the live broadcast image to obtain a live broadcast stream; a second processing module 703, configured to, when it is determined that a target event occurs in the live broadcast scene, add indication information of the target event to the live broadcast stream to obtain a target live broadcast stream; a sending module 704, configured to send the target live stream to a client, or send the target live stream to a server, so that the server sends the target live stream to the client, so that the client prompts the target event according to the indication information when performing live broadcast according to the target live stream.

Optionally, the target event includes: a score change event; as shown in fig. 8, the second processing module 703 includes:

the first acquisition submodule 7031 is configured to acquire an image of a score in the live broadcast scene; a first analysis sub-module 7032 for analyzing the images of the melding cards; a first determining sub-module 7033 configured to determine that the score change event has occurred in a case where it is determined that the score has changed according to the analysis result; a first obtaining sub-module 7034, configured to add a first identifier to a first predetermined position of the live stream, to obtain the target live stream; the first identifier is used for indicating that the score change event occurs.

Optionally, the target event includes: penalty events; as shown in fig. 9, the second processing module 703 includes: a second collecting submodule 7035, which collects images of referees in the live scene; a detection submodule 7036, configured to detect whether an image of a penalty appliance appears in the image of the referee; a second determining sub-module 7037 for determining that the penalty event has occurred in the case where it is determined that the image of the penalty appliance has occurred; a second obtaining sub-module 7038, configured to add a third identifier to a first predetermined position of the live stream to obtain the target live stream; the third indication is for indicating that the penalty event occurred.

Optionally, the target event includes: the appearance of a target character; as shown in fig. 10, the second processing module 703 includes: an identification submodule 7039 configured to perform face identification on the live broadcast image; a comparison submodule 70310, configured to compare the identified face image with a preset face image; a third determining sub-module 70311, configured to determine that the target person appears when the comparison result meets a predetermined condition; a third obtaining sub-module 70312, configured to add a fourth identifier to the first predetermined location of the live stream to obtain the target live stream, where the fourth identifier is used to indicate that the target person appears.

Optionally, on the basis shown in fig. 10, the second processing module further includes: the fourth determining submodule is used for determining the normalized coordinates of the central point of the face contour in the human body image frame where the target person is located; a fifth determining submodule, configured to determine a first parameter according to the normalized coordinate; and the adding submodule is used for adding the first parameter at a second preset position of the live broadcast stream, and the first parameter represents the relative position of the central point of the face contour in a live broadcast picture.

The working principle of the device of the embodiment of the invention can refer to the description of the embodiment of the method.

Referring to fig. 11, fig. 11 is a structural diagram of an information presentation device according to an embodiment of the present invention. The information prompt device is applied to a server side. As shown in fig. 11, the information presentation apparatus 1100 includes:

a receiving module 1101, configured to receive a target live stream sent by an image acquisition end, where the target live stream includes indication information of a target event; a sending module 1102, configured to send the target live stream to a client, so that the client prompts the target event according to the indication information when performing live broadcast according to the target live stream.

Optionally, in a case that the target event does not include the occurrence of the target person, as shown in fig. 12, the apparatus further includes: an obtaining module 1103, configured to decode the target live stream and obtain a live image; the recognition module 1104 is used for carrying out face recognition on the live broadcast image; a first determining module 1105, configured to determine whether a target person appears according to a result of the face recognition; a first adding module 1106, configured to, in an event that it is determined that the target person is present, add a fourth identifier to a first predetermined location of the live stream, where the fourth identifier indicates that the target person is present.

Optionally, on the basis shown in fig. 12, the apparatus may further include: the second determination module is used for determining the normalized coordinates of the central point of the face contour in the human body image frame where the target person is located; the third determining module is used for determining a first parameter according to the normalized coordinate; and the second adding module is used for adding the first parameter at a second preset position of the live broadcast stream, and the first parameter represents the relative position of the central point of the face contour in a live broadcast picture.

Referring to fig. 13, fig. 13 is a structural diagram of an information presentation device according to an embodiment of the present invention. The information prompting device is applied to a client. As shown in fig. 13, the information presentation apparatus 1300 includes:

a receiving module 1301, configured to receive a target live stream sent by an image acquisition end or a server, where the target live stream includes indication information of a target event; a prompting module 1302, configured to prompt the target event according to the indication information when performing live broadcast according to the target live broadcast stream.

Optionally, as shown in fig. 14, the prompt module 1302 includes:

a first obtaining sub-module 13021, configured to obtain prompt selection information of a user; a second obtaining sub-module 13022, configured to obtain a prompting mode pre-selected by the user when the prompting selection information matches the indication information; the prompt submodule 13023 is configured to prompt the target event by using the prompt mode.

Optionally, as shown in fig. 15, the prompt sub-module 13023 includes: a determining unit 130231, configured to determine a face area of a target person when the indication information indicates that the target person appears and the prompting manner includes identifying the target person; an identifying unit 130232, configured to identify the face contour of the target person in the face area.

Wherein the determining unit includes: a first obtaining subunit, configured to obtain a first parameter included in a second predetermined position of the target live stream; the second acquisition subunit is used for acquiring the normalized coordinates of the central point of the face contour in the human body image frame where the target person is located according to the first parameter; the third acquiring subunit is used for acquiring a second parameter according to the normalized coordinate; a dividing subunit, configured to divide the human body image frame into at least one region, each region having a corresponding matrix; and the determining subunit is configured to traverse the matrix, and in a target matrix, if an element value corresponding to the second parameter is a target value, a target area corresponding to the target matrix is a face area.

The identification unit includes: the determining subunit is used for determining pixel points corresponding to the face contour in the face region; and the identification subunit is used for highlighting the target pixel point.

As shown in fig. 16, the information prompt apparatus according to the embodiment of the present invention is applied to an image acquisition end, and includes: the processor 1600, which is used to read the program in the memory 1620, executes the following processes:

collecting a live broadcast image of a live broadcast scene; encoding the live broadcast image to obtain a live broadcast stream; under the condition that a target event occurs in the live broadcast scene, adding indication information of the target event in the live broadcast stream to obtain a target live broadcast stream; the target live stream is sent to a client through the transceiver 1610, or the target live stream is sent to a server, so that the server sends the target live stream to the client, and the client prompts the target event according to the indication information when performing live broadcasting according to the target live stream.

A transceiver 1610 for receiving and transmitting data under the control of the processor 1600.

In fig. 16, among other things, the bus architecture may include any number of interconnected buses and bridges, with one or more processors represented by the processor 1600 and various circuits of the memory represented by the memory 1620 linked together. The bus architecture may also link together various other circuits such as peripherals, voltage regulators, power management circuits, and the like, which are well known in the art, and therefore, will not be described any further herein. The bus interface provides an interface. The transceiver 1610 can be a plurality of elements including a transmitter and a transceiver providing a means for communicating with various other apparatus over a transmission medium. The processor 1600 is responsible for managing the bus architecture and general processing, and the memory 1620 may store data used by the processor 1600 in performing operations.

The processor 1600 is responsible for managing the bus architecture and general processing, and the memory 1620 may store data used by the processor 1600 in performing operations.

The target event comprises: a score change event; the processor 1600 is further configured to read the computer program and execute the following steps:

collecting images of the score cards in the live broadcast scene;

analyzing the image of the score card;

The target event comprises: a penalty event; the processor 1600 is further configured to read the computer program and execute the following steps:

acquiring a panoramic image of the live broadcast scene;

detecting a position of a game ball in the panoramic image;

adding a second identifier at a first preset position of the live stream to obtain the target live stream; the second identifier is used for indicating that the penalty ball event occurs.

The target event comprises the following steps: penalty events; the processor 1600 is further configured to read the computer program and execute the following steps:

collecting images of referees in the live scenes;

detecting whether an image of a penalty tool appears in the image of the judge;

The target event comprises: the appearance of a target character; the processor 1600 is further configured to read the computer program and execute the following steps:

carrying out face recognition on the live broadcast image;

comparing the recognized face image with a preset face image;

The processor 1600 is further configured to read the computer program and execute the following steps:

determining a first parameter according to the normalized coordinates;

As shown in fig. 17, the information prompt apparatus according to the embodiment of the present invention is applied to a server, and includes: a processor 1700 configured to read the program in the memory 1720 and execute the following processes:

receiving a target live stream sent by an image acquisition end through a transceiver 1710, wherein the target live stream comprises indication information of a target event; and sending the target live stream to a client so that the client prompts the target event according to the indication information when the client carries out live broadcasting according to the target live stream.

A transceiver 1710 for receiving and transmitting data under the control of the processor 1700.

In fig. 17, among other things, the bus architecture may include any number of interconnected buses and bridges, with one or more processors represented by processor 1700 and various circuits of memory represented by memory 1720 being linked together. The bus architecture may also link together various other circuits such as peripherals, voltage regulators, power management circuits, and the like, which are well known in the art, and therefore, will not be described any further herein. The bus interface provides an interface. The transceiver 1710 may be a plurality of elements including a transmitter and a transceiver, providing a means for communicating with various other apparatus over a transmission medium. The processor 1700 is responsible for managing the bus architecture and general processing, and the memory 1720 may store data used by the processor 1700 in performing operations.

The processor 1700 is responsible for managing the bus architecture and general processing, and the memory 1720 may store data used by the processor 1700 in performing operations.

In case the target event does not comprise the presence of a target person, the processor 1700 is further configured to read the computer program and perform the following steps:

decoding the target live broadcast stream to obtain a live broadcast image;

carrying out face recognition on the live broadcast image;

The processor 1700 is further configured to read the computer program and perform the following steps:

determining the normalized coordinates of the central point of the face contour in the human body image frame where the target person is located;

determining a first parameter according to the normalized coordinates;

as shown in fig. 18, the information prompt apparatus according to the embodiment of the present invention is applied to a client, and includes: the processor 1800, which reads the program stored in the memory 1820, executes the following processes:

receiving a target live broadcast stream sent by an image acquisition end or a server end through a transceiver 1810, wherein the target live broadcast stream comprises indication information of a target event; and prompting the target event according to the indication information when the live broadcast is carried out according to the target live broadcast stream.

A transceiver 1810 for receiving and transmitting data under the control of the processor 1800.

In fig. 18, among other things, the bus architecture may include any number of interconnected buses and bridges with various circuits including one or more processors, represented by the processor 1800, and memory, represented by the memory 1820. The bus architecture may also link together various other circuits such as peripherals, voltage regulators, power management circuits, and the like, which are well known in the art, and therefore, will not be described any further herein. The bus interface provides an interface. The transceiver 1810 may be a number of elements including a transmitter and a receiver that provide a means for communicating with various other apparatus over a transmission medium. The user interface 1830 may also be an interface capable of interfacing externally to a desired device for different user devices, including but not limited to a keypad, display, speaker, microphone, joystick, etc.

The processor 1800 is responsible for managing the bus architecture and general processing, and the memory 1820 may store data used by the processor 1800 in performing operations.

When the indication information indicates that the target person is present and the prompting mode includes identification of the target person, the processor 1800 is further configured to read the computer program and perform the following steps:

acquiring prompt selection information of a user;

and prompting the target event by utilizing the prompting mode.

The processor 1800 is further configured to read the computer program and perform the following steps:

determining a face area of the target person;

and identifying the face contour of the target person in the face area.

acquiring a normalized coordinate of the center point of the face contour in a human body image frame where the target person is located according to the first parameter;

acquiring a second parameter according to the normalized coordinate;

determining pixel points corresponding to the face contour in the face region;

and highlighting the target pixel point.

Furthermore, a computer-readable storage medium of an embodiment of the present invention stores a computer program that is executable by a processor to implement:

collecting a live broadcast image of a live broadcast scene;

encoding the live broadcast image to obtain a live broadcast stream;

and sending the target live stream to a client, or sending the target live stream to a server so that the server sends the target live stream to the client, thereby prompting the target event according to the indication information when the client carries out live broadcasting according to the target live stream.

collecting images of the score cards in the live broadcast scene;

analyzing the images of the score cards;

Wherein the target event comprises: a penalty event; adding the indication information of the target event in the live stream to obtain a target live stream, wherein the method comprises the following steps:

acquiring a panoramic image of the live broadcast scene;

detecting a position of a game ball in the panoramic image;

collecting images of referees in the live scenes;

detecting whether an image of a penalty tool appears in the image of the judge;

Wherein the target event comprises: the appearance of a target character; adding the indication information of the target event in the live stream to obtain a target live stream, wherein the step of adding the indication information of the target event in the live stream comprises the following steps:

carrying out face recognition on the live broadcast image;

comparing the recognized face image with a preset face image;

determining a first parameter according to the normalized coordinates;

Furthermore, a computer-readable storage medium of an embodiment of the present invention stores a computer program executable by a processor to implement:

decoding the target live broadcast stream to obtain a live broadcast image;

carrying out face recognition on the live broadcast image;

determining a first parameter according to the normalized coordinates;

acquiring prompt selection information of a user;

and prompting the target event by utilizing the prompting mode.

determining a face area of the target person;

and identifying the face contour of the target person in the face area.

Wherein the determining the face area of the target person comprises:

acquiring a second parameter according to the normalized coordinate;

Wherein, in the face region, identifying the face contour of the target person includes:

determining pixel points corresponding to the face contour in the face region;

and highlighting the target pixel point.

In the several embodiments provided in the present application, it should be understood that the disclosed method and apparatus may be implemented in other ways. For example, the above-described apparatus embodiments are merely illustrative, and for example, the division of the units is only one logical division, and other divisions may be realized in practice, for example, a plurality of units or components may be combined or integrated into another system, or some features may be omitted, or not executed. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection through some interfaces, devices or units, and may be in an electrical, mechanical or other form.

In addition, functional units in the embodiments of the present invention may be integrated into one processing unit, or each unit may be separately and physically included, or two or more units may be integrated into one unit. The integrated unit can be realized in a form of hardware, or in a form of hardware plus a software functional unit.

The integrated unit implemented in the form of a software functional unit may be stored in a computer readable storage medium. The software functional unit is stored in a storage medium and includes several instructions to enable a computer device (which may be a personal computer, a server, or a network device) to execute some steps of the transceiving method according to various embodiments of the present invention. And the aforementioned storage medium includes: various media capable of storing program codes, such as a usb disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk, or an optical disk.

While the foregoing is directed to the preferred embodiment of the present invention, it will be understood by those skilled in the art that various changes and modifications may be made without departing from the spirit and scope of the invention as defined in the appended claims.

Claims

1. An information prompting method is applied to an image acquisition end and is characterized by comprising the following steps:

collecting a live broadcast image of a live broadcast scene;

encoding the live broadcast image to obtain a live broadcast stream;

sending the target live stream to a client, or sending the target live stream to a server, so that the server sends the target live stream to the client, and the client prompts the target event according to the indication information when performing live broadcast according to the target live stream;

wherein the target event comprises: score change events, penalty ball events, penalty events and target characters appear;

the indication information is located in the header information of the RTMP block.

2. The method of claim 1, wherein the target event comprises: a score change event; adding the indication information of the target event in the live stream to obtain a target live stream, wherein the step of adding the indication information of the target event in the live stream comprises the following steps:

collecting images of the score cards in the live broadcast scene;

analyzing the images of the score cards;

3. The method of claim 1, wherein the target event comprises: a penalty event; adding the indication information of the target event in the live stream to obtain a target live stream, wherein the step of adding the indication information of the target event in the live stream comprises the following steps:

acquiring a panoramic image of the live broadcast scene;

detecting a position of a game ball in the panoramic image;

4. The method of claim 1, wherein the target event comprises: penalty events; adding the indication information of the target event in the live stream to obtain a target live stream, wherein the method comprises the following steps:

collecting images of referees in the live scenes;

detecting whether an image of a penalty tool appears in the image of the judge;

in a case where it is determined that the image of the penalty appliance appears, it is determined that the penalty event has occurred;

5. The method of claim 1, wherein the target event comprises: the appearance of a target character; adding the indication information of the target event in the live stream to obtain a target live stream, wherein the step of adding the indication information of the target event in the live stream comprises the following steps:

carrying out face recognition on the live broadcast image;

comparing the recognized face image with a preset face image;

6. The method of claim 5, wherein after determining that the target person appears if the comparison result satisfies a predetermined condition, the method further comprises:

determining a first parameter according to the normalized coordinates;

7. An information prompting method is applied to a server side and is characterized by comprising the following steps:

sending the target live stream to a client so that the client prompts the target event according to the indication information when the client carries out live broadcasting according to the target live stream;

8. The method of claim 7, wherein in a case that the target event does not include an occurrence of a target person, prior to the sending of the target live stream to the client, the method further comprises:

decoding the target live broadcast stream to obtain a live broadcast image;

carrying out face recognition on the live broadcast image;

9. The method of claim 8, wherein after determining whether the target person is present according to the result of the face recognition, the method further comprises:

determining a first parameter according to the normalized coordinates;

10. An information prompting method is applied to a client side and is characterized by comprising the following steps:

when the live broadcast is carried out according to the target live broadcast stream, prompting the target event according to the indication information;

wherein the indication information is located in the header information of the RTMP block.

11. The method of claim 10, wherein the prompting the target event according to the indication information comprises:

acquiring prompt selection information of a user;

and prompting the target event by utilizing the prompting mode.

12. The method of claim 11, wherein when the indication indicates that a target person is present and when the prompting mode includes identifying the target person, the prompting the target event using the prompting mode comprises:

determining a face area of the target person;

and identifying the face contour of the target person in the face area.

13. The method of claim 12, wherein determining the face region of the target person comprises:

acquiring a second parameter according to the normalized coordinate;

14. The method of claim 12, wherein identifying the face contour of the target person in the face region comprises:

determining pixel points corresponding to the face contour in the face region;

and highlighting the target pixel point.

15. An information prompting device, comprising: a transceiver, a memory, a processor, and a computer program stored on the memory and executable on the processor; it is characterized in that the preparation method is characterized in that,

the processor for reading the program in the memory to implement the steps in the method of any one of claims 1 to 6; or implementing a step in a method according to any one of claims 7 to 9; or implementing a step in a method according to any of claims 10 to 14.

16. A computer-readable storage medium for storing a computer program, wherein the computer program, when executed by a processor, implements the steps in the method according to any one of claims 1 to 6; or implementing a step in a method according to any one of claims 7 to 9; or implementing a step in a method according to any of claims 10 to 14.

17. An information prompting system, comprising: the system comprises an image acquisition end and at least one client;

the client is used for receiving a target live broadcast stream sent by the image acquisition end and prompting the target event according to the indication information when the target live broadcast stream is subjected to live broadcast;

wherein the target event comprises: a score change event, a penalty event and the appearance of a target figure;

18. An information prompting system, comprising: the system comprises an image acquisition end, a server end and at least one client end;

the server is used for sending the target live stream to the client;

the client is used for receiving a target live stream sent by the server and prompting the target event according to the indication information when the target live stream is broadcasted directly;