A positioning management method and system for screen recording video conference

Abstract

The present application discloses a method and system for positioning management of screen recording video conferences, which belongs to the field of artificial intelligence. The method includes: collecting screen recording videos, performing frame-by-frame text recognition on the screen recording videos, and obtaining text recognition results; obtaining initial segmentation identification results based on the text recognition results; performing segment matching based on the user's input segmentation data and text recognition results to obtain segmentation identification results; performing character feature recognition and object feature recognition on the screen recording videos, and constructing character feature recognition identification data and object feature identification data respectively; performing identification processing on the time nodes of the screen recording videos, and performing positioning management based on the time node identification processing results. The present application solves the technical problems in the prior art that the positioning management of screen recording video conferences has low accuracy and intelligence, and cannot quickly locate the content of the conference. The technical effect of quickly locating the content of the screen recording conference and improving the quality of video viewing is achieved.

Description

Technical Field

[0001] The present application relates to the field of artificial intelligence, and in particular to a positioning management method and system for screen recording video conferencing. Background Art

[0002] At present, people at different addresses use the Internet to achieve face-to-face meetings, file sharing and file transfer in different places. At the same time, the content of the video conference can be recorded, which is convenient for people who did not attend the meeting to watch it in time and for those who attended the meeting to watch it again, thus providing great convenience for people's work and life. However, due to the large amount of recorded video content, when it is necessary to find a video with specific content, it is impossible to accurately locate the corresponding position, and it is necessary to keep dragging the progress bar to search for content, thereby reducing viewing efficiency. There are technical problems in the prior art that the positioning management accuracy and intelligence of screen recording video conferences are low, and the meeting content cannot be quickly located. Summary of the invention

[0003] The purpose of this application is to provide a method and system to solve the technical problems in the prior art of low accuracy and intelligence in the positioning management of screen recording video conferences and the inability to quickly locate the content of the conference.

[0004] In view of the above problems, the present application provides a positioning management method and system for screen recording video conferencing.

[0005] In a first aspect, the present application provides a positioning management method for a screen recording video conference, wherein the method includes: collecting a screen recording video, performing frame-by-frame text recognition on the screen recording video, and obtaining a text recognition result, and the text recognition result includes time identification information; performing initial section identification based on the text recognition result to obtain an initial section identification result; obtaining the user's input section section data, performing section matching based on the input section section data and the text recognition result to obtain a section section identification result; performing character feature recognition on the screen recording video, and constructing character feature recognition identification data; performing object feature recognition on the screen recording video, and constructing object feature identification data based on the recognition result; performing time node identification processing on the screen recording video through the initial section identification result, the section section identification result, the character feature recognition identification data, and the object feature identification data, and performing positioning management based on the time node identification processing result.

[0006] On the other hand, the present application also provides a positioning management system for screen recording video conferences, wherein the system includes: a text recognition module, which is used to collect screen recording videos, perform frame-by-frame text recognition on the screen recording videos, and obtain text recognition results, and the text recognition results include time identification information; a separate identification module, which is used to perform initial section identification based on the text recognition results to obtain initial section identification results; a plate matching module, which is used to obtain the user's input plate section data, perform plate matching based on the input plate section data and the text recognition results, and obtain a plate section identification result. ; A person feature recognition module, the person feature recognition module is used to perform person feature recognition on the screen recording video and construct person feature recognition identification data; an object feature recognition module, the object feature recognition module is used to perform object feature recognition on the screen recording video and construct item feature identification data based on the recognition result; a positioning management module, the positioning management module is used to perform time node identification processing on the screen recording video through the initial section identification result, the section section identification result, the person feature recognition identification data and the item feature identification data, and perform positioning management based on the time node identification processing result.

[0007] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0008] After obtaining the screen recording video based on big data, this application first recognizes the text on each frame of the screen recording video to pave the way for subsequent video positioning based on text, divides the chapters of the video conference according to the results of text recognition, and performs initial section identification, and then obtains the section section identification result according to the section section data input by the user, and then matches the content in the text recognition result, and then identifies the character features in the screen recording video to obtain character feature identification data, which can be used to subsequently locate the video content based on the character, and then identify the objects in the screen recording video, identify the object features, and then perform time node identification processing of the screen recording video according to the initial section identification result, the section section identification result, the character feature identification data and the object feature identification data, and perform video positioning management according to the processing result. Thus, the technical effect of improving the accuracy of positioning and improving management efficiency and quality is achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] In order to more clearly illustrate the technical solutions in the present application or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are only exemplary, and for ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying any creative work.

[0010] Figure 1 A flowchart of a method for positioning management of a screen recording video conference provided in an embodiment of the present application;

[0011] Figure 2 A schematic diagram of a process for obtaining an initial segment identification result in a location management method for a screen recording video conference provided in an embodiment of the present application;

[0012] Figure 3 A schematic diagram of a flow chart of time node identification processing for screen recording videos in a positioning management method for screen recording video conferences provided in an embodiment of the present application;

[0013] Figure 4 A schematic diagram of the structure of a positioning management system for screen recording video conferencing for this application;

[0014] Explanation of the reference numerals: text recognition module 11 , identification module 12 , plate matching module 13 , person feature recognition module 14 , object feature recognition module 15 , positioning management module 16 . DETAILED DESCRIPTION

[0015] This application provides a method and system for positioning management of screen recording video conferences, which solves the technical problems in the prior art that the positioning management accuracy and intelligence of screen recording video conferences are low and the conference content cannot be quickly positioned. The technical effect of quickly positioning the content of the screen recording conference and improving the quality of video viewing is achieved.

[0016] The acquisition, storage, use, and processing of data in the technical solution of this application comply with the relevant provisions of national laws and regulations.

[0017] Below, the technical solutions in the present application will be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments of the present application. It should be understood that the present application is not limited to the example embodiments described herein. Based on the embodiments of the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present application. It should also be noted that, for the convenience of description, only the parts related to the present application are shown in the accompanying drawings, rather than all of them.

[0018] Embodiment 1

[0019] like Figure 1 As shown, the present application provides a method for positioning management of screen recording video conference, wherein the method includes:

[0020] Step S100: collecting screen recording video, performing frame-by-frame text recognition on the screen recording video, and obtaining a text recognition result, wherein the text recognition result includes time mark information;

[0021] Specifically, a screen recording video is formed by recording a video conference, and then the screen recording video is collected based on big data. The screen recording video is any video that needs to be identified by marking the time node to locate the content. The text contained in each frame of the screen recording video is recognized to obtain a text recognition result, and time mark information is added to the text recognition result. Optionally, the text in the video can be recognized by OCR recognition, and a time mark is added to the text recognition result to ensure that the text corresponds to the frame in the video. Among them, the text recognition result refers to the position, size, format, correlation between texts and other information obtained after the text in each frame of the video is recognized. The time mark information is the time point at which the recognized text is located in the video, and the position of the text in the video is located by marking the time point. As a result, the technical effect of paving the way for content analysis and identification based on the text recognition result is achieved.

[0022] Step S200: performing initial section marking based on the text recognition result to obtain an initial section marking result;

[0023] Further, such as Figure 2 As shown, initial section identification is performed based on the text recognition result to obtain an initial section identification result. Step S200 of the embodiment of the present application further includes:

[0024] Step S210: constructing a digital correlation feature set based on big data;

[0025] Step S220: performing digital association combination of the text recognition results using the digital association feature set to obtain a combination solution set;

[0026] Step S230: Evaluate the magnitude of each combination number in the combination solution set, and screen the combination solution set based on the numerical magnitude evaluation result to obtain a screened solution set;

[0027] Step S240: Obtain the initial section identification result through the screening scheme set.

[0028] Furthermore, step S200 of the embodiment of the present application further includes:

[0029] Step S250: recognizing the space feature before the number in the text recognition result, performing space feature matching constraint based on the space feature recognition result, and generating a space constraint condition;

[0030] Step S260: recognizing the symbol features before and after the numbers in the text recognition result, and obtaining symbol constraint conditions based on the symbol feature recognition result;

[0031] Step S270: modifying the screening solution set by using the space constraint and the symbol constraint to obtain a modified solution set;

[0032] Step S280: Obtain the initial section identification result through the correction solution set.

[0033] Specifically, in the process of conducting a video conference, in order to highlight the content of the meeting and to facilitate the participants to understand the content of the meeting and grasp the spirit of the meeting, different titles are set for the content of the meeting to be told according to different narration points. Then, the content is made clearer by arranging the titles digitally. Therefore, by analyzing the numbers in the text recognition results, the overall meeting content can be divided into chapters.

[0034] Specifically, the digital association feature set is the association relationship between the chapter distinction numbers that are common to the conference content obtained based on big data, and optionally includes: a first-level title format, a second-level title format, a third-level title format, etc. Exemplarily, the first-level title format is a title with a serial number of "one," "two," "three," etc., with the largest font size, and the first-level title is a title formed after the main content of the video discussion is refined. The second-level title format is a title with a serial number of "(one)", "(two)," "(three)," etc., with a medium font size, and the second-level title is a title obtained after the sub-arguments under the first-level title are refined. According to the interrelated features of the numbers in the digital association feature set, the numbers in the text recognition results are associated and combined, and the numbers with the same digital features are classified into the same combination, thereby obtaining the combination scheme set. Among them, the combination scheme set is a set obtained by classifying and combining all the numbers in the text recognition results according to the same digital features. Then, the digital size of each combination number is evaluated. The larger the number, the more the central content is reflected in the video content. Then, the combination scheme set is screened according to the data size evaluation result to obtain the combination scheme with the largest digital size evaluation result. According to the time identification information of each number in the combination scheme, the video content is initially segmented to obtain the initial segmentation identification result of the video. The initial segmentation identification is to divide the video content into levels according to the main content as a whole, and to identify the alternation of each chapter.

[0035] Specifically, the screening scheme set is modified by comprehensively considering the space feature before the number and the symbol feature before and after the number. This is mainly because in addition to the title position in the video content, numbers will also appear in the conference content. At the same time, the size of the number is consistent with the size of the number in the title, which will affect the number combination. The space feature refers to the number of spaces before the number. The space constraint condition is a condition that limits the number of spaces before the number in the screening scheme set. The specific limit on how many spaces can be before the number is set by the staff and is not limited here. The identification of the symbol feature before and after the number refers to the punctuation before and after the number, which reflects the specific role of the number, whether it is used for counting or in the title. The symbol constraint condition refers to the condition that limits the number and type of symbols before and after the number. Then, according to the space constraint condition and the symbol constraint condition, the number combination in the screening scheme set is modified and adjusted to obtain a set of modified schemes. Therefore, the initial segmentation identification result obtained according to the modified scheme can more accurately divide the video into layers, thereby improving the accuracy of positioning.

[0036] Step S300: obtaining the input section segmentation data of the user, performing section matching based on the input section segmentation data and the text recognition result, and obtaining a section segmentation identification result;

[0037] Specifically, the user's input section segmentation data is data input by the user to divide the video content into sections. Section matching is then performed based on the section segmentation data and the text recognition result, and the content contained in the section segmentation data is matched with the data content in the text recognition result. When the match is successful, the video content can be sectioned and located according to the section based on the time mark information in the text recognition result, thereby achieving the goal of dividing and locating the video content from the user's perspective.

[0038] For example, if a user wants to hold a thesis report discussion meeting, the thesis report can be divided into the following sections according to its content: subject background, project research significance, literature research overview, main design content, and progress plan. By matching the data corresponding to the sections with the text recognition results, for example, the progress plan section corresponds to all progress-related content, and by matching it with the progress data in the text recognition results, the video content occupied by the progress plan section in the discussion meeting can be obtained based on the first and last occurrences of the progress data.

[0039] Step S400: performing character feature recognition on the screen recording video and constructing character feature recognition identification data;

[0040] Further, such as Figure 3 As shown, step S400 of the embodiment of the present application also includes:

[0041] Step S410: Collecting voiceprint feature information of meeting participants;

[0042] Step S420: performing voice feature recognition of the screen recording video using the voiceprint feature information, and generating personnel voiceprint identification information based on the recognition result;

[0043] Step S430: collecting facial feature information of the participants, wherein the facial feature information has a one-to-one correspondence with the voiceprint feature information;

[0044] Step S440: constructing the character feature recognition identification data through the facial feature information, and performing auxiliary identification of the character feature recognition identification data based on the person voiceprint identification information to obtain an auxiliary identification result;

[0045] Step S450: perform time node identification processing on the screen recording video using the auxiliary identification result.

[0046] Furthermore, step S410 of the embodiment of the present application further includes:

[0047] Step S411: constructing a voiceprint quantity constraint value;

[0048] Step S412: determining whether the voiceprint quantity information at the same time node satisfies the voiceprint quantity constraint value;

[0049] Step S413: when the voiceprint quantity information at the same time node can satisfy the voiceprint quantity constraint value, a time interval is formed based on the time node that can satisfy the voiceprint quantity constraint value;

[0050] Step S414: Mark the conference discussion of the screen recording video according to the time interval.

[0051] Specifically, the voiceprint feature information is the pitch, intensity, length, and timbre of the sound emitted by a person, which reflects the voice characteristics of different people. Different people can be identified according to the voiceprint features. The sound characteristics of the screen recording video are identified according to the voiceprint feature information, and the person and the sound in the screen recording video are matched one by one to obtain the person's voiceprint identification information. Among them, the person's voiceprint identification information is the information obtained after the person's voice in the screen recording video is identified, which can reflect the identity of the person who made the sound. Then, the facial feature information of the participants is collected, including: facial features proportion, face shape, facial features shape, color, etc. By matching the facial feature information with the voiceprint feature information one by one, preparation is made for the subsequent establishment of auxiliary identification. Different participants can be identified according to the facial feature information, and then the character feature identification identification data is constructed to identify the person appearing in the video, and then the corresponding person's voiceprint identification information is obtained to obtain the auxiliary identification result, and the identification processing is performed on the time node where the person who makes the sound in the screen recording video is located.

[0052] Specifically, during the meeting, there will be a discussion session on the content of the meeting. During the discussion, different participants will discuss the same issue, and multiple voices will appear in the video. By judging the number of voiceprints, it is determined whether a meeting discussion is being conducted. Among them, the voiceprint number constraint value is to determine the minimum number of voices that appear during the meeting discussion. If the number of voiceprints at the same time node is lower than the voiceprint number constraint value, it indicates that there is no meeting discussion at this time, and if it is higher than the voiceprint number constraint value, it indicates that there is a meeting discussion at this time. Therefore, by judging whether the voiceprint number information at the same time node meets the voiceprint number constraint value, it can be determined whether a meeting discussion is being conducted at the current time node. By determining the time node that meets the voiceprint number constraint value, the first and last time nodes are taken in time order to obtain the time interval, and the time interval is the time period for the meeting discussion. Furthermore, according to the time interval, the meeting discussion mark is performed in the screen recording video, so as to achieve the goal of meeting positioning in the video, and achieve the technical effect of improving the accuracy of video positioning management.

[0053] Step S500: performing object feature recognition on the screen recording video, and constructing object feature identification data based on the recognition result;

[0054] Specifically, the objects that appear in the video are sometimes the focus of discussion in the meeting, so it is necessary to identify the features of the objects in the video. The features of the objects include size, shape, color, etc. According to the recognition results, the objects in the video are identified to obtain the object feature identification data. The object feature identification data is used to quickly locate the video content related to the object in the video according to the identification. In this way, the goal of locating the video content from multiple aspects is achieved, and the technical effect of improving the accuracy of video positioning is achieved.

[0055] Step S600: Perform time node identification processing on the screen recording video using the initial section identification result, the section section identification result, the person feature recognition identification data and the object feature identification data, and perform positioning management based on the time node identification processing result.

[0056] Furthermore, step S600 in the embodiment of the present application further includes:

[0057] Step S610: Obtaining the authority information of the person viewing the screen recording video;

[0058] Step S620: obtaining feedback identification data of the viewing personnel;

[0059] Step S630: Adjust the identification of the screen recording video according to the personnel authority information and the feedback identification data.

[0060] Furthermore, step S630 of the embodiment of the present application further includes:

[0061] Step S631: When the personnel authority information is administrator authority, the overall identification of the screen recording video is adjusted based on the feedback identification data;

[0062] Step S632: When the personnel authority information is non-administrator authority, the local video identification of the viewing person is adjusted based on the feedback identification data.

[0063] Specifically, the time nodes in the screen recording video are marked according to the initial section marking results, the section section marking results, the character feature identification data and the object feature identification data, so as to achieve the goal of multi-level differentiation of video content from multiple angles such as chapters, sections, characters and objects, and achieve the technical effect of accurate video positioning. Then, according to the time node identification processing results in the video, it is possible to directly jump to the corresponding content of the video according to the identification information, and find the corresponding video content by searching for the corresponding identification, so as to perform positioning management on the video.

[0064] Specifically, the personnel authority information reflects the scope of adjustment that the personnel who watch the screen recording video can make to the identification information in the video. Different authorities have different degrees of adjustment to the identification, and the degree of influence on other viewers after the identification adjustment is different. The feedback identification information is the feedback information obtained by evaluating the identification result and the accuracy of positioning presented by the existing time node identification processing result to increase or delete the identification result. When the personnel authority information is the administrator authority, the overall identification of the screen recording video can be adjusted according to the feedback identification information. After the identification information is adjusted, the identification information seen by the personnel who watch the screen recording video later is the identification adjusted by the administrator. When the personnel authority information is non-administrator authority, the identification of the local video can only be adjusted according to the feedback identification information. The adjusted identification information will only appear in the video of the adjuster himself, and will not affect the viewing of other personnel. As a result, it is not only ensured that the identification information of the video will not be tampered with at will, affecting the accuracy of positioning and improving the management level, but also can meet the personalized needs of different viewers for the video and improve the diversity of management.

[0065] In summary, the positioning management method for screen recording video conference provided by this application has the following technical effects:

[0066] 1. In the embodiment of the present application, the video content is analyzed from multiple angles such as chapters, sections, characters, and objects through multi-angle analysis of the screen recording video, thereby meeting the user's needs for video positioning. First, the text appearing in the video is identified, and the chapter features are obtained according to the text content, so as to perform initial segmentation. Then, according to the data obtained after the user inputs the section division of the meeting content, the corresponding text is matched, and the section segmentation identification result is obtained. Then, the character features are identified, and the voiceprint features are used to assist in identifying the meeting discussion time period. In addition, the object feature identification data is obtained by identifying the object. Finally, the time node identification of the screen recording video is processed according to the above feature identification results, and then the video positioning management is performed. The technical effect of improving the accuracy of video positioning and meeting the user's positioning needs is achieved.

[0067] 2. The embodiment of the present application determines the chapter division of the meeting based on the digital features in the video. This is mainly done by establishing digital association features and then matching them with the text recognition results. The numbers with the same features that appear in the video are grouped together to obtain a digital combination scheme, and then the size of the numbers is evaluated to determine the combination scheme representing the chapters, thereby obtaining the initial section identification result. This achieves the technical effect of dividing the video into sections and improving the accuracy of video positioning management.

[0068] Embodiment 2

[0069] Based on the same inventive concept as the location management method for screen recording video conference in the aforementioned embodiment, Figure 4 As shown, the present application also provides a location management system for screen recording video conference, wherein the system includes:

[0070] A text recognition module 11, which is used to collect screen recording video, perform frame-by-frame text recognition on the screen recording video, and obtain a text recognition result, and the text recognition result includes time identification information;

[0071] A separate identification module 12, the separate identification module 12 is used to perform initial section identification based on the text recognition result to obtain an initial section identification result;

[0072] A section matching module 13, the section matching module 13 is used to obtain the user's input section segmentation data, perform section matching based on the input section segmentation data and the text recognition result, and obtain a section segmentation identification result;

[0073] A character feature recognition module 14, wherein the character feature recognition module 14 is used to perform character feature recognition on the screen recording video and construct character feature recognition identification data;

[0074] An object feature recognition module 15, the object feature recognition module 15 is used to perform object feature recognition on the screen recording video, and construct object feature identification data based on the recognition result;

[0075] The positioning management module 16 is used to perform time node identification processing of the screen recording video through the initial section identification result, the section section identification result, the person feature recognition identification data and the object feature identification data, and perform positioning management based on the time node identification processing result.

[0076] Furthermore, the system also includes:

[0077] A correlation feature construction unit, the correlation feature construction unit is used to construct a digital correlation feature set based on big data;

[0078] A combination scheme obtaining unit, the combination scheme obtaining unit is used to perform digital association combination of the text recognition results through the digital association feature set to obtain a combination scheme set;

[0079] A scheme screening unit, the scheme screening unit is used to evaluate the numerical magnitude of each combination number in the combination scheme set, and screen the combination scheme set based on the numerical magnitude evaluation result to obtain a screening scheme set;

[0080] A section identification obtaining unit is used to obtain the initial section identification result through the screening scheme set.

[0081] Furthermore, the system also includes:

[0082] A space constraint generating unit, the space constraint generating unit is used to identify the space feature before the digit in the text recognition result, perform space feature matching constraint based on the space feature recognition result, and generate a space constraint condition;

[0083] A symbol feature recognition unit, the symbol feature recognition unit is used to recognize the symbol features before and after the numbers in the text recognition result, and obtain the symbol constraint condition based on the symbol feature recognition result;

[0084] A solution modification unit, the solution modification unit is used to modify the screening solution set according to the space constraint condition and the symbol constraint condition to obtain a modified solution set;

[0085] The identification result obtaining unit is used to obtain the initial section identification result through the correction scheme set.

[0086] Furthermore, the system also includes:

[0087] A voiceprint feature collection unit, which is used to collect voiceprint feature information of participants;

[0088] A sound feature recognition unit, the sound feature recognition unit is used to recognize the sound features of the screen recording video through the voiceprint feature information, and generate personnel voiceprint identification information based on the recognition result;

[0089] A facial feature collection unit, the facial feature collection unit is used to collect facial feature information of the participants, wherein the facial feature information has a one-to-one correspondence with the voiceprint feature information;

[0090] An auxiliary identification unit, the auxiliary identification unit is used to construct the character feature identification data through the facial feature information, and perform auxiliary identification of the character feature identification data based on the person's voiceprint identification information to obtain an auxiliary identification result;

[0091] A time node identification processing unit, wherein the time node identification processing unit is used to perform time node identification processing on the screen recording video through the auxiliary identification result.

[0092] Furthermore, the system also includes:

[0093] A voiceprint quantity constraint unit, wherein the voiceprint quantity constraint unit is used to construct a voiceprint quantity constraint value;

[0094] A voiceprint quantity determination unit, the voiceprint quantity determination unit is used to determine whether the voiceprint quantity information at the same time node meets the voiceprint quantity constraint value;

[0095] A time node obtaining unit, wherein when the voiceprint quantity information at the same time node can satisfy the voiceprint quantity constraint value, the time node obtaining unit is used to form a time interval based on the time nodes that can satisfy the voiceprint quantity constraint value;

[0096] A discussion identification unit, wherein the discussion identification unit is used to identify the conference discussion of the screen recording video according to the time interval.

[0097] Furthermore, the system also includes:

[0098] A personnel authority obtaining unit, the personnel authority obtaining unit is used to obtain the personnel authority information of the person viewing the screen recording video;

[0099] A feedback identification data obtaining unit, the feedback identification data obtaining unit is used to obtain the feedback identification data of the viewing personnel;

[0100] An identification adjustment unit, wherein the identification adjustment unit is used to adjust the identification of the screen recording video according to the personnel authority information and the feedback identification data.

[0101] Furthermore, the system also includes:

[0102] An overall identification adjustment unit, wherein the overall identification adjustment unit is used to adjust the overall identification of the screen recording video based on the feedback identification data when the personnel authority information is administrator authority;

[0103] A local video identification adjustment unit, wherein the local video identification adjustment unit is used to adjust the local video identification of the viewing person based on the feedback identification data when the person authority information is non-administrator authority.

[0104] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. Figure 1 The location management method and specific examples of a screen recording video conference in the first embodiment are also applicable to the location management system of a screen recording video conference in the present embodiment. Through the above detailed description of the location management method of a screen recording video conference, those skilled in the art can clearly know the location management system of a screen recording video conference in the present embodiment, so for the sake of brevity of the specification, it will not be described in detail here. For the device disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part description.

[0105] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for positioning and managing screen recording video conference, characterized in that: The method comprises: Collecting screen recording video, performing frame-by-frame text recognition on the screen recording video, and obtaining a text recognition result, wherein the text recognition result includes time mark information; Performing initial section marking based on the text recognition result to obtain an initial section marking result; Obtaining the input section segmentation data of the user, performing section matching based on the input section segmentation data and the text recognition result, and obtaining a section segmentation identification result; Performing character feature recognition on the screen recording video and constructing character feature recognition identification data; Performing object feature recognition on the screen recording video, and constructing object feature identification data based on the recognition result; Perform time node identification processing on the screen recording video through the initial section identification result, the section section identification result, the character feature recognition identification data and the object feature identification data, and perform positioning management based on the time node identification processing result; The method further comprises: Construct digital correlation feature sets based on big data; Performing digital association combination of the text recognition results through the digital association feature set to obtain a combination scheme set; Evaluate the magnitude of each combination number in the combination solution set, and screen the combination solution set based on the numerical magnitude evaluation result to obtain a screened solution set; Obtaining the initial section identification result through the screening scheme set; Recognize the space feature before the number in the text recognition result, perform space feature matching constraints based on the space feature recognition result, and generate space constraint conditions; Recognize the symbol features before and after the numbers in the text recognition result, and obtain the symbol constraint condition based on the symbol feature recognition result; Modifying the screening solution set by using the space constraint condition and the symbol constraint condition to obtain a modified solution set; Obtaining the initial segment identification result through the correction scheme set; Collect voiceprint feature information of participants; Performing voice feature recognition of the screen recording video using the voiceprint feature information, and generating personnel voiceprint identification information based on the recognition result; Collecting facial feature information of the participants, wherein the facial feature information has a one-to-one correspondence with the voiceprint feature information; Constructing the character feature recognition identification data through the facial feature information, and performing auxiliary identification of the character feature recognition identification data based on the person voiceprint identification information to obtain an auxiliary identification result; Perform time node identification processing on the screen recording video using the auxiliary identification result; Construct the constraint value of the number of voiceprints; Determine whether the voiceprint quantity information at the same time node meets the voiceprint quantity constraint value; When the voiceprint quantity information at the same time node can satisfy the voiceprint quantity constraint value, a time interval is formed based on the time node that can satisfy the voiceprint quantity constraint value; Identify the conference discussion of the screen recording video according to the time interval; Obtaining the personnel authority information of the person viewing the screen recording video; Obtaining feedback identification data of the viewing personnel; The identification of the screen recording video is adjusted according to the personnel authority information and the feedback identification data.

2. The method according to claim 1, characterized in that The method further comprises: When the personnel authority information is administrator authority, the overall identification of the screen recording video is adjusted based on the feedback identification data; When the personnel authority information is non-administrator authority, the local video identification of the viewing person is adjusted based on the feedback identification data.

3. A location management system for screen recording video conference, characterized in that: The system comprises: A text recognition module, wherein the text recognition module is used to collect screen recording videos, perform frame-by-frame text recognition on the screen recording videos, and obtain text recognition results, and the text recognition results include time identification information; A separate identification module, the separate identification module is used to perform initial section identification based on the text recognition result to obtain an initial section identification result; A section matching module, the section matching module is used to obtain the user's input section segmentation data, perform section matching based on the input section segmentation data and the text recognition result, and obtain a section segmentation identification result; A character feature recognition module, the character feature recognition module is used to perform character feature recognition on the screen recording video and construct character feature recognition identification data; An object feature recognition module, the object feature recognition module is used to perform object feature recognition on the screen recording video, and construct object feature identification data based on the recognition result; A positioning management module, the positioning management module is used to perform time node identification processing of the screen recording video through the initial section identification result, the section section identification result, the character feature recognition identification data and the object feature identification data, and perform positioning management based on the time node identification processing result; A correlation feature construction unit, the correlation feature construction unit is used to construct a digital correlation feature set based on big data; A combination scheme obtaining unit, the combination scheme obtaining unit is used to perform digital association combination of the text recognition results through the digital association feature set to obtain a combination scheme set; A scheme screening unit, the scheme screening unit is used to evaluate the magnitude of the numbers of each combination in the combination scheme set, and screen the combination scheme set based on the numerical magnitude evaluation result to obtain a screening scheme set; A section identification obtaining unit, the section identification obtaining unit is used to obtain the initial section identification result through the screening scheme set; A space constraint generating unit, the space constraint generating unit is used to identify the space feature before the digit in the text recognition result, perform space feature matching constraint based on the space feature recognition result, and generate a space constraint condition; A symbol feature recognition unit, the symbol feature recognition unit is used to recognize the symbol features before and after the numbers in the text recognition result, and obtain the symbol constraint condition based on the symbol feature recognition result; A solution modification unit, the solution modification unit is used to modify the screening solution set according to the space constraint condition and the symbol constraint condition to obtain a modified solution set; an identification result obtaining unit, the identification result obtaining unit being used to obtain the initial segment identification result through the correction scheme set; A voiceprint feature collection unit, which is used to collect voiceprint feature information of participants; A sound feature recognition unit, the sound feature recognition unit is used to recognize the sound features of the screen recording video through the voiceprint feature information, and generate personnel voiceprint identification information based on the recognition result; A facial feature collection unit, the facial feature collection unit is used to collect facial feature information of the participants, wherein the facial feature information has a one-to-one correspondence with the voiceprint feature information; An auxiliary identification unit, the auxiliary identification unit is used to construct the character feature identification data through the facial feature information, and perform auxiliary identification of the character feature identification data based on the person's voiceprint identification information to obtain an auxiliary identification result; A time node identification processing unit, the time node identification processing unit is used to perform time node identification processing on the screen recording video using the auxiliary identification result; A voiceprint quantity constraint unit, wherein the voiceprint quantity constraint unit is used to construct a voiceprint quantity constraint value; A voiceprint quantity determination unit, the voiceprint quantity determination unit is used to determine whether the voiceprint quantity information at the same time node meets the voiceprint quantity constraint value; A time node obtaining unit, wherein when the voiceprint quantity information at the same time node can satisfy the voiceprint quantity constraint value, the time node obtaining unit is used to form a time interval based on the time nodes that can satisfy the voiceprint quantity constraint value; A discussion identification unit, the discussion identification unit is used to identify the conference discussion of the screen recording video according to the time interval; A personnel authority obtaining unit, the personnel authority obtaining unit is used to obtain the personnel authority information of the person viewing the screen recording video; A feedback identification data obtaining unit, the feedback identification data obtaining unit is used to obtain the feedback identification data of the viewing personnel; An identification adjustment unit, wherein the identification adjustment unit is used to adjust the identification of the screen recording video according to the personnel authority information and the feedback identification data.

Images