Digital anchor virtual image generation method and system based on big data artificial intelligence

Abstract

The application relates to a digital anchor virtual image generation method and system based on big data artificial intelligence, and relates to the field of computer technology. The method comprises the following steps: obtaining news reporting text and reporting video, performing word segmentation on the reporting text to obtain a plurality of first word combinations of the reporting text, performing text recognition and feature recognition on the reporting video to obtain second word combinations and features in the reporting video, determining the news field to which the news belongs based on the first word combinations, the second word combinations and the features, determining the dressing of a reference digital anchor based on the news field and mapping the dressing to the reference digital anchor to obtain a target digital anchor, determining the position of the target digital anchor based on the reporting video, and mapping the target digital anchor to the position in the reporting video. The application has the effect of reducing information display loss caused by the shielding of the digital anchor.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of computer technology, in particular to a digital anchor virtual image generation method and system based on big data artificial intelligence. BACKGROUND

[0002] With the development of artificial intelligence, AI creation and other technologies, news reporting through virtual images and digital anchors in the digital media field has emerged as the times require, so that news reporting no longer needs to be prepared by real people to go out and host, reducing labor costs and reporting errors.

[0003] However, when a digital anchor reports news, the position of the digital anchor in the news video is usually set by relevant personnel, but the image of the digital anchor will block part of the news video, resulting in a lack of information display in the news video. Therefore, how to reduce the lack of information display caused by the blocking of the digital anchor has become a problem. SUMMARY

[0004] In order to reduce the lack of information display caused by the blocking of the digital anchor, the present application provides a digital anchor virtual image generation method and system based on big data artificial intelligence.

[0005] In a first aspect, the present application provides a digital anchor virtual image generation method based on big data artificial intelligence, which adopts the following technical solution:

[0006] The digital anchor virtual image generation method based on big data artificial intelligence comprises:

[0007] Obtaining the reporting text and the reporting video of the news;

[0008] Performing word segmentation processing on the reporting text to obtain a plurality of first word combinations of the reporting text;

[0009] Performing text recognition and feature recognition on the reporting video to obtain second word combinations and features in the reporting video;

[0010] Determining the news field to which the news belongs based on the first word combinations, the second word combinations and the features;

[0011] Determining the dress of a reference digital anchor based on the news field to which the news belongs and mapping the dress to the reference digital anchor to obtain a target digital anchor;

[0012] Determining the position of the target digital anchor based on the reporting video and mapping the target digital anchor to the position in the reporting video.

[0013] By adopting the technical solutions, the broadcast text and the broadcast video are convenient for subsequent analysis. The broadcast text is subjected to word segmentation to obtain a plurality of first word combinations constituting the broadcast text. The broadcast video is subjected to character recognition and feature recognition to obtain a second word combination in the broadcast video. The first word combination, the second word combination, and the feature in the broadcast video are all key factors reflecting the news field to which the news belongs. Therefore, the news field to which the news belongs can be accurately determined according to the first word combination, the second word combination, and the feature. After the required news field is determined, the clothing on the reference digital anchor is determined, so that the image of the digital anchor is more suitable for the news field. Then, the clothing is mapped to the reference digital anchor to obtain a target digital anchor. Since the pictures at different positions in the broadcast video are different, the position of the target digital anchor on the broadcast video is determined according to the broadcast video, that is, the position with the least information loss in the news broadcast. The target digital anchor is mapped to the determined position, and finally the effect of reducing the information display loss caused by the shielding of the digital anchor is realized.

[0014] In another possible implementation manner, each news field corresponds to a preset word library and a preset feature library. The news field to which the news belongs is determined based on the first word combination, the second word combination, and the feature, and includes the following steps.

[0015] A first quantity of the plurality of first word combinations and the second word combination in each preset word library is determined.

[0016] A second quantity of the plurality of features in each preset feature library is determined.

[0017] A plurality of target data sets are determined. Each target data set is each second word combination and a feature in a picture of the broadcast video at a time point at which each second word combination appears.

[0018] A news field hit by each target data set and a third quantity of a number of each news field hit by each target data set are determined, and a news field with the largest third quantity is determined as a preferred field of each target data set.

[0019] The news field to which the news belongs is determined based on the first quantity, the second quantity, and the preferred field.

[0020] In another possible implementation manner, the news field to which the news belongs is determined based on the first quantity, the second quantity, and the preferred field, and includes the following steps.

[0021] The first quantity is sorted to obtain a target preset word library with the largest first quantity.

[0022] The second quantity is sorted to obtain a target preset feature library with the largest second quantity.

[0023] The target preference field of all target data sets is summarized to obtain the target preference field with the highest occurrence frequency;

[0024] If there are at least two same fields in the news field of the target preset word library, the news field of the target preset feature library, and the target preference field, the at least two same fields are determined as the news field of the news.

[0025] In another possible implementation manner, the position of the target digital anchor is determined based on the broadcast video, including:

[0026] Edge detection is performed on the target digital anchor to obtain a contour of the target digital anchor;

[0027] The contour is mapped to a starting position at the lower left corner of the broadcast video, and the contour is horizontally translated according to a preset step size to obtain a plurality of candidate regions, the candidate regions including a region where the contour is located at the starting position;

[0028] The broadcast video is segmented according to each candidate region to obtain a region video of each candidate region;

[0029] Each frame of picture of each region video is determined, a first information entropy of each frame of picture and a similarity between adjacent pictures are calculated;

[0030] A target region is drawn with the contour as the center and according to a preset width, and each frame of picture of the broadcast video is segmented according to the target region to obtain a target region picture of each candidate region;

[0031] A pixel difference value of the same position between two adjacent frames of picture is determined to obtain an absolute difference graph, and a difference entropy of each absolute difference graph is calculated;

[0032] A first importance degree of each candidate region is determined based on the first information entropy, the difference entropy, and the similarity;

[0033] A second information entropy of each target region picture of each region video is calculated, and a gray value of a pixel on both sides of the contour is determined;

[0034] A second importance degree of each candidate region is determined based on the second information entropy and the gray value;

[0035] A total importance degree is determined based on the first importance degree and the second importance degree, and a candidate region with the lowest total importance degree is determined as the position of the target digital anchor.

[0036] In another possible implementation manner, the first importance degree of each candidate region is determined based on the first information entropy, the difference entropy, and the similarity, including:

[0037] calculating a first average value of information entropy of each regional video and a difference entropy average value of each absolute difference map;

[0038] calculating a second average value of similarity of adjacent frames of each regional video and a similarity variance;

[0039] determining a first importance degree based on the first average value, the difference entropy average value, the second average value, the similarity variance and respective corresponding coefficients.

[0040] In another possible implementation manner, the determining the second importance degree of each candidate region based on the second information entropy and the gray value includes:

[0041] dividing each target region frame of each regional video into a plurality of sub-region frames;

[0042] determining a third average value of respective corresponding gray values of each sub-region frame on both sides of the contour and determining a difference value of the third average value;

[0043] summing up the difference value of each sub-region frame to obtain a total value of each target region frame with respect to the difference value;

[0044] determining a fourth average value of each candidate region with respect to the total value based on the total value of each sub-region frame, determining a fifth average value of each candidate region with respect to the second information entropy based on the second information entropy of each sub-region frame, and determining the second importance degree of each candidate region based on the fourth average value, the fifth average value and respective corresponding coefficients.

[0045] In another possible implementation manner, each news field corresponds to a plurality of preset dressing combinations, and each preset dressing combination corresponds to a plurality of labels, and the determining the dressing of the reference digital anchor based on the news field includes:

[0046] determining a fourth number of labels of each preset dressing combination hit by the first word combination, the second word combination and the feature;

[0047] determining the preset dressing combination with the largest fourth number as the dressing of the reference digital anchor.

[0048] In a second aspect, the present application provides a digital anchor virtual image generation system based on big data artificial intelligence, which adopts the following technical scheme:

[0049] The digital anchor virtual image generation system based on big data artificial intelligence includes:

[0050] an acquisition module configured to acquire a broadcast text and a broadcast video of news;

[0051] The first processing module is configured to perform word segmentation processing on the broadcast text to obtain a plurality of first word combinations of the broadcast text.

[0052] The second processing module is configured to perform text recognition and feature recognition on the broadcast video to obtain second word combinations and features in the broadcast video.

[0053] The field determining module is configured to determine a news field of the news based on the first word combinations, the second word combinations and the features.

[0054] The dressing mapping module is configured to determine a dressing of a reference digital anchor based on the news field of the news and map the dressing to the reference digital anchor to obtain a target digital anchor.

[0055] The position mapping module is configured to determine a position of the target digital anchor based on the broadcast video and map the target digital anchor to the position in the broadcast video.

[0056] By using the above technical solutions, the acquisition module acquires the broadcast text and the broadcast video to facilitate subsequent analysis. The first processing module performs word segmentation processing on the broadcast text to obtain a plurality of first word combinations constituting the broadcast text. The second processing module performs text recognition and feature recognition on the broadcast video to obtain second word combinations in the broadcast video. The first word combinations, the second word combinations and the features in the broadcast video are all key factors reflecting the news field of the news. Therefore, the field determining module can accurately determine the news field of the news based on the first word combinations, the second word combinations and the features. After determining the required news field, the dressing mapping module determines the dressing on the reference digital anchor to make the image of the digital anchor more consistent with the news field. Then, the dressing is mapped to the reference digital anchor to obtain the target digital anchor. Since the pictures of different positions in the broadcast video are different, the position mapping module determines the position of the target digital anchor in the broadcast video based on the broadcast video, i.e., the position with the least information loss in the news broadcast, and maps the target digital anchor to the determined position. Finally, the effect of reducing information display loss caused by the occlusion of the digital anchor is achieved.

[0057] In another possible implementation, each news field corresponds to a preset word library and a preset feature library. When determining the news field of the news based on the word combinations and the features, the field determining module is specifically configured to:

[0058] determine a first number of the plurality of first word combinations and the second word combinations in each preset word library;

[0059] determine a second number of the plurality of features in each preset feature library;

[0060] determining a plurality of target data sets, each target data set being for each second word combination and a feature in a frame of the broadcast video at a time point at which the second word combination appears;

[0061] determining a third quantity of a news field hit by each target data set and a number of times each news field is hit, and determining a news field with a largest third quantity as a preferred field of each target data set;

[0062] determining a news field to which the news belongs based on the first quantity, the second quantity and the preferred field.

[0063] In another possible implementation, when the field determining module determines the news field to which the news belongs based on the first quantity, the second quantity and the preferred field, the field determining module is specifically configured to:

[0064] sorting the first quantity to obtain a target preset word library with a largest first quantity;

[0065] sorting the second quantity to obtain a target preset feature library with a largest second quantity;

[0066] summarizing the preferred fields of all target data sets to obtain a target preferred field with a largest number of occurrences;

[0067] if there are at least two same fields in the target preset word library, the target preset feature library and the target preferred field, determining the at least two same fields as the news field to which the news belongs.

[0068] In another possible implementation, when the position mapping module determines the position of the target digital anchor based on the broadcast video, the position mapping module is specifically configured to:

[0069] performing edge detection on the target digital anchor to obtain an outline of the target digital anchor;

[0070] mapping the outline to a starting position at a lower left corner of the broadcast video, and performing horizontal translation on the outline according to a preset step size to obtain a plurality of candidate regions, the candidate regions including a region in which the outline is located when the outline is at the starting position;

[0071] segmenting the broadcast video according to each candidate region to obtain a region video of each candidate region;

[0072] determining each frame of picture of each region video, calculating a first information entropy of each frame of picture and a similarity between adjacent frames of picture;

[0073] drawing a target region with the outline as a center and according to a preset width, and segmenting each frame of picture of the broadcast video according to the target region to obtain a target region picture of each candidate region;

[0074] determining a pixel difference value between the same position of two adjacent frames of pictures to obtain an absolute difference graph, and calculating a difference entropy of each absolute difference graph;

[0075] determining a first importance degree of each candidate region based on the first information entropy, the difference entropy and the similarity;

[0076] calculating a second information entropy of each target region picture of each region video and determining a gray value of a pixel on both sides of the contour;

[0077] determining a second importance degree of each candidate region based on the second information entropy and the gray value;

[0078] determining a total importance degree based on the first importance degree and the second importance degree, and determining a candidate region with the lowest total importance degree as the position of the target digital anchor.

[0079] In another possible implementation, when the position mapping module determines the first importance degree of each candidate region based on the first information entropy, the difference entropy and the similarity, the position mapping module is specifically configured to:

[0080] calculating a first average value of the information entropy of each region video and a difference entropy average value of each absolute difference graph;

[0081] calculating a second average value of the similarity of adjacent pictures of each region video and a similarity variance;

[0082] determining the first importance degree based on the first average value, the difference entropy average value, the second average value, the similarity variance and respective corresponding coefficients.

[0083] In another possible implementation, when the position mapping module determines the second importance degree of each candidate region based on the second information entropy and the gray value, the position mapping module is specifically configured to:

[0084] dividing each target region picture of each region video into a plurality of sub-region pictures;

[0085] determining a third average value of respective corresponding gray values of each sub-region picture on both sides of the contour and determining a difference value of the third average value;

[0086] summing up the difference value of each sub-region picture to obtain a total sum value of each target region picture with respect to the difference value;

[0087] determine a fourth average value of each candidate region with respect to the total value based on the total value of each sub-region picture, determine a fifth average value of each candidate region with respect to the second information entropy based on the second information entropy of each sub-region picture, and determine the second importance of each candidate region based on the fourth average value, the fifth average value and the respective corresponding coefficient.

[0088] In another possible implementation, each news field corresponds to a plurality of preset dressing combinations, each preset dressing combination corresponds to a plurality of labels, and the dressing mapping module, when determining the dressing of the reference digital anchor based on the news field, is specifically configured to:

[0089] determine a fourth number of labels of each preset dressing combination hit by the first word combination, the second word combination and the feature;

[0090] determine the preset dressing combination with the largest fourth number as the dressing of the reference digital anchor.

[0091] In a third aspect, the present application provides an electronic device, which adopts the following technical solution:

[0092] An electronic device, comprising:

[0093] at least one processor;

[0094] a memory;

[0095] at least one application program, wherein the at least one application program is stored in the memory and is configured to be executed by the at least one processor, and the at least one processor is configured to execute the method for generating a digital anchor virtual image based on big data artificial intelligence according to any one of the possible implementation manners of the first aspect.

[0096] In a fourth aspect, the present application provides a computer readable storage medium, which adopts the following technical solution:

[0097] A computer readable storage medium, when the computer program is executed in the computer, the computer executes the method for generating a digital anchor virtual image based on big data artificial intelligence according to any one of the first aspect.

[0098] In summary, the present application includes at least one of the following beneficial technical effects:

[0099] The acquisition of the broadcast text and the broadcast video facilitates subsequent analysis. The broadcast text is subjected to word segmentation to obtain a plurality of first word combinations constituting the broadcast text. The broadcast video is subjected to character recognition and feature recognition to obtain second word combinations in the broadcast video. The first word combinations, the second word combinations, and the features in the broadcast video are all key factors reflecting the news field of the news. Therefore, the news field of the news can be accurately determined according to the first word combinations, the second word combinations, and the features. After the required news field is determined, the clothing on the reference digital anchor is determined, so that the image of the digital anchor is more suitable for the news field. Then, the clothing is mapped to the reference digital anchor to obtain a target digital anchor. Since the pictures at different positions in the broadcast video are different, the position of the target digital anchor on the broadcast video is determined according to the broadcast video, that is, the position with the least information loss in the news broadcast. The target digital anchor is mapped to the determined position, and finally the effect of reducing the information display loss caused by the occlusion of the digital anchor is realized. BRIEF DESCRIPTION OF DRAWINGS

[0100] Figure 1 FIG. 1 is a flow diagram of a digital anchor virtual image generation method based on big data artificial intelligence according to an embodiment of the present application.

[0101] Figure 2 FIG. 2 is a structural diagram of a digital anchor virtual image generation system based on big data artificial intelligence according to an embodiment of the present application.

[0102] Figure 3 FIG. 3 is a structural diagram of an electronic device according to an embodiment of the present application. DETAILED DESCRIPTION

[0103] The present application will be further described below in conjunction with the accompanying drawings.

[0104] Those skilled in the art can make modifications to the present embodiments without creative contribution after reading the present specification, but as long as the modifications are within the scope of the claims of the present application, they are protected by the patent law.

[0105] To make the purpose, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be described clearly and completely below in conjunction with the drawings of the embodiments of the present application. Obviously, the described embodiments are some of the embodiments of the present application, not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0106] In addition, the term "and / or" in this document merely describes an association relationship of associated objects, which means that there can be three relationships, for example, A and / or B can represent three cases of existence of A alone, existence of A and B simultaneously, and existence of B alone. In addition, the character " / " in this document generally represents an "or" relationship between the front and rear associated objects unless otherwise specified.

[0107] The embodiments of the present application will be further described in detail below with reference to the accompanying drawings.

[0108] The embodiments of the present application provide a digital anchor virtual image generation method based on big data artificial intelligence, which is executed by an electronic device. The electronic device can be a server or a terminal device. The server can be a stand-alone physical server, a server cluster composed of multiple physical servers or a distributed system, or a cloud server providing cloud computing services. The terminal device can be a smart phone, a tablet computer, a notebook computer, a desktop computer, etc., but is not limited thereto. The terminal device and the server can be directly or indirectly connected through wired or wireless communication, and the embodiments of the present application do not limit this. As shown in the figure, the method comprises steps S101, S102, S103, S104, S105 and S106, wherein, Figure 1

[0109] S101, obtaining a broadcast text and a broadcast video of news.

[0110] For the embodiments of the present application, the broadcast text and the broadcast video can be edited and produced in advance by staff and then input into the electronic device so that the electronic device obtains the broadcast text and the broadcast video.

[0111] S102, performing word segmentation processing on the broadcast text to obtain a plurality of first word combinations of the broadcast text.

[0112] For the embodiments of the present application, the electronic device performs word segmentation processing according to a hidden Markov model, a conditional random field, a deep learning model, etc. to obtain a plurality of first word combinations, or performs word segmentation processing on the broadcast text by using jieba, HanLP, etc. plug-in tools to obtain a plurality of first word combinations.

[0113] S103, performing text recognition and feature recognition on the broadcast video to obtain a second word combination and a feature in the broadcast video.

[0114] ​For the embodiment of the present application, there is audio in the broadcast video, so the electronic device parses the broadcast video to obtain the audio file in the broadcast video, and then inputs the audio file into the network model for audio-to-text to perform text recognition to obtain the text information corresponding to the audio in the broadcast video, and then obtains the second word combination in the broadcast video through the method recorded in step S102. The electronic device inputs each frame of picture of the broadcast video into the trained network model for feature recognition, thereby obtaining the features appearing in the broadcast video.

[0115] S104, determining the belonging news field of the news based on the first word combination, the second word combination and the features.

[0116] For the embodiment of the present application, the first word combination, the second word combination and the features appearing in the broadcast video are all key factors describing the news field required by the news, so the electronic device can accurately determine the required news field of the news by combining the first word combination, the second word combination and the features. The news field includes international, sports, entertainment, automobile, film, etc.

[0117] S105, determining the dressing of the reference digital anchor based on the belonging news field and mapping the dressing to the reference digital anchor to obtain the target digital anchor.

[0118] For the embodiment of the present application, in order to make the image of the digital anchor more suitable for the required news field, the electronic device determines the dressing of the reference digital anchor according to the required news field, the reference digital anchor is only a human body model and a basic clothing attached to the human body model, the basic clothing does not belong to any news field, and after the electronic device determines the dressing of the reference digital anchor, the target digital anchor can be obtained by mapping the dressing to the reference digital anchor, the target digital anchor is more suitable for the field of the news, thereby improving the viewing experience and effect.

[0119] S106, determining the position of the target digital anchor based on the broadcast video and mapping the target digital anchor to the position in the broadcast video.

[0120] For the embodiment of the present application, since the information amount and picture change of each part in the broadcast video are different, the electronic device determines the position of the target digital anchor in the broadcast video according to the broadcast video, which is the position where the target digital anchor blocks the broadcast video with the least missing of the overall broadcast video information display and the least broadcast impact, and then maps the target digital anchor to the position, thereby realizing reducing the information display missing caused by the blocking of the digital anchor.

[0121] In a possible implementation of the embodiment, each news field corresponds to a preset vocabulary library and a preset feature library, and the news field to which the news belongs is determined based on the first vocabulary combination, the second vocabulary combination and the features in step S104. Specifically, the determination includes steps S1041 (not shown in the figure), S1042 (not shown in the figure), S1043 (not shown in the figure), S1044 (not shown in the figure) and S1045 (not shown in the figure), wherein

[0122] In S1041, the first quantity of the first vocabulary combination and the second vocabulary combination in each preset vocabulary library is determined.

[0123] For the embodiment, the first vocabulary combination and the second vocabulary combination are both descriptions of the news in the aspect of words, and the words in the preset vocabulary library corresponding to each news field are classified and used to describe the news field by staff. Therefore, the electronic device fuses the first vocabulary combination and the second vocabulary combination, and then matches and counts all the vocabulary combinations with the preset vocabulary library of each news field, so as to determine the first quantity of all the vocabulary combinations in each preset vocabulary library. The greater the first quantity corresponding to a preset vocabulary library is, the greater the possibility that the news belongs to the news field of the preset vocabulary library is.

[0124] In S1042, the second quantity of the features in each preset feature library is determined.

[0125] For the embodiment, the features in the preset feature library corresponding to each news field are also classified and used to describe the news field by staff. The electronic device can determine the second quantity of the features in each preset feature library in the manner described in step S1041. The greater the second quantity corresponding to a preset feature library is, the greater the possibility that the news belongs to the news field of the preset feature library is.

[0126] In S1043, a plurality of target data sets are determined.

[0127] Each target data set is each second vocabulary combination and the features in the picture of the broadcast video at the time point of the appearance of each second vocabulary combination.

[0128] For the embodiment, the electronic device determines the time point of the appearance of the second vocabulary combination in the broadcast video and the picture corresponding to the time point. The features in the picture at the same time point have a strong correlation with the second vocabulary combination at the time point, and therefore the electronic device determines a plurality of target data sets.

[0129] S1044, determine a third quantity of the number of each news field hit by each target data set and the number of each news field hit, and determine the news field with the largest third quantity as the preferred field of each target data set.

[0130] For the embodiments of the present application, if the second word combination and the feature hit in a certain target data set are in the same news field, that is, the number of news fields hit by the target data set is 1, it indicates that the target data set of the picture reacts the possibility of the news belonging to the hit news field is larger, if the number of news fields hit is greater than 1, it indicates that the target data set of the picture reacts the news may belong to different news fields, therefore, the number of each news field hit by the second word combination and the feature in the target data set is determined, that is, the third quantity, the news field with the largest third quantity is the news field to which the target data set of the picture reacts the news belonging to the news field with the largest possibility, that is, the preferred field. If there are at least two news fields with the same third quantity in the multiple news fields hit, the electronic device determines the preferred field as the news field with the same third quantity.

[0131] S1045, determine the news field to which the news belongs based on the first quantity, the second quantity and the preferred field.

[0132] For the embodiments of the present application, in summary, the first quantity, the second quantity and the preferred field are all key factors representing the real news field to which the news belongs, therefore, the electronic device can accurately determine the required news field of the news by combining the first quantity, the second quantity and the preferred field.

[0133] In one possible implementation of the embodiments of the present application, the step of determining the news field to which the news belongs based on the first quantity, the second quantity and the preferred field in step S1045 specifically includes step S1 (not shown in the figure), step S2 (not shown in the figure), step S3 (not shown in the figure) and step S4 (not shown in the figure), wherein,

[0134] S1, sort the first quantity to obtain the target preset word library with the largest first quantity.

[0135] For the embodiments of the present application, the electronic device sorts the first quantity from large to small to obtain the target preset word library that matches the first word combination and the second word combination most, the number of the first word combination and the second word combination hit in the target preset word library is the largest, which indicates that the news field corresponding to the target preset word library in the first word combination and the second word combination is most consistent with the real field to which the news belongs.

[0136] S2, sort the second quantity to obtain the target preset feature library with the largest second quantity.

[0137] For the embodiment of the present application, the electronic device sorts the second quantity from large to small, obtains the target preset feature library with the most matched feature aspect, and the target preset feature library has the most matched features, which indicates that the target preset word library corresponding to the feature in the broadcast video is most consistent with the true belonging field of the news.

[0138] S3, the preference field of the entire target data set is summarized to obtain the target preference field with the most occurrences.

[0139] For the embodiment of the present application, the electronic device counts the preference field of the entire target data set, determines the number of occurrences of each preference field, and then sorts the number of occurrences to obtain the target preference field with the most occurrences. The target preference field is the most consistent with the true belonging field of the news in the target data set (strong correlation between the second word combination and the feature in the same moment picture).

[0140] S4, if the news field of the target preset word library, the news field of the target preset feature library, and the target preference field have at least two same fields, the at least two same fields are determined as the belonging news field of the news.

[0141] For the embodiment of the present application, if there are at least two same fields, it means that a certain news field accounts for the majority of the situation, so the belonging news field of the news is determined as the majority of the same news field. If the above three fields are not the same, the corresponding news field of the highest priority of the above target preset word library, target preset feature library, and preference field is determined as the belonging news field of the news. The priority of the above three fields can be set by the staff according to the demand or the actual situation. The belonging news field determined by the first word combination, the second word combination, and the feature is more accurate.

[0142] In one possible implementation of the embodiment of the present application, the step S106 of determining the position of the target digital anchor based on the broadcast video includes steps S1061 (not shown in the figure), S1062 (not shown in the figure), S1063 (not shown in the figure), S1064 (not shown in the figure), S1065 (not shown in the figure), S1066 (not shown in the figure), S1067 (not shown in the figure), S1068 (not shown in the figure), S1069 (not shown in the figure), and S10610 (not shown in the figure), wherein,

[0143] S1061, edge detection is performed on the target digital anchor to obtain the contour of the target digital anchor.

[0144] For the embodiments of the present application, the electronic device acquires the image of the target digital anchor, and then uses an edge detection algorithm, such as Canny edge detection, to determine the edge by finding points with sharp changes in gray intensity in the image.

[0145] S1062, map the contour to a starting position at the lower left corner of the broadcast video, and perform horizontal translation of the contour by a preset step size to obtain a plurality of candidate regions.

[0146] The candidate region includes a region in which the contour is located when the contour is located at the starting position.

[0147] For the embodiments of the present application, the digital anchor is usually placed in the lower region of the broadcast video, so the electronic device maps the contour of the digital anchor to the starting position at the lower left corner of the broadcast video, and the preset step size can be a specified number of pixels or a distance, and then the electronic device controls the contour to perform horizontal translation to the right in the broadcast video by the preset step size, thereby determining a plurality of candidate regions of the contour of the digital anchor on the broadcast video.

[0148] S1063, segment the broadcast video according to each candidate region to obtain a region video of each candidate region.

[0149] For the embodiments of the present application, after the electronic device determines the candidate regions, the broadcast video is segmented according to the candidate regions, thereby obtaining a region video of each candidate region.

[0150] S1064, determine each frame of picture of each region video, and calculate the first information entropy of each frame of picture and the similarity between adjacent pictures.

[0151] For the embodiments of the present application, the electronic device can read the video frame by frame through the library function, thereby determining each frame of picture of each region video. Then the electronic device calculates the first information entropy of each frame of picture using the information entropy calculation formula. Specifically, the electronic device traverses all pixels of each frame of picture, counts the number of occurrences of each pixel value, and calculates the probability thereof, and according to the probability of each pixel value, uses the formula H(X) = -sum(p(x) x log2(p(x))) to calculate the information entropy, wherein H(X) is the information entropy, and p(x) is the probability of the pixel value. The electronic device calculates the information entropy of each color channel of each frame of picture, and then adds them to obtain the first information entropy of each frame of picture. The greater the first information entropy in the picture, the more information and uncertainty the picture contains, and the greater the amount of information, and the more serious the information loss when the digital anchor blocks the region.

[0152] The electronic device can input adjacent pictures of the region video into the trained network model for similarity calculation, represent the adjacent pictures in vector form, and then calculate the cosine distance between the vectors to determine the similarity between the adjacent pictures. The higher the similarity, the less drastic the change in the pictures of the region video, and the more stable the information change. When the digital anchor blocks the region, the information loss is less.

[0153] S1065, drawing the target region centered on the contour and according to a preset width, and segmenting each frame picture of the broadcast video according to the target region to obtain a target region picture of each candidate region.

[0154] For the embodiments of the present application, the preset width can be a specified number of pixels. The electronic device extends the contour by a number of pixels on both sides of the contour according to the preset width, so that the number of pixels on both sides of the contour is consistent, that is, the width on both sides of the contour is consistent. The electronic device segments the broadcast video in this way to obtain the target region, and further obtains each frame picture of the target region, that is, the target region picture.

[0155] S1066, determining the pixel difference value of the same position between adjacent two frame pictures to obtain an absolute difference graph, and calculating the difference entropy of each absolute difference graph.

[0156] For the embodiments of the present application, the electronic device determines the adjacent pictures of the region video and calculates the pixel difference value of the same position of the adjacent pictures to obtain the absolute difference graph. A new difference graph image, that is, the absolute difference graph, is generated by calculating the absolute difference value between the corresponding pixels of the two images. In the difference graph image, the area with a higher pixel value represents a larger difference between the original images. Then, the electronic device can determine the difference entropy of each absolute difference graph by using the information entropy calculation formula. The larger the difference entropy, the greater the change between the adjacent pictures, and further, the more unstable the information change. The blocking of the digital anchor will result in a more serious information display loss.

[0157] S1067, determining the first importance of each candidate region based on the first information entropy, the difference entropy, and the similarity.

[0158] For the embodiments of the present application, as summarized above, the first information entropy, the difference entropy, and the similarity are all key factors representing the information amount and the information change degree in each region video. Therefore, the electronic device determines the first importance of each candidate region more accurately according to the first information entropy and the other three factors.

[0159] S1068, calculating the second information entropy of each target region picture of each candidate region and determining the gray value of the pixels on both sides of the contour.

[0160] For the embodiment of the present application, the second information entropy of the target region picture is also calculated using the information entropy calculation formula. The greater the second information entropy, the greater the amount of broadcast video information at the edge of the digital anchor, and the more important the information. The electronic device performs grayscale transformation on each target region picture to obtain the grayscale values of the pixels on both sides of the contour. The greater the difference in grayscale values, the richer the content of the target region picture, and the greater the change in information at the contour.

[0161] S1069, determining the second importance of each candidate region based on the second information entropy and the grayscale value.

[0162] For the embodiment of the present application, the second information entropy represents the amount of information of the video picture at the contour and the uncertainty randomness of the pixels, and the grayscale value represents the richness of the picture at the contour and the change of the pixels. Therefore, the second information entropy and the grayscale value are both key factors affecting the amount of video information at the contour and whether the information is important. Therefore, the electronic device determines the second importance of each region video based on the second information entropy and the grayscale value.

[0163] S10610, determining the total importance based on the first importance and the second importance, and determining the candidate region with the lowest total importance as the position of the target digital anchor.

[0164] For the embodiment of the present application, the first importance represents the overall information change and the amount of information of the region video (candidate region), and the second importance represents the information change and the amount of information of the video at the contour. Therefore, the electronic device can determine the total importance of each candidate region based on the first importance and the second importance. Specifically, the electronic device can sum the first importance and the second importance to obtain the total importance. The higher the total importance, the more important the information, the more intense the information change, and the greater the amount of information. Therefore, the electronic device can determine the candidate region with the lowest total importance as the position of the target digital anchor. Through the determination of the first importance and the second importance, the position of the target digital anchor is more accurately selected from the candidate regions.

[0165] In one possible implementation of the embodiment of the present application, the first importance of each candidate region is determined based on the first information entropy, the difference entropy, and the similarity in step S1067, specifically including steps Sa (not shown in the figure), step Sb (not shown in the figure), and step Sc (not shown in the figure), wherein,

[0166] Sa, calculating the first average value of the information entropy of each region video and the average value of the difference entropy of each absolute difference map.

[0167] For the embodiment of the present application, the information entropy of each region includes the information entropy of each frame of picture, so the electronic device determines the first average value through the average value calculation formula, and the greater the first average value is, the greater the information quantity of the region video as a whole is, the greater the randomness of the pixels is, and the more important the region video is. The difference entropy average value of the absolute difference graph is determined through the average value calculation formula, and the greater the difference entropy average value is, the greater the pixel change of the region video is, the more important the information is, and the more important the region video is.

[0168] Sb, a second average value of the similarity of adjacent pictures of each region video and a similarity variance are calculated.

[0169] For the embodiment of the present application, the electronic device calculates the second average value of the similarity through the average value calculation formula, and determines the similarity variance using the variance calculation formula. The greater the second average value is, the closer the adjacent pictures are, the smaller the change degree of the picture over time is, the smaller the information change is, and the less important the region video is. The greater the variance is, the more violent the picture change is, the greater the information change is, and the more important the region video is.

[0170] Sc, a first importance degree is determined based on the first average value, the difference entropy average value, the second average value, the similarity variance, and respective corresponding coefficients.

[0171] For the embodiment of the present application, in summary, the first average value, the difference entropy average value, the second average value, and the similarity variance are all key factors affecting the importance degree of the to-be-selected region, so the staff sets respective corresponding coefficients for the above four factors and stores them in the electronic device, the electronic device calls the respective corresponding coefficients and performs weighted calculation on the above four factors to obtain a score, and the score represents the first importance degree. The first importance degree obtained through comprehensive calculation and analysis of the four factors is more accurate.

[0172] In one possible implementation manner of the embodiment of the present application, the second importance degree of each to-be-selected region is determined based on the second information entropy and the gray value in step S1069, and specifically includes the following steps one, two, three, and four.

[0173] Step one, each target region picture of each region video is divided into a plurality of sub-region pictures.

[0174] For the embodiment of the present application, the electronic device can divide the target region picture into a plurality of sub-region pictures along the contour according to a preset length.

[0175] Step two, a third average value of respective corresponding gray values of each sub-region picture on both sides of the contour is determined, and a difference value of the third average value is determined.

[0176] For the embodiment of the present application, the electronic device determines the third average value of the respective corresponding gray scale values of each sub-region picture on both sides of the contour through an average value calculation formula, and then obtains the difference value by subtracting the two third average values. The greater the difference value, the greater the change of the pixels on both sides of the contour, the greater the information quantity, and the more important the sub-region picture.

[0177] Step three, summing up the difference values of each sub-region picture to obtain the total sum value of each target region picture with respect to the difference value.

[0178] For the embodiment of the present application, the electronic device sums up the difference values of each sub-region picture to obtain the total sum value representing the information change of the target region picture on both sides of the contour.

[0179] Step four, determining the fourth average value of each candidate region with respect to the total sum value based on the total sum value of each sub-region picture, determining the fifth average value of each candidate region with respect to the second information entropy based on the second information entropy of each sub-region picture, and determining the second importance degree of each candidate region based on the fourth average value, the fifth average value, and the respective corresponding coefficients.

[0180] For the embodiment of the present application, the electronic device averages all the second information entropies, i.e., the fifth average value, to represent the information quantity and the uncertainty of the pixels of each candidate region at the contour of the video. In summary, the fourth average value and the fifth average value are both key factors affecting whether each candidate region is important at the contour of the video. The respective coefficients of the fourth average value and the fifth average value are set by the staff and stored in the electronic device, and then the electronic device calls the respective corresponding coefficients to perform weighted calculation on the fourth average value and the fifth average value to obtain a score, which represents the second importance degree. The higher the second importance degree, the more important the information at the contour, and the less suitable it is as the position of the target digital anchor. The determination of the second importance degree through the comprehensive determination of the fourth average value and the fifth average value is more accurate.

[0181] In one possible implementation of the embodiment of the present application, each news field corresponds to a plurality of preset dressing combinations, and each preset dressing combination corresponds to a plurality of labels. In step S105, the dressing of the reference digital anchor is determined based on the news field, specifically including steps five and six, wherein,

[0182] Step five, determining the fourth number of labels of each preset dressing combination hit by the first word combination, the second word combination, and the feature.

[0183] Step six, determining the preset dressing combination with the largest fourth number as the dressing of the reference digital anchor.

[0184] For the embodiments of the present application, each news field corresponds to a plurality of preset dressing combinations, and the labels of the preset dressing combinations are different, so the electronic device determines the first word combination, the second word combination, and the fourth number of the labels hit by the features of each preset dressing combination. The more the fourth number of a certain preset dressing combination is, the more the preset dressing combination fits the current news, and the higher the fitting degree with the news is, so the electronic device can determine the preset dressing combination with the largest fourth number as the dressing of the reference digital anchor.

[0185] The above embodiments introduce a digital anchor virtual image generation method based on big data artificial intelligence from the perspective of method flow, and the following embodiments introduce a digital anchor virtual image generation system based on big data artificial intelligence from the perspective of virtual modules or virtual units. For details, see the following embodiments.

[0186] The embodiments of the present application provide a digital anchor virtual image generation system 20 based on big data artificial intelligence, as shown in the figure, which specifically can include: Figure 2

[0187] The acquisition module 201 is configured to acquire the broadcast text and the broadcast video of the news.

[0188] The first processing module 202 is configured to perform word segmentation processing on the broadcast text to obtain a plurality of first word combinations of the broadcast text.

[0189] The second processing module 203 is configured to perform text recognition and feature recognition on the broadcast video to obtain the second word combination and the features in the broadcast video.

[0190] The field determination module 204 is configured to determine the news field to which the news belongs based on the first word combination, the second word combination, and the features.

[0191] The dressing mapping module 205 is configured to determine the dressing of the reference digital anchor based on the news field and map the dressing to the reference digital anchor to obtain the target digital anchor.

[0192] The position mapping module 206 is configured to determine the position of the target digital anchor based on the broadcast video and map the target digital anchor to the position in the broadcast video.

[0193] ​The embodiment of the application discloses a digital anchor virtual image generation system 20 based on big data artificial intelligence, wherein a obtaining module 201 obtains reporting text and reporting video for subsequent analysis, a first processing module 202 performs word segmentation processing on the reporting text to obtain a plurality of first word combinations constituting the reporting text, a second processing module 203 performs text recognition and feature recognition on the reporting video to obtain a second word combination in the reporting video, the first word combination, the second word combination and the feature in the reporting video are all key factors reflecting the news field of the news, therefore, a field determining module 204 can accurately determine the news field of the news according to the first word combination, the second word combination and the feature, after determining the required news field, a dressing mapping module 205 determines the dressing on the reference digital anchor, so that the image of the digital anchor is more suitable for the news field, and then the dressing is mapped to the reference digital anchor to obtain a target digital anchor, since the picture changes at different positions in the reporting video are different, a position mapping module 206 determines the position of the target digital anchor on the reporting video according to the reporting video, that is, the position with the least information loss in the news reporting, and maps the target digital anchor to the determined position, finally, the effect of reducing the information display loss caused by the shielding of the digital anchor is realized.

[0194] In a possible implementation manner of the embodiment of the application, each news field corresponds to a preset word library and a preset feature library, and the field determining module 204, when determining the news field of the news based on the word combination and the feature, is specifically used for:

[0195] determining a first quantity of the plurality of first word combinations and the second word combination in each preset word library;

[0196] determining a second quantity of the plurality of features in each preset feature library;

[0197] determining a plurality of target data sets, each target data set being each second word combination and the feature in the picture of the reporting video at the appearance time point of each second word combination;

[0198] determining a third quantity of the news field hit by each target data set and the number of each news field hit by each target data set, and determining the news field with the largest third quantity as the preferred field of each target data set;

[0199] determining the news field of the news based on the first quantity, the second quantity and the preferred field.

[0200] In a possible implementation manner of the embodiment of the application, when the field determining module 204 determines the news field of the news based on the first quantity, the second quantity and the preferred field, the field determining module 204 is specifically used for:

[0201] The first quantity is sorted to obtain a target preset word library with the largest first quantity;

[0202] The second quantity is sorted to obtain a target preset feature library with the largest second quantity;

[0203] The preference fields of all target data sets are summarized to obtain a target preference field with the largest number of occurrences;

[0204] If there are at least two same fields in the news field of the target preset word library, the news field of the target preset feature library, and the target preference field, the at least two same fields are determined as the belonging news field of the news.

[0205] In a possible implementation of the embodiment of the application, the position mapping module 206 is specifically configured to:

[0206] Edge detection is performed on the target digital anchor to obtain a contour of the target digital anchor;

[0207] The contour is mapped to a starting position at the lower left corner of the broadcast video, and the contour is horizontally translated according to a preset step size to obtain a plurality of candidate regions, the candidate regions including a region where the contour is located at the starting position;

[0208] Each candidate region is used to segment the broadcast video to obtain a region video of each candidate region;

[0209] Each frame of picture of each region video is determined, and a first information entropy of each frame of picture and a similarity between adjacent pictures are calculated;

[0210] A target region is drawn with the contour as the center and according to a preset width, and each frame of picture of the broadcast video is segmented according to the target region to obtain a target region picture of each candidate region;

[0211] A pixel difference value of the same position between adjacent two frames of picture is determined to obtain an absolute difference graph, and a difference entropy of each absolute difference graph is calculated;

[0212] A first importance degree of each candidate region is determined based on the first information entropy, the difference entropy, and the similarity;

[0213] A second information entropy of each target region picture of each region video is calculated, and a gray value of a pixel on both sides of the contour is determined;

[0214] A second importance degree of each candidate region is determined based on the second information entropy and the gray value;

[0215] A total importance degree is determined based on the first importance degree and the second importance degree, and a candidate region with the lowest total importance degree is determined as the position of the target digital anchor.

[0216] In a possible implementation of the embodiment, when determining the first importance degree of each candidate region based on the first information entropy, the difference entropy and the similarity, the position mapping module 206 is specifically configured to:

[0217] calculate a first average value of the information entropy of each region video and a difference entropy average value of each absolute difference map;

[0218] calculate a second average value of the similarity of adjacent pictures of each region video and a similarity variance;

[0219] determine the first importance degree based on the first average value, the difference entropy average value, the second average value, the similarity variance and respective corresponding coefficients.

[0220] In a possible implementation of the embodiment, when determining the second importance degree of each candidate region based on the second information entropy and the gray value, the position mapping module 206 is specifically configured to:

[0221] divide each target region picture of each region video into a plurality of sub-region pictures;

[0222] determine a third average value of the respective corresponding gray values of each sub-region picture on both sides of the contour and determine a difference value of the third average value;

[0223] sum the difference values of each sub-region picture to obtain a sum value of each target region picture with respect to the difference value;

[0224] determine a fourth average value of each candidate region with respect to the sum value based on the sum value of each sub-region picture, determine a fifth average value of each candidate region with respect to the second information entropy based on the second information entropy of each sub-region picture, and determine the second importance degree of each candidate region based on the fourth average value, the fifth average value and respective corresponding coefficients.

[0225] In a possible implementation of the embodiment, each news field corresponds to a plurality of preset dressing combinations, each preset dressing combination corresponds to a plurality of labels, and when determining the dressing of the reference digital anchor based on the news field, the dressing mapping module 205 is specifically configured to:

[0226] determine a fourth number of labels of each preset dressing combination hit by the first word combination, the second word combination and the feature;

[0227] determine the preset dressing combination with the largest fourth number as the dressing of the reference digital anchor.

[0228] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the digital anchor virtual image generation system 20 based on big data artificial intelligence described above can refer to the corresponding process in the foregoing method embodiments, which will not be described here.

[0229] This application provides an electronic device, such as... Figure 3 As shown, Figure 3 The illustrated electronic device 30 includes a processor 301 and a memory 303. The processor 301 and the memory 303 are connected, for example, via a bus 302. Optionally, the electronic device 30 may also include a transceiver 304. It should be noted that in practical applications, the transceiver 304 is not limited to one type, and the structure of this electronic device 30 does not constitute a limitation on the embodiments of this application.

[0230] Processor 301 may be a CPU (Central Processing Unit), a general-purpose processor, a DSP (Digital Signal Processor), an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Gate Array), or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof. It can implement or execute the various exemplary logic blocks, modules, and circuits described in conjunction with the disclosure of this application. Processor 301 may also be a combination that implements computational functions, such as including one or more microprocessor combinations, a combination of a DSP and a microprocessor, etc.

[0231] Bus 302 may include a pathway for transmitting information between the aforementioned components. Bus 302 may be a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus, etc. Bus 302 can be divided into address bus, data bus, control bus, etc. For ease of representation, Figure 3 The symbol is represented by a single thick line, but this does not mean that there is only one bus or one type of bus.

[0232] The memory 303 can be a ROM (Read Only Memory) or other type of static storage device that can store static information and instructions, a RAM (Random Access Memory) or other type of dynamic storage device that can store information and instructions, an EEPROM (Electrically Erasable Programmable Read Only Memory), a CD-ROM (Compact Disc Read Only Memory) or other optical disk storage, a magnetic disk storage medium or other magnetic storage device, or any other medium that can be used to carry or store desired program code in the form of instructions or data structures and that can be accessed by a computer, but is not limited thereto.

[0233] The memory 303 is configured to store application program codes for implementing the solutions of the present application, and the processor 301 is configured to control the execution of the application program codes. The processor 301 is configured to execute the application program codes stored in the memory 303 to implement the content shown in the foregoing method embodiments.

[0234] The electronic device includes, but is not limited to, a mobile terminal such as a mobile phone, a notebook computer, a digital broadcast receiver, a PDA (Personal Digital Assistant), a PAD (Tablet Personal Computer), a PMP (Portable Multimedia Player), a car terminal (e.g., a car navigation terminal), and the like, and a stationary terminal such as a digital TV, a desktop computer, and the like. It can also be a server or the like. Figure 3 The electronic device shown is merely an example and should not impose any limitation on the functions and use range of the embodiments of the present application.

[0235] The computer readable storage medium provided in the embodiment of the present application stores a computer program, and when the computer program is run on a computer, the computer can execute the corresponding content in the foregoing method embodiment. Compared with the related art, the broadcast text and the broadcast video are acquired in the embodiment of the present application, which facilitates subsequent analysis. The broadcast text is subjected to word segmentation processing to obtain a plurality of first word combinations constituting the broadcast text, and the broadcast video is subjected to character recognition and feature recognition to obtain a second word combination in the broadcast video. The first word combination, the second word combination, and the feature in the broadcast video are all key factors reflecting the news field of the news. Therefore, the news field of the news can be accurately determined according to the first word combination, the second word combination, and the feature. After the required news field is determined, the clothing on the reference digital anchor is determined, so that the image of the digital anchor is more suitable for the news field. Then, the clothing is mapped to the reference digital anchor to obtain a target digital anchor. Since the picture changes at each position in the broadcast video are different, the position of the target digital anchor on the broadcast video is determined according to the broadcast video, that is, the position with the least information loss in the news broadcast. The target digital anchor is mapped to the determined position, and finally the effect of reducing the information display loss caused by the shielding of the digital anchor is realized.

[0236] It should be understood that, although each step in the flowchart of the accompanying drawings is shown in sequence according to the direction of the arrow, these steps are not necessarily executed in sequence according to the direction of the arrow. Unless otherwise specified herein, the execution of these steps is not strictly limited in sequence, and they can be executed in other sequences. Moreover, at least part of the steps in the flowchart of the accompanying drawings can include a plurality of sub-steps or stages, which are not necessarily executed at the same time, but can be executed at different times, and the execution sequence is not necessarily sequential, but can be executed in rotation or alternation with at least part of other steps or sub-steps or stages of other steps.

[0237] The above only describes some embodiments of the present application. It should be noted that, for those skilled in the art, without departing from the principles of the present application, a number of improvements and refinements can be made, which should also be considered as the protection scope of the present application.

Claims

1. A method for generating virtual avatars of digital anchors based on big data and artificial intelligence, characterized in that, include: Get the text and video of news broadcasts; The broadcast text is segmented into words to obtain multiple combinations of first words. The broadcast video is subjected to text recognition and feature recognition to obtain the second word combination and features in the broadcast video; The news field to which the news belongs is determined based on the first word combination, the second word combination, and the features; Based on the news field to which the target digital anchor is located, the clothing of the benchmark digital anchor is determined and the clothing is mapped onto the benchmark digital anchor to obtain the target digital anchor. Determine the fourth number of tags that match the first word combination, the second word combination, and the feature matching for each preset clothing combination; The fourth and most frequent preset clothing combination is determined as the clothing of the benchmark digital anchor; The location of the target digital anchor is determined based on the broadcast video, and the target digital anchor is mapped to the location in the broadcast video; Edge detection is performed on the target digital anchor to obtain the outline of the target digital anchor; The contour is mapped to the starting position of the lower left corner of the broadcast video, and the contour is translated laterally according to a preset step size to obtain multiple candidate areas. The candidate areas include the area where the contour is located when it is at the starting position. The broadcast video is segmented according to each candidate region to obtain the region video of each candidate region; Determine each frame of the video for each region, and calculate the first information entropy of each frame and the similarity between adjacent frames; The target area is drawn with the outline as the center and the width is preset. Each frame of the broadcast video is divided according to the target area to obtain the target area image of each candidate area. Determine the pixel difference at the same position between two adjacent frames to obtain an absolute difference map, and calculate the difference entropy of each absolute difference map; The first importance of each candidate region is determined based on the first information entropy, difference entropy, and similarity. Calculate the second information entropy of each target area frame in each region video and determine the grayscale values ​​of the pixels on both sides of the contour; The second importance of each candidate region is determined based on the second information entropy and the gray value. The total importance is determined based on the first importance and the second importance, and the candidate area with the lowest total importance is determined as the position of the target digital anchor.

2. The method for generating a digital anchor virtual image based on big data artificial intelligence according to claim 1, characterized in that, Each news category corresponds to a preset word library and a preset feature library. Determining the news category to which the news belongs based on the word combinations and features includes: Determine the first quantity of the plurality of first word combinations and the second word combinations in each preset word library; Determine the second quantity of the plurality of features in each preset feature library; Multiple target datasets were identified, each target dataset consisting of features for each second word combination and the images in the broadcast video at the time points when each second word combination appeared. Determine the news domains that are hit for each target dataset and the third number of times each news domain is hit, and determine the preferred domain for each target dataset based on the news domain with the highest third number. The news field to which the news belongs is determined based on the first quantity, the second quantity, and the preferred field.

3. The method for generating a digital anchor virtual image based on big data artificial intelligence according to claim 2, characterized in that, Determining the news domain to which the news belongs based on the first quantity, the second quantity, and the preferred domain includes: Sort the first quantity to obtain the target preset word library with the largest first quantity; Sort the second quantity to obtain the target preset feature library with the second largest quantity; The most frequently occurring target preference domains are obtained by summarizing the preference domains of all target datasets. If at least two of the news domains in the target preset word library, the news domains in the target preset feature library, and the target preference domain are the same, then the at least two identical domains are determined as the news domain to which the news belongs.

4. The method for generating a digital anchor virtual image based on big data artificial intelligence according to claim 3, characterized in that, The step of determining the first importance of each candidate region based on the first information entropy, difference entropy, and similarity includes: Calculate the first average of the information entropy of each region's video and the average of the difference entropy of each absolute difference map; Calculate the second average similarity and similarity variance of adjacent frames in each region of the video; The first importance is determined based on the first average value, the average difference entropy value, the second average value, the similarity variance, and their respective coefficients.

5. The method for generating a digital anchor virtual image based on big data artificial intelligence according to claim 1, characterized in that, The determination of the second importance of each candidate region based on the second information entropy and grayscale value includes: Each target area frame in each region video is divided into multiple sub-region frames; Determine the third average value of the grayscale values ​​corresponding to each sub-region on both sides of the outline and determine the difference of the third average value; The summation of the differences between the images of each sub-region is used to obtain the total sum of the differences between the images of each target region. Based on the sum of the images of each sub-region, a fourth average value of each candidate region with respect to the sum is determined. Based on the second information entropy of the images of each sub-region, a fifth average value of each candidate region with respect to the second information entropy is determined. Based on the fourth average value, the fifth average value, and their respective corresponding coefficients, a second importance of each candidate region is determined.

6. The method for generating a digital anchor virtual image based on big data artificial intelligence according to claim 1, characterized in that, Each news category corresponds to multiple preset clothing combinations, and each preset clothing combination corresponds to multiple tags.

7. A digital anchor virtual avatar generation system based on big data and artificial intelligence, characterized in that, include: The acquisition module is used to acquire the text and video of the news broadcast. The first processing module is used to perform word segmentation on the broadcast text to obtain multiple combinations of first words from the broadcast text; The second processing module is used to perform text recognition and feature recognition on the broadcast video to obtain the second word combination and features in the broadcast video; The domain determination module is used to determine the news domain to which the news belongs based on the first word combination, the second word combination, and features; The clothing mapping module is used to determine the clothing of a benchmark digital anchor based on the news field to which the anchor belongs and to map the clothing onto the benchmark digital anchor to obtain the target digital anchor. Determine the fourth number of tags that match the first word combination, the second word combination, and the feature matching for each preset clothing combination; The fourth and most frequent preset clothing combination is determined as the clothing of the benchmark digital anchor; A location mapping module is used to determine the location of the target digital anchor based on the broadcast video, and map the target digital anchor to the location in the broadcast video; Edge detection is performed on the target digital anchor to obtain the outline of the target digital anchor; The contour is mapped to the starting position of the lower left corner of the broadcast video, and the contour is translated laterally according to a preset step size to obtain multiple candidate areas. The candidate areas include the area where the contour is located when it is at the starting position. The broadcast video is segmented according to each candidate region to obtain the region video of each candidate region; Determine each frame of the video for each region, and calculate the first information entropy of each frame and the similarity between adjacent frames; The target area is drawn with the outline as the center and the width is preset. Each frame of the broadcast video is divided according to the target area to obtain the target area image of each candidate area. Determine the pixel difference at the same position between two adjacent frames to obtain an absolute difference map, and calculate the difference entropy of each absolute difference map; The first importance of each candidate region is determined based on the first information entropy, difference entropy, and similarity. Calculate the second information entropy of each target area frame in each region video and determine the grayscale values ​​of the pixels on both sides of the contour; The second importance of each candidate region is determined based on the second information entropy and the gray value. The total importance is determined based on the first importance and the second importance, and the candidate area with the lowest total importance is determined as the position of the target digital anchor.

8. An electronic device, characterized in that, It includes: At least one processor; Memory; At least one application, wherein the at least one application is stored in the memory and configured to be executed by the at least one processor, the at least one application being used to execute the method for generating a digital anchor virtual image based on big data artificial intelligence according to any one of claims 1 to 6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed in the computer, the computer is instructed to perform the method for generating a digital anchor virtual image based on big data artificial intelligence as described in any one of claims 1 to 6.

Images