CN110737802B - Pirated video detection method and device, electronic equipment and storage medium - Google Patents

Pirated video detection method and device, electronic equipment and storage medium Download PDF

Info

Publication number
CN110737802B
CN110737802B CN201910977721.0A CN201910977721A CN110737802B CN 110737802 B CN110737802 B CN 110737802B CN 201910977721 A CN201910977721 A CN 201910977721A CN 110737802 B CN110737802 B CN 110737802B
Authority
CN
China
Prior art keywords
detected
video
tested
determining
characteristic
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN201910977721.0A
Other languages
Chinese (zh)
Other versions
CN110737802A (en
Inventor
王飞
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Zhongke Zhiyun Technology Co ltd
Original Assignee
Zhongke Zhiyun Technology Co ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Zhongke Zhiyun Technology Co ltd filed Critical Zhongke Zhiyun Technology Co ltd
Priority to CN201910977721.0A priority Critical patent/CN110737802B/en
Publication of CN110737802A publication Critical patent/CN110737802A/en
Application granted granted Critical
Publication of CN110737802B publication Critical patent/CN110737802B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/70—Information retrieval; Database structures therefor; File system structures therefor of video data
    • G06F16/78—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
    • G06F16/783—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using metadata automatically derived from the content
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/70—Information retrieval; Database structures therefor; File system structures therefor of video data
    • G06F16/71—Indexing; Data structures therefor; Storage structures
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/22—Matching criteria, e.g. proximity measures
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V20/00—Scenes; Scene-specific elements
    • G06V20/40—Scenes; Scene-specific elements in video content
    • G06V20/41—Higher-level, semantic clustering, classification or understanding of video scenes, e.g. detection, labelling or Markovian modelling of sport events or news items
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V20/00—Scenes; Scene-specific elements
    • G06V20/40—Scenes; Scene-specific elements in video content
    • G06V20/46—Extracting features or characteristics from the video content, e.g. video fingerprints, representative shots or key frames

Abstract

The application provides a pirated video detection method, a pirated video detection device, electronic equipment and a storage medium. The pirated video detection method comprises the steps of obtaining a video to be detected, wherein the video to be detected carries an identification to be detected, the identification to be detected is used for identifying the video to be detected, then determining a first characteristic value set to be detected of the video to be detected, and finally determining the pirated video according to the first characteristic value set to be detected and a characteristic database, wherein the characteristic database is a database formed by characteristic values extracted from the original video according to a preset extraction model. Therefore, automatic real-time detection of the pirated video is realized, and the detection accuracy and the detection efficiency are improved.

Description

Pirated video detection method and device, electronic equipment and storage medium
Technical Field
The present application relates to the field of video processing technologies, and in particular, to a method and an apparatus for detecting pirated video, an electronic device, and a storage medium.
Background
With the rapid development of communication technology, a large amount of video resources are emerging continuously, so that users can watch videos conveniently at any time and any place. Meanwhile, the number of pirated videos is increased, and the appearance of the pirated videos brings immeasurable loss to copyright owners of the original videos, so that the detection of the pirated videos is very important.
In the prior art, whether a monitored video is a pirated video or not is generally determined by manually monitoring to see whether the monitored video has similarity with a legal video or not.
However, manual detection has problems of large workload, low efficiency and difficulty in real-time monitoring, and the manual detection process has different degrees of recognition fatigue, resulting in deviation of accuracy of judgment on pirated videos.
Disclosure of Invention
The application provides a pirated video detection method, a pirated video detection device, electronic equipment and a storage medium, which are used for solving the technical problems that the workload of artificially detecting pirated videos is large, the efficiency is low, real-time monitoring is difficult to realize and the detection accuracy has deviation in the prior art.
In a first aspect, the present application provides a method for detecting a pirated video, including:
acquiring a video to be detected, wherein the video to be detected carries an identifier to be detected, and the identifier to be detected is used for identifying the video to be detected;
determining a first set of to-be-detected feature values of the to-be-detected video;
and determining a pirated video according to the first to-be-detected characteristic value set and a characteristic database, wherein the characteristic database comprises characteristic values extracted from the legal video according to a preset extraction model.
In one possible design, before determining a pirated video according to the first set of to-be-detected feature values and the feature database, the method further includes:
the feature database is determined.
In one possible design, the determining the feature database includes:
determining a video frame data set according to a first preset extraction rule and a plurality of legal videos, wherein the video frame data set comprises a plurality of video frame data, and the video frame data comprises a video frame sequence number and a corresponding image;
according to a preset sequence of video frame numbers, extracting characteristic values from each image in sequence according to the preset extraction model;
and determining the characteristic database according to the identifier, the video frame sequence number and the characteristic value, wherein the identifier is used for identifying each legal version video.
In one possible design, the determining the first set of feature values to be detected of the video to be detected includes:
determining a first to-be-detected video frame data set according to a second preset extraction rule and the to-be-detected video, wherein the first to-be-detected video frame data set comprises a plurality of first to-be-detected video frame data, and the first to-be-detected video frame data comprises a first to-be-detected video frame sequence number and a corresponding first to-be-detected image;
according to the preset sequence, sequentially extracting first to-be-detected characteristic values from each first to-be-detected image according to the preset extraction model;
and determining a first set of characteristic values to be detected according to the identifier to be detected, the first video frame serial number to be detected and the first characteristic value to be detected.
In one possible design, the determining a pirated video according to the first set of to-be-detected feature values and a feature database includes:
sequentially determining a first similarity between each first to-be-detected characteristic value in the first to-be-detected characteristic value set and each characteristic value in the characteristic database;
determining a first subset of feature values to be detected according to the first similarity, wherein the first similarity between each first feature value to be detected in the first subset of feature values and each feature value in the feature database is greater than a first preset similarity;
determining a second characteristic value subset to be detected according to the first characteristic value subset to be detected, wherein the first times of occurrence of each first similarity in the second characteristic value subset to be detected are all larger than a first preset time;
determining a third characteristic value subset to be tested according to the second characteristic value subset to be tested, wherein each identifier to be tested in the third characteristic value subset to be tested is consistent with each corresponding identifier in the characteristic database;
and determining whether the frame numbers of the first videos to be tested in the third subset of characteristic values to be tested are sequentially increased, if so, determining that the videos to be tested corresponding to the third subset of characteristic values to be tested are the pirated videos.
In one possible design, after determining a third subset of feature values to be tested from the second subset of feature values to be tested, the method includes:
determining a second video frame data set to be tested according to a third preset extraction rule and a video to be tested corresponding to the third characteristic value subset to be tested, wherein the second video frame data set to be tested comprises a plurality of second video frame data to be tested, and the second video frame data to be tested comprises a second video frame serial number to be tested and a corresponding second image to be tested; the extraction frame rate of the third preset extraction rule is greater than that of the second preset extraction rule;
according to the preset sequence, sequentially extracting second characteristic values to be detected from each second image to be detected according to the preset extraction model so as to generate a second characteristic value set to be detected;
sequentially determining a second similarity between each second characteristic value to be detected in the second characteristic value set to be detected and each characteristic value in the characteristic database;
determining a fourth subset of characteristic values to be detected according to the second similarity, wherein the second similarity between each second characteristic value to be detected in the fourth subset of characteristic values to be detected and each characteristic value in the characteristic database is greater than a second preset similarity; the second preset similarity is greater than the first preset similarity;
determining a fifth subset of characteristic values to be tested according to the fourth subset of characteristic values to be tested, wherein the second times of occurrence of each second similarity in the fifth subset of characteristic values to be tested are all larger than a second preset time; the second preset times are greater than the first preset times;
determining a sixth characteristic value subset to be tested according to the fifth characteristic value subset to be tested, wherein each identifier to be tested in the sixth characteristic value subset to be tested is consistent with each corresponding identifier in the characteristic database;
and determining whether the frame numbers of the second videos to be detected in the sixth subset of the features to be detected are sequentially increased, if so, determining that the videos to be detected corresponding to the sixth subset of the features to be detected are the pirate videos.
Optionally, after determining a pirated video according to the first set of to-be-detected feature values and the feature database, the method further includes:
and generating an alarm message, wherein the alarm message is used for prompting that the video to be detected is the pirated video.
In a second aspect, the present application provides a pirated video detection apparatus, including:
the acquisition module acquires a video to be detected, wherein the video to be detected carries an identifier to be detected, and the identifier to be detected is used for identifying the video to be detected;
the first processing module is used for determining a first set of characteristic values to be detected of the video to be detected;
and the second processing module is used for determining the pirated video according to the first to-be-detected characteristic value set and a characteristic database, wherein the characteristic database comprises characteristic values extracted from the legal video according to a preset extraction model.
In one possible design, the apparatus further includes:
a third processing module for determining the feature database.
In a possible design, the third processing module is specifically configured to:
determining a video frame data set according to a first preset extraction rule and a plurality of legal videos, wherein the video frame data set comprises a plurality of video frame data, and the video frame data comprises a video frame sequence number and a corresponding image;
according to a preset sequence of video frame serial numbers, extracting characteristic values from each image in sequence according to the preset extraction model;
and determining the characteristic database according to the identifier, the video frame sequence number and the characteristic value, wherein the identifier is used for identifying each legal version video.
In one possible design, the first processing module is specifically configured to:
determining a first to-be-detected video frame data set according to a second preset extraction rule and the to-be-detected video, wherein the first to-be-detected video frame data set comprises a plurality of first to-be-detected video frame data, and the first to-be-detected video frame data comprises a first to-be-detected video frame sequence number and a corresponding first to-be-detected image;
according to the preset sequence, sequentially extracting first characteristic values to be detected from each first image to be detected according to the preset extraction model;
and determining a first set of characteristic values to be detected according to the identifier to be detected, the first video frame serial number to be detected and the first characteristic value to be detected.
In one possible design, the second processing module is specifically configured to:
sequentially determining a first similarity between each first to-be-detected characteristic value in the first to-be-detected characteristic value set and each characteristic value in the characteristic database;
determining a first subset of feature values to be detected according to the first similarity, wherein the first similarity between each first feature value to be detected in the first subset of feature values and each feature value in the feature database is greater than a first preset similarity;
determining a second characteristic value subset to be detected according to the first characteristic value subset to be detected, wherein the first times of occurrence of each first similarity in the second characteristic value subset to be detected are all larger than a first preset time;
determining a third characteristic value subset to be tested according to the second characteristic value subset to be tested, wherein each identifier to be tested in the third characteristic value subset to be tested is consistent with each corresponding identifier in the characteristic database;
and determining whether the frame numbers of the first videos to be tested in the third subset of characteristic values to be tested are sequentially increased, if so, determining that the videos to be tested corresponding to the third subset of characteristic values to be tested are the pirated videos.
In one possible design, the second processing module includes a processing subunit, which is specifically configured to:
determining a second video frame data set to be tested according to a third preset extraction rule and a video to be tested corresponding to the third characteristic value subset to be tested, wherein the second video frame data set to be tested comprises a plurality of second video frame data to be tested, and the second video frame data to be tested comprises a second video frame serial number to be tested and a corresponding second image to be tested; the extraction frame rate of the third preset extraction rule is greater than that of the second preset extraction rule;
according to the preset sequence, sequentially extracting second characteristic values to be detected from each second image to be detected according to the preset extraction model so as to generate a second characteristic value set to be detected;
sequentially determining a second similarity between each second characteristic value to be detected in the second characteristic value set to be detected and each characteristic value in the characteristic database;
determining a fourth subset of characteristic values to be detected according to the second similarity, wherein the second similarity between each second characteristic value to be detected in the fourth subset of characteristic values to be detected and each characteristic value in the characteristic database is greater than a second preset similarity; the second preset similarity is greater than the first preset similarity;
determining a fifth subset of characteristic values to be tested according to the fourth subset of characteristic values to be tested, wherein the second times of occurrence of each second similarity in the fifth subset of characteristic values to be tested are all larger than a second preset time; the second preset times are greater than the first preset times;
determining a sixth characteristic value subset to be tested according to the fifth characteristic value subset to be tested, wherein each identifier to be tested in the sixth characteristic value subset to be tested is consistent with each corresponding identifier in the characteristic database;
and determining whether the frame numbers of the second videos to be detected in the sixth subset of the features to be detected are sequentially increased, if so, determining that the videos to be detected corresponding to the sixth subset of the feature values to be detected are the pirated videos.
Optionally, the apparatus further comprises:
and the generating module is used for generating an alarm message, and the alarm message is used for prompting that the video to be detected is the pirated video.
In a third aspect, the present application provides an electronic device, comprising:
at least one processor; and
a memory communicatively coupled to the at least one processor; wherein the content of the first and second substances,
the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the method of detecting pirated video according to the first aspect and optional aspects.
In a fourth aspect, the present application provides a non-transitory computer-readable storage medium storing computer instructions for causing a computer to perform the method for detecting pirated video according to the first aspect and optional aspects.
The application provides a pirated video detection method, a device, an electronic device and a storage medium, wherein a video to be detected is firstly obtained, the video to be detected carries an identification to be detected, the identification to be detected is used for identifying the video to be detected, then a first set of characteristic values to be detected of the video to be detected is determined, and finally the pirated video is determined according to the first set of characteristic values to be detected and a characteristic database, wherein the characteristic database is a database formed by characteristic values extracted from a legal video according to a preset extraction model. Compared with the prior art, the pirated video detection method can realize automatic real-time detection of the pirated video and improve the detection accuracy and the detection efficiency.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present disclosure and together with the description, serve to explain the principles of the disclosure.
Fig. 1 is a schematic view of an application scenario of a pirate video detection method according to an embodiment of the present application;
fig. 2 is a schematic flowchart of a pirated video detection method according to an embodiment of the present application;
fig. 3 is a schematic flowchart of determining a feature database according to an embodiment of the present application;
fig. 4 is a schematic flowchart of determining a first set of feature values to be detected according to an embodiment of the present application;
fig. 5 is a schematic flowchart of determining pirated video according to an embodiment of the present application;
fig. 6 is a schematic flow chart of another method for determining pirated video according to an embodiment of the present application;
fig. 7 is a schematic structural diagram of a pirate video detection apparatus according to an embodiment of the present application;
fig. 8 is a schematic structural diagram of an electronic device according to an embodiment of the present application.
With the foregoing drawings in mind, certain embodiments of the disclosure have been shown and described in more detail below. These drawings and written description are not intended to limit the scope of the disclosed concepts in any way, but rather to illustrate the concepts of the disclosure to those skilled in the art by reference to specific embodiments.
Detailed Description
Reference will now be made in detail to the exemplary embodiments, examples of which are illustrated in the accompanying drawings. When the following description refers to the accompanying drawings, like numbers in different drawings represent the same or similar elements unless otherwise indicated. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present application. Rather, they are merely examples of methods and apparatus consistent with certain aspects of the present application, as detailed in the appended claims.
The terms "first," "second," "third," "fourth," and the like in the description and in the claims of the present application and in the above-described drawings (if any) are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used is interchangeable under appropriate circumstances such that the embodiments of the application described herein are, for example, capable of operation in sequences other than those illustrated or otherwise described herein. Furthermore, the terms "comprises," "comprising," and "having," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed, but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.
With the rapid development of communication technology, a large amount of video resources are emerging continuously, so that users can watch the video conveniently at any time and any place. Accordingly, more and more pirated videos appear, and bring immeasurable loss to copyright owners of the original videos, so that the detection of the pirated videos is particularly important. In the prior art, whether a monitored video is similar to a legal video or not is generally checked through manual monitoring, and whether the monitored video is a pirated video or not is further determined. However, manual detection has problems of large workload, low efficiency and difficulty in real-time monitoring, and different degrees of recognition fatigue may occur in the manual detection process, resulting in deviation of accuracy of judgment on pirated videos.
In order to solve the above problems in the prior art, the present application provides a pirated video detection method, an apparatus, an electronic device, and a storage medium, where a video to be detected is obtained, where the video to be detected carries an identifier to be detected, and the identifier to be detected is used to identify the video to be detected, then a first set of characteristic values to be detected of the video to be detected is determined, and finally the pirated video is determined according to the first set of characteristic values to be detected and a characteristic database, where the characteristic database is a database formed by characteristic values extracted from a legal video according to a preset extraction model. Compared with the prior art, the pirated video detection method provided by the application can realize automatic real-time detection of the pirated video and improve the detection accuracy and the detection efficiency.
The technical solution of the present application will be described in detail below with specific examples. These several specific embodiments may be combined with each other below, and details of the same or similar concepts or processes may not be repeated in some embodiments.
Fig. 1 is a schematic view of an application scenario of a pirate video detection method according to an embodiment of the present application. As shown in fig. 1, the pirate video detection method provided by the present application is executed by an electronic device, where the electronic device may be a mobile phone, a computer, a tablet computer, and the like, and fig. 1 illustrates a computer 1 as an example. By the pirated video detection method, automatic real-time detection on whether the video to be detected is the pirated video can be realized.
Firstly, video stream resources needing to be detected are obtained and used as videos to be detected, videos which are played in a notebook computer 2 shown in the figure 1 are the videos to be detected, the videos to be detected carry identifiers to be detected, the identifiers to be detected are used for identifying the videos to be detected, then first characteristic values to be detected of the videos to be detected are extracted and combined, and finally pirated videos are determined according to the first characteristic value sets to be detected and a characteristic database, wherein the characteristic database is a database which is formed by multiple characteristic values extracted from the original videos according to a preset extraction model, so that automatic real-time monitoring of the pirated videos is achieved, and detection accuracy and detection efficiency are improved.
Fig. 2 is a schematic flowchart of a method for detecting pirated videos according to an embodiment of the present disclosure. As shown in fig. 2, the pirated video detection method provided by this embodiment is executed by an electronic device, and the method includes:
s21: and acquiring a video to be detected.
The video to be detected carries an identifier to be detected, and the identifier to be detected is used for identifying the video to be detected.
And acquiring the video stream resource to be detected, and taking the acquired video stream resource as the video to be detected. The method comprises the steps that when video stream resources are obtained, the video stream resources can carry identifiers to be tested, and the identifiers to be tested are used for identifying videos to be tested. It can be understood that the identifier to be detected is an Identity Identifier (ID) of the video to be detected, and each video to be detected has one or only one identifier to be detected according to the playing media in which the video to be detected is located. Therefore, the video to be detected is obtained, the video to be detected carries the identification to be detected, the video to be detected and the identification to be detected have a one-to-one correspondence relationship, and the identification to be detected is used for identifying the video to be detected.
S22: and determining a first set of characteristic values to be detected of the video to be detected.
After the video to be detected is obtained, a first set of characteristic values to be detected of the video to be detected is determined. The first to-be-detected characteristic value is a characteristic value extracted through a preset extraction model, and the extracted characteristic value is a minimum unit for distinguishing each frame of image. And extracting the characteristic value of each frame of image through a preset extraction model, and taking the characteristic value as a first characteristic value to be detected. It can be understood that the video to be detected is composed of each frame of image, each frame of image corresponds to one video frame number and one first characteristic value to be detected corresponding to the video frame number, multiple frames of images can be extracted from one video to be detected, each frame of image has one video frame number, the first characteristic value to be detected is extracted from each frame of image, and a data set composed of the extracted first characteristic value to be detected, the corresponding video frame number and the identification to be detected of the video to be detected is determined as the first characteristic value set to be detected of the video to be detected.
S23: and determining a pirated video according to the first set of feature values to be detected and the feature database.
The feature database comprises feature values extracted from the original video according to a preset extraction model.
After the first set of feature values to be detected is extracted, the feature database comprises feature values extracted from the original video according to a preset extraction model, so that the first set of feature values to be detected in the first set of feature values is compared with the feature database according to a certain preset rule, and whether the video to be detected is a pirated video or not can be determined relative to the original video. It will be appreciated that the feature database includes a plurality of feature values for each of the original videos.
The pirated video detection method provided by the embodiment of the application comprises the steps of firstly obtaining a video to be detected, wherein the video to be detected carries an identification to be detected, the identification to be detected is used for identifying the video to be detected, then determining a first characteristic value set to be detected of the video to be detected, and finally determining the pirated video according to the first characteristic value set to be detected and a characteristic database, wherein the characteristic database is a database formed by characteristic values extracted from the original video according to a preset extraction model. Therefore, automatic real-time detection of pirated videos is realized, and the problems that manual detection in the prior art is large in workload, low in detection efficiency and difficult to realize real-time monitoring are solved. Furthermore, the pirated video detection method provided by the application does not need manual intervention, so that compared with the prior art, identification fatigue does not occur in the detection process, and the detection accuracy is further improved.
In one possible design, before determining the pirated video according to the first set of to-be-detected feature values and the feature database, the method further includes:
a feature database is determined.
The pirated video detection method provided by the embodiment of the application determines the pirated video based on the first to-be-detected feature value set and the feature database, and can determine the feature database before detecting the to-be-detected video to determine whether the to-be-detected video is the pirated video.
The feature database is relative to the original video, and the determination of the feature database can be to collect feature values uniquely characterizing the original video so as to form the feature database. Or extracting characteristic values from the original video according to a certain preset extraction model, and forming a characteristic database by the characteristic values and data representing the original video corresponding to the characteristic values on the basis of obtaining the characteristic values.
Optionally, fig. 3 is a schematic flowchart of a process for determining a feature database according to an embodiment of the present application, and as shown in fig. 3, the method for determining a feature database according to the embodiment includes:
s31: and determining a video frame data set according to a first preset extraction rule and a plurality of legal videos.
The video frame data set comprises a plurality of video frame data, and the video frame data comprise video frame serial numbers and corresponding images.
The plurality of original videos mean that the same original video can be signed with different media to play, and the videos from different media with the signed play all belong to the original videos.
It is understood that, in the art, for continuous image conversion, when the number of frames per second exceeds 24 frames, the continuous image looks smooth and continuous due to the persistence of vision, and thus the continuous image is called video. The number of still pictures played in the video format per second is referred to as frame rate (fps).
Based on the above, the method for determining the feature database provided by this embodiment determines a video frame data set according to a first preset extraction rule and a plurality of original videos, where the video frame data set includes a plurality of video frame data, and each video frame data includes a video frame number and an image corresponding to the video frame number. In other words, the video frame data of the original video is determined from the plurality of original videos according to the first predetermined extraction rule, the amount of the obtained video frame data is multiple, and thus the determined result is a set of video frame data, and for each video frame data, it includes a video frame number and a corresponding image.
The first preset extraction rule is to extract video frame data in the original video according to a high frame rate. The specific value of the high frame rate may be determined according to various data such as the image quality of the original video, and the specific value is not limited in the embodiment of the present application, but the high frame rate in the embodiment is to extract corresponding video frame data from a plurality of original videos according to an extraction rule of the video to be detected, for example, at 1fps, and the extracted data includes a video frame number and a corresponding image.
S32: and according to the preset sequence of the video frame serial numbers, sequentially extracting characteristic values from each image according to a preset extraction model.
In the video frame data set determined according to the first preset extraction rule, each video frame data comprises a video frame sequence number and a corresponding image, and characteristic values are sequentially extracted from the images by adopting a preset extraction model according to a preset sequence of the video frame sequence numbers. It can be understood that the video frame number is the determined number of each video frame, and the feature value is extracted from each image according to the preset sequence, that is, the feature value corresponding to the video frame number is extracted from the image corresponding to each video frame number according to the preset sequence. It should be understood that the preset sequence is a sequence of video frame numbers, and the number may be from small to large, or from large to small, which is not limited in this embodiment of the present application.
It should be noted that the preset extraction model is a model capable of extracting a feature value in the art, where the feature value extracted by the extraction model is a minimum unit for distinguishing each image, and for example, the preset extraction model may be a depth feature extraction model. For the training of the preset extraction model, so that the preset extraction model can extract the feature values of the difference images, the training process of the preset extraction model is not described herein again.
S33: and determining a characteristic database according to the identification, the video frame sequence number and the characteristic value.
Wherein the identification is used to identify each of the genuine videos.
After the characteristic values are sequentially extracted from the images corresponding to the video frame serial numbers by adopting a preset extraction model according to the preset sequence of the video frame serial numbers, the identifiers, the video frame serial numbers and the characteristic values corresponding to the video frame serial numbers are stored in a database, and the database is determined as a characteristic database.
As described above, a plurality of original videos refer to that the same original video can be associated with different media to make a play agreement, and for each playing media, an identifier belonging to the playing media is added to the played original video, so that for the playing media, the identifier can identify each original video.
The method for determining the feature database includes the steps of determining a video frame data set according to a first preset extraction rule and a plurality of legal videos, wherein the determined video frame data set includes a plurality of video frame data, each video frame data includes a video frame number and an image corresponding to the video frame number, then sequentially extracting feature values from each image by using a preset extraction model according to a preset sequence of the video frame numbers, and finally determining the feature database according to an identifier, the video frame number and the feature values, wherein the identifier can identify each legal video. Thus, a feature database is determined. The feature database determined in this embodiment includes an identifier for identifying each original video, a video frame number, and a feature value of a minimum unit capable of distinguishing images corresponding to each video frame number.
Based on the foregoing embodiment, a possible implementation manner of determining the first set of feature values to be detected of the video to be detected, which is described in step S22, is shown in fig. 4, where fig. 4 is a schematic flowchart of a process for determining the first set of feature values to be detected provided by the embodiment of the present application, and the method includes:
s221: and determining a first video frame data set to be tested according to the second preset extraction rule and the video to be tested.
The first set of video frame data to be tested comprises a plurality of first video frame data to be tested, and the first video frame data to be tested comprises a first video frame number to be tested and a corresponding first image to be tested.
The implementation manner and principle of step S221 are similar to those of step S31, except that step S221 employs a second preset extraction rule, and the extracted object is a video to be tested. Specifically, the second preset extraction rule is to extract video frame data in the video to be detected according to the low frame rate, that is, the first video frame data to be detected, where the extracted multiple first video frame data to be detected form a first video frame data set to be detected, and each first video frame data to be detected includes a first video frame number to be detected and a corresponding first image to be detected.
It should be noted that the second preset extraction rule is extraction according to a low frame rate, and the specific value of the low frame rate is not limited in the embodiment of the present application as long as the value is smaller than the high frame rate adopted by the first preset extraction rule. For example, the second preset extraction rule may be to extract the first video frame data to be detected from the video to be detected at 0.1 fps.
S222: and sequentially extracting the first to-be-detected characteristic values from each first to-be-detected image according to a preset sequence and a preset extraction model.
Step S222 is implemented in a manner similar to step S32, and when the first to-be-detected feature values are extracted from each first to-be-detected image, the sequence is consistent with the preset sequence in step S32, and the adopted preset extraction model is also the preset extraction model in step S32.
S223: and determining a first set of characteristic values to be detected according to the identifier to be detected, the first video frame number to be detected and the first characteristic value to be detected.
The data set composed of the identifier to be detected, the first video frame number to be detected and the first characteristic value to be detected is determined as the first characteristic value set to be detected, wherein the identifier to be detected is carried by the first video to be detected, the identifier to be detected can identify the video to be detected, and the first video frame number to be detected corresponds to the first characteristic value to be detected one by one.
The method for extracting the first set of characteristics values of the video to be detected according to the embodiment of the application includes determining a first set of video frame data according to a second preset extraction rule and the video to be detected, where the determined first set of video frame data includes a plurality of first video frame data, each first video frame data includes a first video frame number and a corresponding first image to be detected, then extracting corresponding first characteristic values from each first image to be detected of the video to be detected according to the preset sequence and the preset extraction model in step S32, and finally determining a data set composed of the identifier to be detected, the first video frame number and the first characteristic values as the first set of characteristics values to be detected. Thus, a first set of to-be-detected feature values is determined.
On the basis of the foregoing embodiment, a possible implementation manner of determining a pirated video according to the first set of to-be-detected feature values and the feature database in step S23 is shown in fig. 5, where fig. 5 is a schematic flow diagram of determining a pirated video provided by an embodiment of the present application, and the implementation manner includes:
s231: and sequentially determining a first similarity between each first to-be-detected characteristic value in the first to-be-detected characteristic value set and each characteristic value in the characteristic database.
For the video to be detected, a corresponding first characteristic value to be detected exists for each first video frame number in the determined first characteristic value set to be detected, and similarly, for the legal video, a characteristic value corresponding to each video frame number exists in the characteristic database. Therefore, the whole feature database is traversed, the similarity between each first feature value to be detected and each feature value in the feature database is determined in sequence, and the determined similarity is defined as the first similarity.
Alternatively, when determining the first similarity, the cosine similarity may be used to determine the similarity between each first feature value to be measured and each feature value in the feature database. It can be understood that each video to be detected is composed of a plurality of first characteristic values to be detected, and similarly, each original video in the characteristic database is also composed of a plurality of characteristic values, so that the cosine similarity can be adopted to calculate the similarity between the first characteristic values to be detected and the characteristic values in the characteristic database, so as to determine whether the video to be detected is a pirated video more accurately.
S232: and determining a first subset of the characteristic values to be detected according to the first similarity.
And the first similarity between each first to-be-detected characteristic value in the first to-be-detected characteristic value subset and each characteristic value in the characteristic database is greater than a first preset similarity.
After the similarity between the first characteristic value to be detected and each characteristic value in the characteristic database is determined, namely after the first similarity is determined, screening out the first characteristic value to be detected, which corresponds to the first similarity, and forming a first characteristic value subset to be detected by using a data set consisting of the screened first characteristic value to be detected, the first video frame serial number to be detected, and the corresponding identification to be detected.
It should be understood that the first preset similarity may be set by a person skilled in the art according to the specific related original video and/or the video to be detected, and the embodiment of the present application is not limited thereto. For example, the first preset similarity may be set to 80%, and the above description may be understood as that the similarity between each of the first subset of the determined first measured characteristic values and each of the characteristic values in the characteristic database is greater than 80%.
S233: and determining a second subset of the characteristic values to be detected according to the first subset of the characteristic values to be detected.
And the first times of occurrence of each first similarity in the second characteristic value subset to be detected are all larger than the first preset times.
After the first subset of the feature values to be measured is determined, the first feature values to be measured which appear for a plurality of times in the first subset of the feature values to be measured are counted, the number of times of appearance is calculated, and the number of times is defined as the first number of times. And when the first time is greater than a first preset time, screening out a corresponding first characteristic value to be detected, and determining a data set consisting of the screened first characteristic value to be detected, a first video frame number to be detected corresponding to the screened first characteristic value to be detected and a corresponding first identification to be detected as a second characteristic value subset to be detected.
It should be understood that the first preset number may be set by a person skilled in the art according to specific situations, and the embodiments of the present application are not limited thereto. For example, the first preset number of times may be set to 5 times, and the above description may be understood that the number of times of occurrence of the first similarity of each first feature value in the determined second subset of feature values to be detected is greater than 5 times.
S234: and determining a third subset of characteristic values to be tested according to the second subset of characteristic values to be tested.
And each identifier to be detected in the third subset of characteristic values to be detected is consistent with each corresponding identifier in the characteristic database.
After the second characteristic value subset to be detected is determined, whether each identification to be detected in the second characteristic value subset to be detected is consistent with each corresponding identification in the characteristic database or not is sequentially determined, the identification to be detected corresponding to the consistency is screened out, and a data set formed by the screened identification to be detected, the first characteristic value to be detected corresponding to the screened identification to be detected and the corresponding first video frame serial number to be detected is determined as a third characteristic value subset to be detected.
It should be understood that, when the first subset of feature values to be measured is determined in step S232, the feature values in the feature database corresponding to the feature values with the first similarity greater than the first preset similarity, and the video frame numbers and the identifiers corresponding to the feature values need to be stored. Similarly, when the second subset of feature values to be measured is determined in step S233, the feature values in the feature database corresponding to the first similarity, the video frame numbers and the identifiers corresponding to the feature values, which are greater than the number of times of occurrence of the first similarity, also need to be stored. Therefore, when determining the third feature value subset to be measured in step S234, it is necessary to compare each first identifier to be measured in the third feature value subset to each identifier in the feature database stored in step S233.
S235: and determining whether the frame numbers of the first videos to be tested in the third subset of characteristic values to be tested are sequentially increased, and if so, determining that the video to be tested corresponding to the third subset of characteristic values to be tested is a pirated video.
After the third characteristic value subset to be detected is determined, whether the frame numbers of the first video to be detected in the third characteristic value subset to be detected are sequentially increased, if so, determining that the video to be detected corresponding to the first characteristic value to be detected in the third characteristic value subset to be detected is a pirated video.
It is understood that, when the third subset of feature values to be tested is determined in step S234, the first feature value to be tested and the first video frame number corresponding to the screened identifier to be tested are stored according to the preset sequence in step S21 and step S22.
The method for determining pirated videos provided in the embodiment of the present application first determines similarity between each first to-be-detected feature value in the first to-be-detected feature value set and each feature value in the feature database in sequence, screens out a first to-be-detected feature value with similarity greater than a first preset similarity, determines a data set composed of the screened first to-be-detected feature value, a corresponding first to-be-detected video frame number, and a first to-be-detected identifier as a first to-be-detected feature value subset, then determining a second characteristic value subset to be tested according to the first characteristic value subset to be tested, determining a third characteristic value subset to be tested according to the second characteristic value subset to be tested, finally determining whether the sequence numbers of the first video frames to be tested in the third characteristic value subset to be tested are sequentially increased, if so, and determining that the video to be tested corresponding to the first video frame number in the third subset of characteristic values to be tested is a pirated video. Therefore, automatic real-time detection of pirated videos is realized, and the problems that manual detection in the prior art is large in workload, low in detection efficiency and difficult to realize real-time monitoring are solved. Furthermore, the pirated video detection method provided by the embodiment does not need manual intervention, and identification fatigue does not occur in the detection process, so that the detection accuracy is improved.
In order to further improve the accuracy of detecting the pirated video, after step S234, the embodiment of the present application provides another implementation manner for determining the pirated video, as shown in fig. 6, fig. 6 is another schematic flow chart for determining the pirated video provided by the embodiment of the present application, and the implementation manner includes:
s61: and determining a second video frame data set to be tested according to a third preset extraction rule and the video to be tested corresponding to the third characteristic value subset to be tested.
The second to-be-detected video frame data set comprises a plurality of second to-be-detected video frame data, and the second to-be-detected video frame data comprise a second to-be-detected video frame sequence number and a corresponding second to-be-detected image; the extraction frame rate of the third preset extraction rule is greater than the extraction frame rate of the second preset extraction rule.
The implementation manner and principle of step S61 are similar to those of step S221, except that step S61 employs a third preset extraction rule, the extracted object is the video to be tested corresponding to the third subset of feature values to be tested, and the result determined by the third preset extraction rule is the second video frame number to be tested and the second image to be tested corresponding to the second video frame number.
The embodiment shown in fig. 6 is developed on the basis of fig. 5 to further improve the accuracy of detection, and therefore, the extraction frame rate of the third preset extraction rule is greater than that of the second preset extraction rule. For example, when the decimation frame rate of the second preset decimation rule is 0.1fps, the decimation frame rate of the third preset decimation rule can be 1 fps. It should be noted that, the specific value of the extraction frame rate of the third preset extraction rule is not limited in the embodiment of the present application, and only the extraction frame rate of the third preset extraction rule is greater than the extraction frame rate of the second preset extraction rule.
S62: and according to a preset sequence, sequentially extracting second characteristic values to be detected from each second image to be detected according to a preset extraction model so as to generate a second characteristic value set to be detected.
Step S62 is the same as step S222 in the preset order and the preset extraction model, and the implementation manner and principle thereof are similar, except that the second to-be-detected feature value set generated in step S62 includes the second to-be-detected feature value, the second to-be-detected video frame number corresponding to the second to-be-detected feature value, and the to-be-detected identifier.
S63: and sequentially determining a second similarity between each second characteristic value to be detected in the second characteristic value set to be detected and each characteristic value in the characteristic database.
S64: and determining a fourth subset of the characteristic values to be detected according to the second similarity.
And the second similarity between each second characteristic value to be detected in the fourth characteristic value subset to be detected and each characteristic value in the characteristic database is greater than the second preset similarity.
S65: and determining a fifth subset of characteristic values to be tested according to the fourth subset of characteristic values to be tested.
And the second times of occurrence of each second similarity in the fifth characteristic value subset to be detected are all larger than the second preset times.
S66: and determining a sixth characteristic value subset to be tested according to the fifth characteristic value subset to be tested.
And each identifier to be tested in the sixth characteristic value subset to be tested is consistent with each corresponding identifier in the characteristic database.
S67: and determining whether the frame numbers of the second videos to be tested in the sixth subset of the features to be tested are sequentially increased, and if so, determining that the videos to be tested corresponding to the sixth subset of the feature values to be tested are pirate videos.
The implementation principle and effect of steps S63 to S67 are similar to those of steps S231 to S235 in the embodiment shown in fig. 5, and the specific process may refer to the embodiment shown in fig. 5, which is not described herein again. It should be noted that the second preset similarity in this embodiment may be the same as or different from the first preset similarity in the embodiment shown in fig. 5. Similarly, the second preset number of times in this embodiment may be the same as or different from the first preset number of times. The present embodiment is not limited to this.
It should be noted that the embodiments of determining a pirated video described in fig. 5 and fig. 6 of the present application only exemplify two rules, and those skilled in the art may set other rules according to specific situations based on the first feature value set of the video to be detected and the feature database to determine a pirated video.
On the basis of the foregoing embodiments, optionally, after determining a pirated video, the method for detecting a pirated video according to the embodiment of the present application further includes:
and generating an alarm message.
The alarm message is used for prompting that the video to be detected is a pirated video.
It can be understood that, after determining that the video to be detected is the pirated video, the pirated video detection method provided by this embodiment can also generate an alarm message, where the generated alarm message can prompt the user that the monitored video to be detected is the pirated video, so that the user can timely learn and make corresponding measures, thereby reducing the loss caused by the pirated video and better maintaining the legitimate rights and interests of the video copyright owner.
Fig. 7 is a schematic structural diagram of a pirate video detection apparatus according to an embodiment of the present application. The pirated video detection apparatus provided in this embodiment may execute the pirated video detection method provided in each of the embodiments. As shown in fig. 7, the pirated video detection apparatus 100 provided by the present embodiment includes:
the acquiring module 101 is configured to acquire a video to be detected, where the video to be detected includes an identifier to be detected, and the identifier to be detected is used to identify the video to be detected.
The first processing module 102 is configured to determine a first set of feature values to be detected of a video to be detected.
The second processing module 103 is configured to determine a pirated video according to the first set of feature values to be detected and a feature database, where the feature database includes feature values extracted from the legal video according to a preset extraction model.
The implementation principle and effect of the pirated video detection apparatus provided in this embodiment are similar to those of the method embodiment shown in fig. 2, and are not described herein again.
In one possible design, the pirated video detection apparatus 100 further comprises:
a third processing module 104 for determining a feature database.
Optionally, the third processing module 104 is specifically configured to:
determining a video frame data set according to a first preset extraction rule and a plurality of legal videos, wherein the video frame data set comprises a plurality of video frame data, and the video frame data comprises a video frame sequence number and a corresponding image;
according to a preset sequence of video frame serial numbers, extracting characteristic values from each image in sequence according to a preset extraction model;
and determining a characteristic database according to the identifier, the video frame sequence number and the characteristic value, wherein the identifier is used for identifying each legal version video.
The implementation principle and the effect of the present embodiment are similar to those of the method embodiment shown in fig. 3, and are not described herein again.
In one possible design, the first processing module 102 is specifically configured to:
determining a first to-be-detected video frame data set according to a second preset extraction rule and a to-be-detected video, wherein the first to-be-detected video frame data set comprises a plurality of first to-be-detected video frame data, and the first to-be-detected video frame data comprise a first to-be-detected video frame sequence number and a corresponding first to-be-detected image;
sequentially extracting first to-be-detected characteristic values from each first to-be-detected image according to a preset sequence and a preset extraction model;
and determining a first set of characteristic values to be detected according to the identifier to be detected, the first video frame number to be detected and the first characteristic value to be detected.
The implementation principle and the effect of this embodiment are similar to those of the method embodiment shown in fig. 4, and are not described herein again.
In one possible design, the second processing module 103 is specifically configured to:
sequentially determining a first similarity between each first to-be-detected characteristic value in the first to-be-detected characteristic value set and each characteristic value in the characteristic database;
determining a first to-be-detected characteristic value subset according to the first similarity, wherein the first similarity between each first to-be-detected characteristic value in the first to-be-detected characteristic value subset and each characteristic value in the characteristic database is greater than a first preset similarity;
determining a second characteristic value subset to be detected according to the first characteristic value subset to be detected, wherein the first times of occurrence of each first similarity in the second characteristic value subset to be detected are all larger than the first preset times;
determining a third characteristic value subset to be tested according to the second characteristic value subset to be tested, wherein each identification to be tested in the third characteristic value subset to be tested is consistent with each corresponding identification in the characteristic database;
and determining whether the frame numbers of the first videos to be tested in the third subset of characteristic values to be tested are sequentially increased, and if so, determining that the videos to be tested corresponding to the third subset of characteristic values to be tested are pirate videos.
The implementation principle and the effect of this embodiment are similar to those of the method embodiment shown in fig. 5, and are not described herein again.
Optionally, the second processing module 103 includes a processing subunit 1030, where the processing subunit 1030 is specifically configured to:
determining a second video frame data set to be detected according to a third preset extraction rule and a video to be detected corresponding to a third characteristic value subset to be detected, wherein the second video frame data set to be detected comprises a plurality of second video frame data to be detected, and the second video frame data to be detected comprises a second video frame serial number to be detected and a corresponding second image to be detected; the extraction frame rate of the third preset extraction rule is greater than that of the second preset extraction rule;
according to a preset sequence, sequentially extracting second characteristic values to be detected from each second image to be detected according to a preset extraction model to generate a second characteristic value set to be detected;
sequentially determining a second similarity between each second characteristic value to be detected in the second characteristic value set to be detected and each characteristic value in the characteristic database;
determining a fourth characteristic value subset to be tested according to the second similarity, wherein the second similarity between each second characteristic value to be tested in the fourth characteristic value subset to be tested and each characteristic value in the characteristic database is greater than a second preset similarity;
determining a fifth subset of characteristic values to be tested according to the fourth subset of characteristic values to be tested, wherein the second times of occurrence of each second similarity in the fifth subset of characteristic values to be tested are all larger than the second preset times;
determining a sixth characteristic value subset to be tested according to the fifth characteristic value subset to be tested, wherein each identifier to be tested in the sixth characteristic value subset to be tested is consistent with each corresponding identifier in the characteristic database;
and determining whether the frame numbers of the second video to be tested in the sixth subset of the feature values to be tested are sequentially increased, if so, determining that the video to be tested corresponding to the sixth subset of the feature values to be tested is a pirated video.
The implementation principle and the effect of the present embodiment are similar to those of the method embodiment shown in fig. 6, and are not described herein again.
Optionally, the pirated video detection apparatus 100 further includes:
and the generating module 105 is configured to generate an alarm message, where the alarm message is used to prompt that the video to be detected is a pirated video.
Fig. 8 is a schematic structural diagram of an electronic device according to an embodiment of the present application. As shown in fig. 8, the electronic device 800 provided in the present embodiment includes:
at least one processor 801; and
a memory 802 communicatively coupled to the at least one processor 801; wherein the content of the first and second substances,
the memory 802 stores instructions executable by the at least one processor 801, and the instructions are executed by the at least one processor 801, so that the at least one processor 801 can execute the steps of the above-mentioned pirated video detection method, which can be referred to in the foregoing method embodiments in detail.
In an exemplary embodiment, the present application provides a non-transitory computer readable storage medium storing computer instructions for causing a computer to perform the steps of the stroke skeleton information extraction method in the above embodiments. For example, the readable storage medium may be a ROM, a Random Access Memory (RAM), a CD-ROM, a magnetic tape, a floppy disk, an optical data storage device, and the like.
Other embodiments of the disclosure will be apparent to those skilled in the art from consideration of the specification and practice of the disclosure disclosed herein. This application is intended to cover any variations, uses, or adaptations of the disclosure following, in general, the principles of the disclosure and including such departures from the present disclosure as come within known or customary practice within the art to which the disclosure pertains. It is intended that the specification and examples be considered as exemplary only, with a true scope and spirit of the disclosure being indicated by the following claims.
It will be understood that the present disclosure is not limited to the precise arrangements described above and shown in the drawings and that various modifications and changes may be made without departing from the scope thereof. The scope of the present disclosure is limited only by the appended claims.

Claims (9)

1. A method for detecting pirated video, comprising:
acquiring a video to be detected, wherein the video to be detected carries an identifier to be detected, and the identifier to be detected is used for identifying the video to be detected;
determining a first set of to-be-detected feature values of the to-be-detected video;
determining a pirated video according to the first to-be-detected feature value set and a feature database, wherein the feature database comprises feature values extracted from the legal video according to a preset extraction model;
determining a pirated video according to the first set of to-be-detected feature values and the feature database, comprising:
sequentially determining a first similarity between each first to-be-detected characteristic value in the first to-be-detected characteristic value set and each characteristic value in the characteristic database;
determining a first subset of feature values to be detected according to the first similarity, wherein the first similarity between each first feature value to be detected in the first subset of feature values and each feature value in the feature database is greater than a first preset similarity;
determining a second characteristic value subset to be detected according to the first characteristic value subset to be detected, wherein the first times of occurrence of each first similarity in the second characteristic value subset to be detected are all larger than a first preset time;
determining a third characteristic value subset to be tested according to the second characteristic value subset to be tested, wherein each identifier to be tested in the third characteristic value subset to be tested is consistent with each corresponding identifier in the characteristic database;
and determining whether the frame numbers of the first videos to be tested in the third subset of characteristic values to be tested are sequentially increased, if so, determining that the videos to be tested corresponding to the third subset of characteristic values to be tested are the pirated videos.
2. The method according to claim 1, wherein before determining the pirated video according to the first set of to-be-detected feature values and the feature database, the method further comprises:
determining the feature database.
3. The pirated video detection method according to claim 2, wherein the determining the feature database comprises:
determining a video frame data set according to a first preset extraction rule and a plurality of legal videos, wherein the video frame data set comprises a plurality of video frame data, and the video frame data comprises video frame serial numbers and corresponding images;
according to a preset sequence of video frame numbers, extracting characteristic values from each image in sequence according to the preset extraction model;
and determining the characteristic database according to the identifier, the video frame sequence number and the characteristic value, wherein the identifier is used for identifying each legal version video.
4. The method according to claim 3, wherein said determining a first set of feature values to be detected of the video to be detected comprises:
determining a first to-be-detected video frame data set according to a second preset extraction rule and the to-be-detected video, wherein the first to-be-detected video frame data set comprises a plurality of first to-be-detected video frame data, and the first to-be-detected video frame data comprises a first to-be-detected video frame sequence number and a corresponding first to-be-detected image;
according to the preset sequence, sequentially extracting first characteristic values to be detected from each first image to be detected according to the preset extraction model;
and determining a first set of characteristic values to be detected according to the identifier to be detected, the first video frame serial number to be detected and the first characteristic value to be detected.
5. The method according to claim 1, wherein determining a third subset of feature values to be tested according to the second subset of feature values to be tested comprises:
determining a second video frame data set to be tested according to a third preset extraction rule and a video to be tested corresponding to the third characteristic value subset to be tested, wherein the second video frame data set to be tested comprises a plurality of second video frame data to be tested, and the second video frame data to be tested comprises a second video frame serial number to be tested and a corresponding second image to be tested; the extraction frame rate of the third preset extraction rule is greater than that of the second preset extraction rule;
according to a preset sequence, sequentially extracting second characteristic values to be detected from each second image to be detected according to the preset extraction model to generate a second characteristic value set to be detected;
sequentially determining a second similarity between each second characteristic value to be detected in the second characteristic value set to be detected and each characteristic value in the characteristic database;
determining a fourth subset of characteristic values to be detected according to the second similarity, wherein the second similarity between each second characteristic value to be detected in the fourth subset of characteristic values to be detected and each characteristic value in the characteristic database is greater than a second preset similarity;
determining a fifth subset of characteristic values to be tested according to the fourth subset of characteristic values to be tested, wherein the second times of occurrence of each second similarity in the fifth subset of characteristic values to be tested are all larger than a second preset time;
determining a sixth characteristic value subset to be tested according to the fifth characteristic value subset to be tested, wherein each identifier to be tested in the sixth characteristic value subset to be tested is consistent with each corresponding identifier in the characteristic database;
and determining whether the frame numbers of the second videos to be detected in the sixth subset of the features to be detected are sequentially increased, if so, determining that the videos to be detected corresponding to the sixth subset of the feature values to be detected are the pirated videos.
6. The pirated video detection method according to any one of claims 1-5, wherein after determining the pirated video according to the first set of to-be-detected feature values and the feature database, the method further comprises:
and generating an alarm message, wherein the alarm message is used for prompting that the video to be detected is the pirated video.
7. A pirated video detection apparatus, comprising:
the device comprises an acquisition module, a detection module and a display module, wherein the acquisition module is used for acquiring a video to be detected, the video to be detected comprises an identifier to be detected, and the identifier to be detected is used for identifying the video to be detected;
the first processing module is used for determining a first set of characteristic values to be detected of the video to be detected;
the second processing module is used for determining a pirated video according to the first to-be-detected characteristic value set and a characteristic database, wherein the characteristic database comprises characteristic values extracted from the legal video according to a preset extraction model;
the second processing module is further configured to sequentially determine a first similarity between each first to-be-detected feature value in the first to-be-detected feature value set and each feature value in the feature database; determining a first subset of feature values to be detected according to the first similarity, wherein the first similarity between each first feature value to be detected in the first subset of feature values and each feature value in the feature database is greater than a first preset similarity; determining a second characteristic value subset to be detected according to the first characteristic value subset to be detected, wherein the first times of occurrence of each first similarity in the second characteristic value subset to be detected are all larger than a first preset time; determining a third characteristic value subset to be tested according to the second characteristic value subset to be tested, wherein each identifier to be tested in the third characteristic value subset to be tested is consistent with each corresponding identifier in the characteristic database; and determining whether the frame numbers of the first videos to be tested in the third subset of characteristic values to be tested are sequentially increased, if so, determining that the videos to be tested corresponding to the third subset of characteristic values to be tested are the pirated videos.
8. An electronic device, comprising:
at least one processor; and
a memory communicatively coupled to the at least one processor; wherein the content of the first and second substances,
the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the pirated video detection method according to any one of claims 1-6.
9. A non-transitory computer readable storage medium storing computer instructions for causing a computer to perform the pirated video detection method according to any one of claims 1 to 6.
CN201910977721.0A 2019-10-15 2019-10-15 Pirated video detection method and device, electronic equipment and storage medium Active CN110737802B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201910977721.0A CN110737802B (en) 2019-10-15 2019-10-15 Pirated video detection method and device, electronic equipment and storage medium

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201910977721.0A CN110737802B (en) 2019-10-15 2019-10-15 Pirated video detection method and device, electronic equipment and storage medium

Publications (2)

Publication Number Publication Date
CN110737802A CN110737802A (en) 2020-01-31
CN110737802B true CN110737802B (en) 2022-06-03

Family

ID=69270025

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201910977721.0A Active CN110737802B (en) 2019-10-15 2019-10-15 Pirated video detection method and device, electronic equipment and storage medium

Country Status (1)

Country Link
CN (1) CN110737802B (en)

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101374234A (en) * 2008-09-25 2009-02-25 清华大学 Method and apparatus for monitoring video copy base on content
CN104143055A (en) * 2014-08-16 2014-11-12 合一网络技术(北京)有限公司 Pirated video monitoring method and system
CN105681899A (en) * 2015-12-31 2016-06-15 北京奇艺世纪科技有限公司 Method and device for detecting similar video and pirated video
CN110222594A (en) * 2019-05-20 2019-09-10 厦门能见易判信息科技有限公司 Pirate video recognition methods and system

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20140211978A1 (en) * 2013-01-30 2014-07-31 Hcl Technologies Limited System and Method to Detect Video Piracy
CN106055574B (en) * 2016-05-19 2019-12-24 微梦创科网络科技（中国）有限公司 Method and device for identifying illegal uniform resource identifier (URL)
CN108447501B (en) * 2018-03-27 2020-08-18 中南大学 Pirated video detection method and system based on audio words in cloud storage environment

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101374234A (en) * 2008-09-25 2009-02-25 清华大学 Method and apparatus for monitoring video copy base on content
CN104143055A (en) * 2014-08-16 2014-11-12 合一网络技术(北京)有限公司 Pirated video monitoring method and system
CN105681899A (en) * 2015-12-31 2016-06-15 北京奇艺世纪科技有限公司 Method and device for detecting similar video and pirated video
CN110222594A (en) * 2019-05-20 2019-09-10 厦门能见易判信息科技有限公司 Pirate video recognition methods and system

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
基于内容的视频拷贝检测的实现;向志敏;《中国优秀博硕士学位论文全文数据库(硕士)信息科技辑》;20100315;第Ⅰ138-706页 *
小波域数字图像水印改进算法及其性能分析;王万良等;《计算机辅助设计与图形学学报》;20040920(第09期);第1235-1239页 *

Also Published As

Publication number Publication date
CN110737802A (en) 2020-01-31

Similar Documents

Publication Publication Date Title
WO2020151489A1 (en) Living body detection method based on facial recognition, and electronic device and storage medium
CN108229321B (en) Face recognition model, and training method, device, apparatus, program, and medium therefor
CN109214337B (en) Crowd counting method, device, equipment and computer readable storage medium
US9224048B2 (en) Scene-based people metering for audience measurement
CN110853033A (en) Video detection method and device based on inter-frame similarity
CN110659604A (en) Video detection method, device, server and storage medium
CN110505498A (en) Processing, playback method, device and the computer-readable medium of video
CN109286848B (en) Terminal video information interaction method and device and storage medium
CN115396705B (en) Screen operation verification method, platform and system
WO2022062968A1 (en) Self-training method, system, apparatus, electronic device, and storage medium
CN106162222B (en) A kind of method and device of video lens cutting
CN111611973B (en) Target user identification method, device and storage medium
CN110737802B (en) Pirated video detection method and device, electronic equipment and storage medium
CN113095178A (en) Bad information detection method, system, electronic device and readable storage medium
Zhang et al. Texture information boosts video quality assessment
Frick et al. Detecting “DeepFakes” in H. 264 Video Data Using Compression Ghost Artifacts
CN114786032B (en) Training video management method and system
CN108229320B (en) Frame selection method and device, electronic device, program and medium
CN106295592A (en) Method and device for identifying subtitles of media file and electronic equipment
CN116109964A (en) Intelligent extraction method and device for video map, storage medium and computer equipment
CN111553408B (en) Automatic test method for video recognition software
CN110163043B (en) Face detection method, device, storage medium and electronic device
CN104185032B (en) A kind of video frequency identifying method and system
CN108133210B (en) Image format identification method and device
EP2136314A1 (en) Method and system for generating multimedia descriptors

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant