Method and system for fast review of publically shown movies

Abstract

This invention relates to audio and video review technology, and discloses a rapid review method and system based on publicly released films. The method includes: forming a publicly released film audio fingerprint sample library; registering the opening audio data of publicly released films to a sound recognition system to form the publicly released film audio fingerprint sample library; the opening audio data of publicly released films includes the opening audio data of the film's logo; marking positional information by marking the positional information of characters in the publicly released film audio fingerprint sample library; and reviewing the fingerprint samples by verifying the presence of fingerprint samples in the publicly released film audio fingerprint sample library through the sound recognition system. If the sound recognition system detects a publicly released film audio fingerprint sample in the library, it is considered a fingerprint sample, and the video is a publicly released film video that has passed review. This invention eliminates the need to analyze the entire video, enabling rapid review of publicly released films; simultaneously, this invention provides accurate and efficient detection of publicly released film videos, effectively saving review personnel's time.

Description

Technical Field

[0001] This invention relates to audio and video review technology, and more particularly to a rapid review method and system based on publicly released films. Background Technology

[0002] In addition to manual review, companies now also use intelligent review systems for film censorship, which use AI technology to inspect video content (images, audio, and text). However, due to the limitations of AI technology, many reports are still generated for films that have already obtained public screening licenses, requiring significant manual intervention after intelligent review, thus failing to effectively reduce the workload.

[0003] For example, in the existing technology, the research and implementation of harmful audio and video consistency detection methods, No. 3, 2016; the existing technology mainly detects audio and video based on sound; it cannot be used for video detection in publicly screened movies. Summary of the Invention

[0004] This invention addresses the limitation of existing audio and video detection technologies, which primarily rely on sound detection and are not well-suited for detecting publicly released film videos. It provides a rapid review method and system for publicly released films.

[0005] To solve the above-mentioned technical problems, the present invention provides the following technical solution:

[0006] A rapid review method based on publicly released films, including a voice recognition system; the method includes:

[0007] The formation of the audio fingerprint sample database for publicly released films involves registering the opening audio data of publicly released films to the sound recognition system, thereby forming the audio fingerprint sample database for publicly released films; the opening audio data of publicly released films includes the opening audio data of the Dragon Mark logo;

[0008] Location information annotation: The character location information in the audio fingerprint sample database of publicly released movies is annotated; fingerprint sample verification: The sound recognition system verifies whether the fingerprint sample exists in the audio fingerprint sample database of publicly released movies. When the sound recognition system finds the audio fingerprint sample database of publicly released movies, it is a fingerprint sample, and the video is a publicly released movie video that has passed the verification.

[0009] As a preferred option, the characters in the audio fingerprint sample database of publicly screened films include "public screening license" and "film review number (xxxx) yyy".

[0010] As a preferred option, it also includes the acquisition of time information, which is obtained through a sound recognition system. The time information includes the start time and end time of the dragon logo intro.

[0011] S1. Determining the sample duration: The sample duration is determined by the start and end times of the Dragon Mark intro. If the sample duration matches the duration of the audio fingerprint sample library of publicly released movies more than the set duration threshold, then the sample is a Dragon Mark movie.

[0012] S2. Extraction of sequence images: Extract sequence images of the start and end times of the dragon logo intro using the video decoder at intervals T.

[0013] S3. Recognition of text and location information: For the acquired sequence images, OCR intelligent recognition technology is used to identify the text and location information in the sequence images;

[0014] S4. Text comparison: The identified text is compared with other strings to check if the characters in the sequence images are from the public-screened movie audio fingerprint sample database. If the sequence images contain character information from the public-screened movie audio fingerprint sample database, the center position of the character in the public-screened movie audio fingerprint sample database is calculated using the position information.

[0015] S5. Comparison of center position: The center position in step S4 is compared with the center position in the public screening movie audio fingerprint sample library as a percentage. If the percentage is less than the set percentage threshold, the year and public screening certificate information in the characters in the public screening movie audio fingerprint sample library are extracted.

[0016] Preferably, the interval T is 1-3 seconds.

[0017] Preferably, the duration threshold is greater than 90%.

[0018] To address the aforementioned technical issues, this application also provides a rapid review system based on publicly screened films, including a voice recognition system; it includes:

[0019] The module for forming the audio fingerprint sample database of publicly released films registers the opening audio data of publicly released films to the sound recognition system, thereby forming the audio fingerprint sample database of publicly released films; the opening audio data of publicly released films includes the opening audio data of the Dragon Mark logo;

[0020] The location information annotation module is used to annotate the character location information in the audio fingerprint sample database of publicly screened movies;

[0021] The fingerprint sample review module uses a voice recognition system to check whether a fingerprint sample exists in the audio fingerprint sample database of publicly released movies. If the voice recognition system finds a fingerprint sample in the audio fingerprint sample database of publicly released movies, then the video is a publicly released movie video that has passed the review.

[0022] To address the aforementioned technical problems, this application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements steps such as a rapid review method based on publicly screened films.

[0023] To address the aforementioned technical problems, this application also provides an electronic device comprising: a memory for storing a computer program; and a processor for executing the computer program to implement steps such as a rapid review method based on publicly screened films.

[0024] This invention, by adopting the above technical solutions, has significant technical effects:

[0025] This invention eliminates the need for analyzing the entire video, enabling rapid review of publicly screened film videos. Furthermore, this invention provides accurate and efficient detection of publicly screened film videos, effectively saving reviewers' time. Attached Figure Description

[0026] Figure 1 This is a flowchart of the present invention.

[0027] Figure 2 This is a flowchart of Embodiment 2 of the present invention. Detailed Implementation

[0028] The present invention will now be described in further detail with reference to the accompanying drawings and embodiments.

[0029] Example 1

[0030] Since all films released in theaters need to obtain a film screening license, commonly known as the "Dragon Mark," the Dragon Mark will appear at the beginning of the film. At the same time, the Dragon Mark video will also display the text information "Screening License" and "Film Approval Number."

[0031] A rapid review method based on publicly released films, including a voice recognition system; the method includes:

[0032] The formation of the audio fingerprint sample database for publicly released films involves registering the opening audio data of publicly released films to the sound recognition system, thereby forming the audio fingerprint sample database for publicly released films; the opening audio data of publicly released films includes the opening audio data of the Dragon Mark logo;

[0033] Location information annotation: Annotate the character location information in the audio fingerprint sample database of publicly released movies;

[0034] The fingerprint sample verification process involves using a voice recognition system to check whether the fingerprint sample exists in the audio fingerprint sample database of publicly released films. If the voice recognition system finds the fingerprint sample in the audio fingerprint sample database of publicly released films, then the video is considered a verified publicly released film video.

[0035] The characters in the audio fingerprint sample database of publicly screened movies include "public screening license" and "film review number (xxxx) yyy".

[0036] Example 2

[0037] Based on Example 1, this example also includes the acquisition of time information, which is obtained through a sound recognition system. The time information includes the start time and end time of the dragon logo intro.

[0038] S1. Determining the sample duration: The sample duration is determined by the start and end times of the Dragon Mark intro. If the sample duration matches the duration of the audio fingerprint sample library of publicly released movies more than the set duration threshold (here, the duration threshold is 90%), then the sample is a Dragon Mark movie.

[0039] S2. Extracting the sequence of images: using the video decoder at intervals T (here, the interval is set to 3 seconds), extracting the sequence of images at the start and end times of the dragon logo intro.

[0040] S3. Recognition of text and location information: For the acquired sequence images, OCR intelligent recognition technology is used to identify the text and location information in the sequence images;

[0041] S4. Text comparison: The identified text is compared with other strings to check if the characters in the sequence images are from the public-screened movie audio fingerprint sample database. If the sequence images contain character information from the public-screened movie audio fingerprint sample database, the center position of the character in the public-screened movie audio fingerprint sample database is calculated using the position information.

[0042] S5. Comparison of center position: The center position in step S4 is compared with the center position of the publicly screened movie audio fingerprint sample library as a percentage. When the compared percentage is less than the set percentage threshold (here, the percentage threshold is 3%), the year and public screening certificate information in the characters of the publicly screened movie audio fingerprint sample library are extracted.

[0043] The obtained year and public screening certificate number are verified through the National Film Administration platform. If the information is normal, it is a publicly screened film; publicly screened films are considered exempt from review.

[0044] Example 3

[0045] This embodiment is a rapid review system based on publicly screened films, including a voice recognition system; it includes:

[0046] The module for forming the audio fingerprint sample database of publicly released films registers the opening audio data of publicly released films to the sound recognition system, thereby forming the audio fingerprint sample database of publicly released films; the opening audio data of publicly released films includes the opening audio data of the Dragon Mark logo;

[0047] The location information annotation module is used to annotate the character location information in the audio fingerprint sample database of publicly screened movies;

[0048] The fingerprint sample review module uses a voice recognition system to check whether a fingerprint sample exists in the audio fingerprint sample database of publicly released movies. If the voice recognition system finds a fingerprint sample in the audio fingerprint sample database of publicly released movies, then the video is a publicly released movie video that has passed the review.

[0049] Example 4

[0050] This embodiment is a computer-readable storage medium storing a computer program that, when executed by a processor, implements steps such as a rapid review method based on publicly screened films.

[0051] Example 5

[0052] This embodiment is an electronic device, which includes: a memory for storing a computer program; and a processor for executing the computer program to implement steps such as a rapid review method based on publicly screened films.

Claims

1. A rapid review method for theatrical release films, including a voice recognition system; The methods include: The formation of the audio fingerprint sample database for publicly released films involves registering the opening audio data of publicly released films to the sound recognition system, thereby forming the audio fingerprint sample database for publicly released films; the opening audio data of publicly released films includes the opening audio data of the Dragon Mark logo; Location information annotation: The character location information in the audio fingerprint sample database of publicly screened movies is annotated; fingerprint sample verification: The sound recognition system verifies whether the fingerprint sample exists in the audio fingerprint sample database of publicly screened movies. When the sound recognition system finds the audio fingerprint sample database of publicly screened movies, it is a fingerprint sample, and the video is a publicly screened movie video that has passed the verification. Time information is obtained through a sound recognition system. The time information includes the start and end times of the dragon logo intro. S1. Determining the sample duration: The sample duration is determined by the start and end times of the Dragon Mark intro. If the sample duration matches the duration of the audio fingerprint sample library of publicly released movies more than the set duration threshold, then the sample is a Dragon Mark movie. S2. Extraction of sequence images: Extract sequence images of the start and end times of the dragon logo intro using the video decoder at intervals T. S3. Recognition of text and location information: For the acquired sequence images, OCR intelligent recognition technology is used to identify the text and location information in the sequence images; S4. Text comparison: The identified text is compared with other strings to check if the characters in the sequence images are from the public-screened movie audio fingerprint sample database. If the sequence images contain character information from the public-screened movie audio fingerprint sample database, the center position of the character in the public-screened movie audio fingerprint sample database is calculated using the position information. S5. Comparison of center position: The center position in step S4 is compared with the center position in the public screening movie audio fingerprint sample library as a percentage. If the percentage is less than the set percentage threshold, the year and public screening certificate information in the characters in the public screening movie audio fingerprint sample library are extracted.

2. The rapid review method for publicly screened films according to claim 1, characterized in that, The characters in the audio fingerprint sample database of publicly screened movies include "public screening license" and "film review number (xxxx) yyy".

3. The rapid review method for publicly screened films according to claim 1, characterized in that, The interval T is 1-3 seconds.

4. The rapid review method for publicly screened films according to claim 1, characterized in that, The duration threshold is greater than 90%.

5. A rapid review system for publicly screened films, including a voice recognition system; characterized in that: The module for forming the audio fingerprint sample database of publicly released films registers the opening audio data of publicly released films to the sound recognition system, thereby forming the audio fingerprint sample database of publicly released films; the opening audio data of publicly released films includes the opening audio data of the Dragon Mark logo; The location information annotation module is used to annotate the character location information in the audio fingerprint sample database of publicly screened movies; The fingerprint sample review module uses a voice recognition system to check whether the fingerprint sample exists in the audio fingerprint sample database of publicly released movies. If the voice recognition system finds the audio fingerprint sample database of publicly released movies, it is a fingerprint sample, and the video is a publicly released movie video that has passed the review. Time information is obtained through a sound recognition system. The time information includes the start and end times of the dragon logo intro. S1. Determining the sample duration: The sample duration is determined by the start and end times of the Dragon Mark intro. If the sample duration matches the duration of the audio fingerprint sample library of publicly released movies more than the set duration threshold, then the sample is a Dragon Mark movie. S2. Extraction of sequence images: Extract sequence images of the start and end times of the dragon logo intro using the video decoder at intervals T. S3. Recognition of text and location information: For the acquired sequence images, OCR intelligent recognition technology is used to identify the text and location information in the sequence images; S4. Text comparison: The identified text is compared with other strings to check if the characters in the sequence images are from the public-screened movie audio fingerprint sample database. If the sequence images contain character information from the public-screened movie audio fingerprint sample database, the center position of the character in the public-screened movie audio fingerprint sample database is calculated using the position information. S5. Comparison of center position: The center position in step S4 is compared with the center position in the public screening movie audio fingerprint sample library as a percentage. If the percentage is less than the set percentage threshold, the year and public screening certificate information in the characters in the public screening movie audio fingerprint sample library are extracted.

6. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the steps of the rapid review method for publicly screened films as described in any one of claims 1 to 4.

7. An electronic device, characterized in that, include: Memory, used to store computer programs; A processor, configured to implement the steps of the rapid review method for publicly screened films as described in any one of claims 1 to 4 when executing the computer program.

Images