A cinema piracy monitoring method and apparatus based on detecting invisible light flicker

Abstract

The application discloses a cinema illegal recording behavior monitoring method and device based on detection of invisible light flicker. The specific method comprises the following steps: acquiring audience seat area video stream collected by a cinema monitoring camera in real time, and performing frame-by-frame preprocessing on the input video; screening candidate flash regions from the video frames through regional brightness change feature analysis; continuously tracking the candidate flash regions obtained through the screening, and extracting a brightness time sequence signal; performing frequency domain analysis on the preprocessed signal through fast Fourier transform, comprehensively evaluating the frequency domain features and periodicity, calculating an illegal recording behavior confidence, and triggering an alarm when the confidence exceeds a preset threshold. The application can realize high-precision active monitoring and timely alarm of the illegal recording behavior without affecting the normal audience viewing experience and without modifying the existing projection equipment of the cinema, thereby providing reliable technical support for cinema copyright protection.

Description

Technical Field

[0001] This application relates to the field of cinema piracy technology, and in particular to a method and device for monitoring cinema piracy behavior based on detecting invisible light flickering within the audience area. Background Technology

[0002] In recent years, cases of high-definition film piracy and recording have occurred frequently in my country, with some cases having a serious impact. Despite numerous joint crackdowns by relevant national departments, the problem persists. The reason for this is the rapid development of the shooting capabilities of smartphones and professional action cameras. Suspects can directly record footage with their phones, producing clear images with good image stabilization, easy distortion correction, and great concealment. In regular theaters, these recordings are almost undetectable by staff. In low light, the phone's laser autofocus module emits infrared laser pulses, calculating the distance to objects by measuring reflection time, achieving fast and accurate focusing. While this infrared light is invisible, it can be detected by the flickering of cameras on other devices. Furthermore, during video recording, the camera needs to continuously adjust focus to maintain image clarity, especially in dynamic scenes, where the infrared laser may flicker frequently to track the distance to the subject. Utilizing this characteristic could provide early warning of film piracy. Summary of the Invention

[0003] In order to solve at least one of the above-mentioned technical problems, this application provides a method for monitoring cinema piracy based on detecting invisible light flicker.

[0004] Firstly, the cinema piracy monitoring method based on detecting invisible light flicker provided in this application adopts the following technical solution: A method for monitoring cinema piracy based on detecting invisible light flicker, the method being executed on a server, comprising: S1: Real-time acquisition of video streams from the audience seating area captured by cinema surveillance cameras, frame-by-frame preprocessing of the input video and output of video frames; S2: By analyzing the regional brightness change characteristics, candidate flash areas are initially screened from the video frames; S3: Continuously track the candidate flash regions obtained from the initial screening and extract the brightness time series signal; S4: Perform detrending, bandpass filtering, and windowing on the brightness time series signal to complete signal preprocessing and enhancement; S5: Perform frequency domain analysis on the preprocessed signal using Fast Fourier Transform, detect the main peak signal within the target frequency range, and calculate the signal-to-noise ratio. S6: Based on autocorrelation analysis, perform consistency evaluation on the detected periodic signals to determine whether they conform to the preset flicker coding characteristics; S7: Calculate the confidence level of the recording behavior by combining the signal-to-noise ratio of the main peak signal within the target frequency range and the periodic consistency evaluation results. When the confidence level exceeds the preset threshold, an alarm is triggered.

[0005] Optionally, step S1 specifically includes: S11, convert the input video frame into a grayscale image; S12, the grayscale image is divided into multiple detection regions according to a fixed step size, and the size of each detection region is... Pixel.

[0006] Optionally, step S2 specifically includes: S21, calculate the average brightness of each region using formula (1): (1) in, The average brightness of the area. For the first in the region grayscale value of each pixel; S22, calculate the luminance difference between adjacent frames using formula (2): (2) in, For brightness difference; S23, use formula (3) to count the number of changes in the sign of the brightness difference: (3) in, For indicator functions, The sign change represents the change in the direction of the brightness signal fluctuation; S24, regions that meet the conditions shown in formula (4) are identified as candidate flash regions: (4) in, For historical brightness sets, The threshold for brightness change. This is the lower limit threshold for the number of sign changes.

[0007] Optionally, step S3 specifically includes: S31, Based on the spatial location of the candidate flash region, the Euclidean distance between the region location in the current frame and the existing ROI is calculated using formula (5): (5) in, This represents the position of the candidate region in the current frame. Historical position of existing ROI; S32, when When the current candidate region is matched to the corresponding ROI, Match the preset position threshold; S33, Extract the brightness time series from the successfully matched ROI regions. , For time frame index.

[0008] Optionally, step S4 specifically includes: S41, use formula (6) to perform detrending processing on the brightness time series to eliminate the DC component in the signal: (6) in, The mean of the signal. The signal length; S42, using formula (7) to design the passband of a bandpass filter to cover the preset range of invisible light flicker frequencies: (7) in, and These are the lower and upper limits of the flicker frequency range, respectively. The video sampling frame rate; S43, after the detrended signal is bandpass filtered, it is then windowed using formula (8): (8) Optionally, step S5 specifically includes: S51 uses Fast Fourier Transform to convert the windowed signal to the frequency domain and obtain the spectral energy distribution. ; S52, within the preset target frequency range The highest peak value in the internal search is denoted as the main peak. The corresponding frequency is ; S53, calculate the signal-to-noise ratio using formula (10): (10) in, This represents the average background noise energy outside the target frequency range.

[0009] Optionally, step S6 specifically includes: S61, perform autocorrelation analysis on the signal using formula (11): (11) in, For time delay; S62, Extracting peak interval sequences from autocorrelation functions Calculate the standard deviation of the peak interval ; S63, calculate the periodic consistency score using formula (12): (12) Optionally, step S7 specifically includes: S71, fused frequency domain peak signal-to-noise ratio and periodic consistency score The confidence level of the recording behavior is calculated using formula (13): (13) in, For normalization function, and These are the weighting coefficients. The signal-to-noise ratio threshold; S72, if The system detects unauthorized recording and triggers an alarm. This is a preset alarm threshold.

[0010] Optionally, the invisible light flicker includes spectral components of the infrared and / or ultraviolet bands, and its flicker frequency is preset in the range of 20Hz to 200Hz. The flicker frequency is set according to the characteristics of the invisible light signal actively emitted by the recording device.

[0011] Secondly, this application provides a cinema piracy monitoring device based on detecting invisible light flicker, employing the following technical solution: A monitoring device for cinema piracy based on detecting invisible light flicker, applied to a server, includes: The video capture module is used to acquire real-time video streams of the audience seating area captured by the cinema's surveillance cameras; The region division module is used to divide the input video frame into multiple monitoring regions; The candidate region filtering module is used to calculate the brightness change characteristics of each region and filter candidate flash regions; The tracking module is used to continuously track candidate flash areas and extract brightness time-series signals; The signal processing module is used to perform detrending, bandpass filtering, and windowing processing on the luminance time series signal; The frequency domain analysis module is used to calculate the signal spectrum through fast Fourier transform and to detect the main peak signal and calculate the signal-to-noise ratio within the target frequency range; The periodicity assessment module is used to assess the periodicity consistency of a signal through autocorrelation analysis. The confidence calculation module is used to calculate the confidence level of the recording behavior by combining the signal-to-noise ratio of the main peak signal within the target frequency range and the periodic consistency evaluation results. The alarm module is used to trigger an alarm when the confidence level exceeds a preset threshold.

[0012] The beneficial technical effects of this invention are as follows: This application provides a method for monitoring cinema piracy based on detecting invisible light flicker. By acquiring real-time video streams from cinema surveillance cameras in the audience seating area, and performing techniques such as video region segmentation, signal tracking and processing, and frequency domain analysis, the method can monitor and alert on instances of piracy in the audience seating area. Without affecting the normal viewing experience of audiences or modifying the cinema's existing projection equipment, this method achieves high-precision proactive monitoring and timely alerts for piracy. It effectively solves the problems of lack of proactive monitoring capabilities, high equipment costs, and environmental limitations on recognition accuracy in existing technologies, providing reliable technical support for cinema copyright protection. Attached Figure Description

[0013] Figure 1 This is a flowchart of a cinema piracy monitoring method based on detecting invisible light flickering, according to the present invention. Figure 2 This is a block diagram of the module architecture of a cinema piracy monitoring device based on detecting invisible light flicker, provided in one embodiment of the present invention. Detailed Implementation To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0014] The present invention will be further described in detail below with reference to the accompanying drawings.

[0015] Reference Figure 1 This invention discloses a method for monitoring cinema piracy based on detecting invisible light flicker. The method is executed on a server and includes the following steps: S1: Real-time acquisition of video streams from the audience seating area captured by cinema surveillance cameras, frame-by-frame preprocessing of the input video, and output of video frames.

[0016] Step S1 specifically includes: S11, convert the input video frame into a grayscale image; S12, the grayscale image is divided into multiple detection regions according to a fixed step size, and the size of each detection region is... Pixel.

[0017] S2: By analyzing the regional brightness change characteristics, candidate flash areas are initially screened from video frames.

[0018] Step S2 specifically includes: S21, calculate the average brightness of each region using formula (1): (1) in, The average brightness of the area. For the first in the region grayscale value of each pixel; S22, calculate the luminance difference between adjacent frames using formula (2): (2) in, For brightness difference; S23, use formula (3) to count the number of changes in the sign of the brightness difference: (3) in, For indicator functions, The sign change represents the change in the direction of the brightness signal fluctuation; S24, regions that meet the conditions shown in formula (4) are identified as candidate flash regions: (4) in, For historical brightness sets, The threshold for brightness change. This is the lower limit threshold for the number of sign changes.

[0019] S3: Continuously track the candidate flash regions obtained from the initial screening and extract the brightness time series signal.

[0020] Step S3 specifically includes: S31, Based on the spatial location of the candidate flash region, the Euclidean distance between the region location in the current frame and the existing ROI is calculated using formula (5): (5) in, This represents the position of the candidate region in the current frame. Historical position of existing ROI; S32, when When the current candidate region is matched to the corresponding ROI, Match the preset position threshold; S33, Extract the brightness time series from the successfully matched ROI regions. , For time frame index.

[0021] S4: Perform detrending, bandpass filtering, and windowing on the luminance time series signal to complete signal preprocessing and enhancement.

[0022] Step S4 specifically includes: S41, use formula (6) to perform detrending processing on the brightness time series to eliminate the DC component in the signal: (6) in, The mean of the signal. The signal length; S42, using formula (7) to design the passband of a bandpass filter to cover the preset range of invisible light flicker frequencies: (7) in, and These are the lower and upper limits of the flicker frequency range, respectively. The video sampling frame rate; S43, after the detrended signal is bandpass filtered, it is then windowed using formula (8): (8) S5: Perform frequency domain analysis on the preprocessed signal using Fast Fourier Transform, detect the main peak signal within the target frequency range, and calculate the signal-to-noise ratio.

[0023] Step S5 specifically includes: S51 uses Fast Fourier Transform to convert the windowed signal to the frequency domain and obtain the spectral energy distribution. ; S52, within the preset target frequency range The highest peak value in the internal search is denoted as the main peak. The corresponding frequency is ; S53, calculate the signal-to-noise ratio using formula (10): (10) in, This represents the average background noise energy outside the target frequency range.

[0024] S6: Based on autocorrelation analysis, perform consistency evaluation on the detected periodic signals to determine whether they conform to the preset flicker coding characteristics.

[0025] Step S6 specifically includes: S61, perform autocorrelation analysis on the signal using formula (11): (11) in, For time delay; S62, Extracting peak interval sequences from autocorrelation functions Calculate the standard deviation of the peak interval ; S63, calculate the periodic consistency score using formula (12): (12) S7: Calculate the confidence level of the recording behavior by combining the signal-to-noise ratio of the main peak signal within the target frequency range and the periodic consistency evaluation results. When the confidence level exceeds the preset threshold, an alarm is triggered.

[0026] Step S7 specifically includes: S71, fused frequency domain peak signal-to-noise ratio and periodic consistency score The confidence level of the recording behavior is calculated using formula (13): (13) in, For normalization function, and These are the weighting coefficients. The signal-to-noise ratio threshold; S72, if The system detects unauthorized recording and triggers an alarm. This is a preset alarm threshold.

[0027] In this embodiment, the invisible light flicker includes spectral components of the infrared band and / or ultraviolet band, and its flicker frequency is preset in the range of 20Hz to 200Hz. The flicker frequency is set according to the characteristics of the invisible light signal actively emitted by the recording device.

[0028] In one embodiment, such as Figure 2 As shown, a cinema piracy monitoring device based on detecting invisible light flicker is provided, including the following program modules: Video acquisition module 201 is used to acquire video streams of the audience seating area captured by cinema surveillance cameras in real time; The region division module 202 is used to divide the input video frame into multiple monitoring regions; The candidate region filtering module 203 is used to calculate the brightness change characteristics of each region and filter candidate flash regions. The tracking module 204 is used to continuously track the candidate flash area and extract the brightness time series signal.

[0029] Signal processing module 205 is used to perform detrending, bandpass filtering and windowing processing on the luminance time series signal; The frequency domain analysis module 206 is used to calculate the signal spectrum through fast Fourier transform and detect the main peak signal and calculate the signal-to-noise ratio within the target frequency range; Periodicity assessment module 207 is used to assess the periodicity consistency of a signal through autocorrelation analysis; The confidence calculation module 208 is used to calculate the confidence level of the recording behavior by comprehensively considering the signal-to-noise ratio of the main peak signal within the target frequency range and the periodic consistency evaluation results. Alarm module 209 is used to trigger an alarm when the confidence level exceeds a preset threshold.

[0030] The above are preferred embodiments of this application, and are not intended to limit the scope of protection of this application. Therefore, all equivalent changes made in accordance with the methods and principles of this application should be covered within the scope of protection of this application.

Claims

1. A method for monitoring cinema piracy based on detecting invisible light flicker, characterized in that, The method is executed on a server and includes: S1: Real-time acquisition of video streams from the audience seating area captured by cinema surveillance cameras, frame-by-frame preprocessing of the input video and output of video frames; S2: By analyzing the regional brightness change characteristics, candidate flash areas are initially screened from the video frames; S3: Continuously track the candidate flash regions obtained from the initial screening and extract the brightness time series signal; S4: Perform detrending, bandpass filtering, and windowing on the brightness time series signal to complete signal preprocessing and enhancement; S5: Perform frequency domain analysis on the preprocessed signal using Fast Fourier Transform, detect the main peak signal within the target frequency range, and calculate the signal-to-noise ratio. S6: Based on autocorrelation analysis, perform consistency evaluation on the detected periodic signals to determine whether they conform to the preset flicker coding characteristics; S7: Calculate the confidence level of the recording behavior by combining the signal-to-noise ratio of the main peak signal within the target frequency range and the periodic consistency evaluation results. When the confidence level exceeds the preset threshold, an alarm is triggered.

2. The method for monitoring cinema piracy based on detecting invisible light flicker according to claim 1, characterized in that, Step S1 specifically includes: S11, convert the input video frame into a grayscale image; S12, the grayscale image is divided into video frames of multiple detection regions according to a fixed step size, and the size of each detection region is [missing information]. Pixel.

3. The method for monitoring cinema piracy based on detecting invisible light flicker according to claim 1, characterized in that, Step S2 specifically includes: S21, calculate the average brightness of each region using formula (1): （1） in, The average brightness of the area. For the first in the region The grayscale value of each pixel; S22, calculate the luminance difference between adjacent frames using formula (2): （2） in, For brightness difference; S23, use formula (3) to count the number of changes in the sign of the brightness difference: （3） in, For indicator functions, The sign change represents the change in the direction of the brightness signal fluctuation; S24, regions that meet the conditions shown in formula (4) are identified as candidate flash regions: （4） in, For historical brightness sets, The threshold for brightness change. This is the lower limit threshold for the number of sign changes.

4. The method for monitoring cinema piracy based on detecting invisible light flicker as described in claim 1, characterized in that, Step S3 specifically includes: S31, Based on the spatial location of the candidate flash region, the Euclidean distance between the region location in the current frame and the existing ROI is calculated using formula (5): （5） in, This represents the position of the candidate region in the current frame. Historical position of existing ROI; S32, when When the current candidate region is matched to the corresponding ROI, Match the preset position threshold; S33, Extract the brightness time series from the successfully matched ROI regions. , For time frame index.

5. The method for monitoring cinema piracy based on detecting invisible light flicker according to claim 1, characterized in that, Step S4 specifically includes: S41, use formula (6) to perform detrending processing on the brightness time series to eliminate the DC component in the signal: （6） in, The mean of the signal. The signal length; S42, using formula (7) to design the passband of a bandpass filter to cover the preset range of invisible light flicker frequencies: （7） in, and These are the lower and upper limits of the flicker frequency range, respectively. The video sampling frame rate; S43, after the detrended signal is bandpass filtered, it is then windowed using formula (8): (8)。 6. The method for monitoring cinema piracy based on detecting invisible light flicker according to claim 1, characterized in that, Step S5 specifically includes: S51 uses Fast Fourier Transform to convert the windowed signal to the frequency domain and obtain the spectral energy distribution. ; S52, within the preset target frequency range The highest peak value in the internal search is denoted as the main peak. The corresponding frequency is ; S53, calculate the main peak signal-to-noise ratio using formula (10): （10） in, This represents the average background noise energy outside the target frequency range.

7. The method for monitoring cinema piracy based on detecting invisible light flicker according to claim 1, characterized in that, Step S6 specifically includes: S61, perform autocorrelation analysis on the signal using formula (11): （11） in, For time delay; S62, Extracting peak interval sequences from autocorrelation functions Calculate the standard deviation of the peak interval. ; S63, calculate the periodic consistency score using formula (12): （12）。 8. The method for monitoring cinema piracy based on detecting invisible light flicker according to claim 1, characterized in that, Step S7 specifically includes: S71, fused frequency domain peak signal-to-noise ratio and periodic consistency score The confidence level of the recording behavior is calculated using formula (13): （13） in, For normalization function, and These are the weighting coefficients. The signal-to-noise ratio threshold; S72, if The system detects unauthorized recording and triggers an alarm. This is a preset alarm threshold.

9. The method for monitoring cinema piracy based on detecting invisible light flicker according to claim 1, characterized in that, The invisible light flicker includes spectral components in the infrared and / or ultraviolet bands, and its flicker frequency is preset in the range of 20Hz to 200Hz. The flicker frequency is set according to the characteristics of the invisible light signal actively emitted by the recording device.

10. A device for detecting cinema piracy based on detecting invisible light flicker, characterized in that, Applied to a server, the device includes: The video capture module is used to acquire real-time video streams of the audience seating area captured by the cinema's surveillance cameras; The region division module is used to divide the input video frame into multiple monitoring regions; The candidate region filtering module is used to calculate the brightness change characteristics of each region and filter candidate flash regions; The tracking module is used to continuously track candidate flash areas and extract brightness time-series signals; The signal processing module is used to perform detrending, bandpass filtering, and windowing processing on the luminance time series signal; The frequency domain analysis module is used to calculate the signal spectrum through fast Fourier transform and to detect the main peak signal and calculate the signal-to-noise ratio within the target frequency range; The periodicity assessment module is used to assess the periodicity consistency of a signal through autocorrelation analysis. The confidence calculation module is used to calculate the confidence level of the recording behavior by combining the signal-to-noise ratio of the main peak signal within the target frequency range and the periodic consistency evaluation results. The alarm module is used to trigger an alarm when the confidence level exceeds a preset threshold.

Images