JP4948667B2 - Video content alteration detection device and system - Google Patents

Description

  The present invention relates to a video content alteration detection device and system, and in particular, a digital watermark detection device, a detection method, and a detection program for detecting alterations such as deletion, addition, and replacement of digital video content using a digital watermark. The present invention relates to a video content falsification verification system and a video content structure.

  Digital video content may be used as evidence video in trials. However, since these digital video contents are easy to process, it is necessary to prove that there is no difference from the original when stored for a long period of time. As a technique for proving the authenticity of content, an electronic signature technique and a digital watermark technique have been put into practical use.

  The creation of an electronic signature requires exchange of data with a certificate authority or the like under high security, which may take time and effort. On the other hand, in the digital watermark, if the authenticity of the content is proved by using the digital watermark technology, the watermark information can be distributed and embedded throughout the content, so even if the encoding format is converted, the embedded watermark information Is hard to disappear. Therefore, when content is stored for a long time, a method of proving authenticity using a digital watermark technique is preferable.

Regarding the use of the digital watermark technology, Patent Documents 1 and 2 disclose a technique for embedding information in content including video data and audio data by digital watermark.
In the technique of Patent Document 1, in order to automatically correct the time lag between video and audio during broadcasting or in use, the same time information is digitally watermarked on the transmitting side on the synchronized digital video and digital audio. The time information extracted from the digital video and the digital audio is compared with each other on the receiving side, and the time lag between the video and the audio is detected and corrected according to the comparison result.

  Further, the technique of Patent Document 2 embeds a feature value obtained from content header information in video data or audio data on the transmission side with a digital watermark, and extracts watermark information embedded in the content on the reception side. A technique for detecting falsification of content when the extracted watermark information and header information do not match is disclosed.

JP 2003-259314 A JP 2003-122726 A

I. Echizen, H. Yoshiura, T. Arai, H. Kimura, T. Takeuchi: General Quality Maintenance Module for Motion Picture Watermarking, IEEE Trans. Consumer Electronics, Vol. 45, No. 4, pp. 1150-1158, ( 1999).

Examples of alteration of video content include deletion, addition, and replacement of video data. For example, you can prevent the detection of medical mistakes by replacing a part of the image of public construction that was delayed in construction with the video of the completion of construction or deleting a part of the image that shot the state of surgery. There is also concern that the video content may be altered for the purpose of illegal acts.
However, the technique of Patent Document 1 is not a technique for detecting falsification of content in the first place, and there is no suggestion about detection of falsification when the content of digital content is deleted, added, or replaced.

  The technique of Patent Document 2 can detect alteration of header information and replacement of video data or audio data by comparing the feature value of the header embedded in the video data or audio data with the header information of the content. However, as in Patent Document 1, tampering when data is partially deleted, added, or replaced cannot be detected.

An object of the present invention is to detect tampering such as deletion, addition, and replacement of digital video content using a digital watermark.
An object of the present invention is to detect tampering of digital video content by managing regularity information embedded in digital video content as electronic watermarks and detecting the regularity information later.

The present invention repeatedly detects the alteration of video content by repeatedly embedding content identification information and regular information such as a time code as a digital watermark in a set of consecutive frame images and detecting the regularity information. To do.
According to a preferred example, the digital watermark detection apparatus according to the present invention is a digital watermark detection apparatus that detects a digital watermark embedded in a moving image content composed of a plurality of frame data, and is a predetermined number of digital watermarks. Detection means for detecting regularity information (regularity information) preliminarily embedded by digital watermark from continuous frame data by digital watermark, and when digital watermark is not detected from a predetermined number of continuous frame data A digital watermark detection apparatus comprising: means for calculating the number of undetected times; and means for determining tampering with a predetermined number of frame data based on the detected regularity information and the number of undetected times. The
Preferably, the detection means detects sequential information or time code having order and embedded as regularity information.
Preferably, the regularity information is embedded in a predetermined first predetermined number of continuous frame data with a digital watermark, and the additional information is added to a predetermined second predetermined number of continuous frame data. Embedded with a digital watermark, the detection means detects regularity information and additional information.
Preferably, the detection unit detects at least one of identification information, position information, and right information of moving image content embedded in advance as additional information in the frame data.

The digital watermark detection method according to the present invention is preferably a digital watermark detection method for detecting a digital watermark embedded in moving image content composed of a plurality of frame data, from a predetermined number of consecutive frame data. A step of detecting regularity information embedded in advance with a digital watermark, a step of calculating the number of undetected times when a digital watermark is not detected from a predetermined number of consecutive frame data, and a detected regularity And a step of determining falsification to a predetermined number of frame data based on the information and the number of undetected times.
Preferably, deletion, addition, or replacement of moving image content is detected by detecting a digital watermark including regularity information.
In one example, after the predetermined number N of acquired frames are divided into sub-segments composed of N / 2 frames that are continuous from the video content, a detection window is set in the sub-segment to detect a digital watermark, It is determined whether or not a digital watermark has been detected from the detection window. If the result of determination is that a digital watermark has not been detected, the watermark non-detection count nD is incremented by 1, and then the video content read position is set to that sub-segment. Set to the reading position of the adjacent sub-segment and proceed to detection processing of the next detection window. If the result of determination is that a digital watermark is detected, the watermark non-detection count nD is set to 0 and the regularity information is updated. To do.

  The program for detecting a digital watermark according to the present invention is preferably a program for detecting a digital watermark for detecting a digital watermark embedded in a moving image content composed of a plurality of frame data. The step of detecting regularity information embedded in advance by digital watermark from a predetermined number of predetermined frame data by digital watermark, and if no digital watermark is detected from the predetermined number of continuous frame data, it is not detected As a digital watermark detection program executable on a computer for executing a step of calculating the number of times and a step of determining falsification to a predetermined number of frame data based on the detected regularity information and the number of undetected times Composed.

The video content falsification verification system according to the present invention is preferably a system including a content processing device that processes video content, and a content verification device that is connected to the content processing device via a network and detects falsification of the video content. The content processing apparatus has means for acquiring regularity information from the outside or inside, and means for embedding the obtained regularity information in a predetermined number of continuous frame data of the moving image content with a digital watermark. And
The content verification device includes a detecting unit that detects regularity information from a predetermined number of consecutive frame data of a moving image content by using a digital watermark, and a digital watermark is not detected from the predetermined number of continuous frame data. A video content falsification verification system comprising: means for calculating the number of undetected times; and means for determining tampering with a predetermined number of frame data based on the detected regularity information and the number of undetected times Is done.
Preferably, the means for embedding embeds regularity information in a predetermined first predetermined number of consecutive frame data with a digital watermark, and further adds it to a predetermined second predetermined number of consecutive frame data. Information is embedded with a digital watermark.
Preferably, the embedding unit embeds sequential information or time code having order as regularity information, and the detection unit detects sequential information or time code.
Preferably, the embedding unit embeds at least one of identification information, position information, and right information of moving image content as additional information, and the detection unit includes identification information, position of moving image content embedded as additional information. At least one of information and right information is detected.

  The structure of the moving image content using the digital watermark according to the present invention is preferably a moving image content structure in which the digital watermark is embedded for verification of the moving image content, and the moving image content is predetermined. Ordered regularity information is embedded in the first predetermined number of consecutive frame data with a digital watermark, and the moving image content identification information is further included in a predetermined second predetermined number of consecutive frame data. , Additional information including at least one of position information and right information is embedded with a digital watermark.

  According to the present invention, tampering of digital video content can be detected with higher accuracy.

The figure which shows the structure of the content delivery system in one Embodiment. FIG. 5 is a diagram showing an example of embedding a digital watermark in video content by the digital watermark embedding device 22. FIG. 3 is a block diagram showing an example of a detailed functional configuration of the digital watermark embedding device 22. FIG. 3 is a block diagram showing an example of a detailed functional configuration of a digital watermark detection apparatus 34. Explanatory drawing which shows the structural example of the embedding information by the digital watermark embedding apparatus 22. FIG. 6 is a flowchart illustrating an example of an operation for embedding a digital watermark. The figure which shows typically the detection processing of the electronic watermark at the time of fraudulent addition of content. The figure which shows typically the detection processing of the digital watermark at the time of the illegal deletion of a content. The figure which shows typically the detection processing of the digital watermark at the time of unauthorized replacement | exchange of content. The figure which shows an example of the Attack determination standard which the digital watermark detection apparatus 34 refers. 7 is a flowchart showing an example of a digital watermark detection processing operation in the digital watermark detection apparatus.

Embodiments of the present invention will be described below with reference to the drawings.
FIG. 1 is a diagram illustrating a configuration of a content distribution system 10 according to an embodiment. The content distribution system 10 includes a content storage / distribution server 12, a content processing device 20, and a content verification device 30 connected to a communication network 11. For example, the content processing device 20 is provided by a content production company, and the content verification device 30 is provided by a content screening facility.

The content processing device 20 includes a digital watermark embedding device 22, an encoder 23, a storage device 24, and a transmission device 25. Here, the content encoding process by the encoder 23 is not necessarily required, and the digital watermark may be directly embedded in the content originally encoded by the digital watermark embedding device 22.
In the content processing device 20, digital video content acquired from a content generation source 21 such as a video device or a camera is input via the communication network 11, and the digital watermark embedding device 22 embeds a digital watermark in this digital video content. The transmission device 25 transmits the content embedded with the digital watermark to the content storage / distribution server 12 via the communication network 11. Note that transfer of content generated by the content processing apparatus 20 may be stored in the server 12 using a medium such as a DVD, a wireless network, or a broadcast wave without going through the communication network 15.

The content verification device 30 includes a reception device 31, a storage device 32, a decoder 33, and a digital watermark detection device 34. Here, the decoding process of the content by the decoder 33 is not necessarily required, and the digital watermark may be directly detected from the content encoded by the digital watermark detection device 34.
The content verification device 30 receives content embedded with a digital watermark from the content storage / distribution server 12 or the content processing device 20 via the communication network 11 and temporarily stores it in the storage device 32. Here, regarding the transfer of content, the content may be received using a medium such as a DVD, a wireless network, or a broadcast wave without going through the communication network 15.
The decoder 33 decodes the content stored in the storage device 32, and the digital watermark detection device 34 detects a digital watermark from the decoded content and outputs a determination result.

  The video content created by the content processing device 20 has a structure as shown in FIG. 2, for example. FIG. 2 schematically shows the digital watermark embedding in the video content. As shown in FIG. 2A, the video content is time-divided as segments composed of N frames (for example, 30 frames per second). Different time codes t1 to tk are embedded in each segment, but the same time code is embedded in each N frame constituting the segment. Here, the time code is, for example, time information consisting of year / month / day / hour / minute / second. However, the time information is not necessarily required, and a number indicating the order of the segments may be used. In general, it may be sequential information with order.

In the example shown in FIG. 2B, in addition to the time codes t1 to tk, additional information such as a content identification ID and GPS information is embedded. Here, the additional information is, for example, an ID (device ID) unique to a video device or a camera, and GPS is position information. The reason why the device ID is embedded is to verify that video content shot by another camera is added to the video content. The reason why the GPS is embedded is to verify that video content shot at another location is added to the video content.
As an example, as shown in FIG. 5, the M-bit embedded information is composed of time information of T (<M) bits and additional information of the remaining MT bits.
The additional information does not need to be embedded on a segment basis, and the same information may be embedded over a plurality of segments, or information obtained by time-sharing the additional information may be embedded over a plurality of segments.

  FIG. 3 is a block diagram illustrating an example of a detailed functional configuration of the digital watermark embedding apparatus 22. The digital watermark embedding device 22 includes a time code acquisition unit 221, a watermark information generation unit 222, a content acquisition unit 223, a watermark information embedding unit 224, and a content output unit 225. These functional units are realized by hardware or software.

Here, the time code acquisition unit 221 acquires a time code from a content generation source such as a video device or a camera, or a time information transmitter inside or outside the apparatus.
The watermark information generation unit 222 generates embedded information from the time code obtained from the time code acquisition unit 221. As shown in FIG. 5, time information is assigned to T (<M) bits of the M-bit embedded information. Additional information such as an identification ID of a content generation source such as a camera or video equipment, an identification ID of content itself, or GPS information is assigned to the remaining MT bits as necessary.

The content acquisition unit 223 acquires video content output from the content generation source. The watermark information embedding unit 224 embeds the watermark information generated by the watermark information generation unit 222 in the video content acquired by the content acquisition unit 223 according to the embedded structure of FIG. The content output unit 225 outputs the watermark embedded video content generated by the watermark information embedding unit 224 to the outside of the apparatus.
Note that an example of a method of embedding watermark information in video data using the digital watermark technique is described in detail in Non-Patent Document 1, and will not be described in detail here.

FIG. 4 is a block diagram illustrating an example of a detailed functional configuration of the digital watermark detection apparatus 34.
The digital watermark detection apparatus 34 includes a content acquisition unit 341, a watermark information detection unit 342, a detection information storage buffer 343, a determination unit 344, and a determination result output unit 345.
The content acquisition unit 341 acquires video content embedded with a digital watermark from the content storage / distribution server 12 or the content processing apparatus 20 via the communication network 11.

The watermark information detection unit 342 divides the acquired video content into N-frame segments, and then detects a digital watermark for each segment. The detection information storage buffer 343 stores and updates additional information such as time codes and IDs for a plurality of continuous segments detected by the watermark information detection unit 342.
The determination unit 344 refers to the information stored in the detection information storage buffer 343 and determines the consistency of information detected between a plurality of consecutive segments. Specifically, it is determined whether or not tampering such as deletion, addition, or replacement of video content is possible. The determination result output unit 345 outputs the determination result determined by the determination unit 344 to the outside of the apparatus.

FIG. 6 is a flowchart showing an example of the operation of the digital watermark embedding device 22.
For the video content input from the content generation source 21, the digital watermark embedding device 22 starts the operation shown in this flowchart.
First, the time code acquisition unit 221 acquires time information as a time code from a content generation source such as a video device or a camera or a transmitter of time information inside or outside the apparatus, and temporarily stores it in a storage area in the time code acquisition unit. (S101). In addition, the content acquisition unit 223 acquires a predetermined number of frames of video content output from the content generation source, and temporarily stores them in the storage area in the time code acquisition unit 223 as well.

  Next, the watermark information generation unit 222 acquires time information stored in a storage area in the time code acquisition unit 221 and generates embedded information (S102). Then, the watermark information embedding unit 224 acquires a segment composed of N frames continuous from the video content temporarily stored in the storage area of the content acquisition unit 223, and applies a watermark information generation unit to a frame image of N frames. The watermark information generated in 222 is embedded according to the embedded structure of FIG. 2 (S103).

Then, the content output unit 225 determines whether or not the watermark information has been embedded in all the segments of the video content (S104). As a result of the determination, when the watermark information is not embedded in the entire video content, the time code acquisition unit 221 and the content acquisition unit 223 execute the process shown in step S101 again.
On the other hand, if the watermark information is embedded in all the segments in the content data, the content output unit 225 notifies the encoder 24 that the watermark information has been embedded (S105). With this notification, the digital watermark embedding apparatus 22 ends the control operation shown in this flowchart.
Here, in the case where the video is video content input in real terms such as a camera, the above determination may be performed by the output stop process of the digital watermark embedding device 22.
Note that the processing of each of the above steps can also be realized by executing a program for performing the processing on a computer constituting the digital watermark embedding device 22.

FIG. 7 schematically illustrates an example of watermark detection processing.
In this example, another video content composed of M frames is illegally inserted and added between the segment in which the time code information t1 is embedded and the segment in which t2 is embedded.
First, the parameters in the detection information storage buffer 343 are initialized. That is, the time code information (TCcur) detected from the target detection window, the time code information (TCprev) detected from the detection window immediately before the target detection window, and the watermark undetected count (nD) are all set to 0. set.

Next, after reading a predetermined number of frames, a detection window is set for the first sub-segment divided into N / 2 frames (detection window 1). In the watermark detection process, a frame accumulation process in which pixel-by-pixel addition is performed on a frame image composed of N / 2 frames in the detection window, and then a watermark is detected from the generated one added watermark image. .
In the example of FIG. 7, since the time code information = t1 is embedded in the area of the detection window 1, in order to detect “t1” as time code information from the window, TCcur = t1 is set. The process proceeds to the detection processing of the detection window (detection window 2).

Also in the detection window 2, in order to detect “t1” as the time code information, after setting the time code information detected from the detection window 1 to TCprev = t1, the information “detected from the detection window 2 which is the current detection window” t1 ″ is set in TCcur (TCcur = t1), and the process proceeds to detection processing for the next detection window.
In the detection window 3, since the watermark cannot be detected, the time code is not updated. After the undetected count nD is incremented by 1 (nD = 1), the process proceeds to the detection process for the next detection window.
Since no watermark is detected even in the detection windows 4 and 5, the undetected count nD is incremented in the same manner as described above.

  Next, in the detection window 6, since the time code information “t2” is detected, the undetected count nD is set to 0, and the time code information is updated (TCprev = t1, TCcur = t2). At the time of determination of each detection window, an attack (No Attack: No Attack, illegal addition part: Addition, illegal deletion part: Delete, Addition) against the detection window from the three types of parameter information stored in the detection information storage buffer 343 Can be determined. In the determination, an attack on the detection window is determined according to the Attack determination criteria as shown in FIG. If Addition is determined in a certain detection window, the difference between TCprev and TCcur is calculated in the next detection window (detection window 6 in the figure). If the difference is 2 unit time or more, the previous detection window Is changed from Addition to Replacement.

  In this detection process, time code information is detected from each detection window composed of a predetermined number of continuous frame data. At the same time as this detection process, an electronic watermark of additional information such as ID and GPS is added from each detection window. May be detected. In this case, the number of frames constituting the detection window that is the detection target of the time code information and the number of frames of the detection window that is the detection target of the additional information are not necessarily the same and may be different. That is, the detection timing of the time code information and the additional information and the number of frames required for one detection do not have to be the same.

Next, with reference to FIG. 8, a description will be given of a watermark detection process when video content is illegally deleted.
In this example, the segment in which the time code information t3 is embedded and a part of the segments in which t2 and t4 are embedded are illegally deleted (Delete).
By the same processing as in FIG. 7, the parameters in the outgoing information storage buffer 343 are first initialized, and are added pixel-by-pixel to the frame image (detection window) composed of N / 2 frames in the detection window. After performing the frame accumulation process, a watermark is detected from the generated one added watermark image.

  The detection windows 1, 2, and 3 in FIG. 8 are determined as No Attack in accordance with the Attack determination criteria shown in FIG. However, in the detection window 4, since TCprev = t2 and TCcur = t4, this detection area is determined to be “Deletion” in accordance with the Attack determination criterion. As in this example, there may be a delay in the location of the detection window that is determined to be the deletion location and the deletion location, but by checking the vicinity of this determination location, the location that was deleted illegally can be detected reliably. can do.

Next, with reference to FIG. 9, a digital watermark detection process when content is illegally replaced will be described.
In this example, the segment in which the time code information t2 is embedded and a part of the segment in which t2 is embedded are illegally replaced with other contents (Replacement). By the same processing as in FIG. 7, the parameters in the outgoing information storage buffer 343 are first initialized, and are added pixel-by-pixel to the frame image (detection window) composed of N / 2 frames in the detection window. After performing the frame accumulation process, a watermark is detected from the generated one added watermark image.

  The detection windows 1, 2 and 3 in FIG. 9 are determined as No Attack in accordance with the Attack determination criteria shown in FIG. However, in the detection window 4, since the number of undetected times nD = 2, the detection area is determined as Addition as a provisional determination according to the Attack determination criterion. In the detection window 5, which is the next detection window after the window 4, since the Addition determination has been made in the previous detection window 4, the difference between TCprev and TCcur is calculated. Is changed from Addition to Replacement.

Next, the digital watermark detection processing operation in the digital watermark detection apparatus 34 will be described with reference to the flowchart of FIG.
When the video watermark embedded with the digital watermark is acquired from the digital watermark detection device 34, the content storage / distribution server 12, or the content processing device 20 via the communication network 11, the digital watermark detection device 31 stores the digital watermark. The detection processing operation is started.

  First, the detection information storage buffer 343 initializes related parameters (201). That is, the time code information (TCcur) detected from the target detection window, the time code information (TCprev) detected from the detection window immediately before the target detection window, and the watermark undetected count (nD) are all set to 0. set. Next, the content acquisition unit 341 reads a predetermined number of frames stored in the storage area of the digital watermark detection device 34 from the content storage / distribution server 12 or the content processing device 20 via the communication network 11 ( S202).

Next, the watermark information detection unit 342 divides the read video content into sub-segments composed of consecutive N / 2 frames, sets a detection window for the sub-segment by the process shown in FIG. A watermark is detected (S203). Then, it is determined whether or not a digital watermark has been detected from this detection window (S204).
If the result of this determination is that a digital watermark has not been detected, the watermark non-detection count nD is incremented by 1, and then the content read position is set to the read location of the sub-segment adjacent to that sub-segment. The process proceeds to the detection process for the next detection window.
On the other hand, if a digital watermark is detected in step S204, parameters in the detection information storage buffer 343 are set (S205). That is, the watermark undetected count nD is set to 0, and the time code information (TCcur, TCprev) is updated.

Next, the parameters in the detection information storage buffer 343 are read, and the Attack in the detection window that matches the determination criteria is determined in light of the Attack determination criteria as shown in FIG. 10 (S206). Thereafter, the determination result of the detection window is output (S207).
Then, it is determined whether or not the watermark detection process has been executed from all the sub-segments (S208). If not, the sub-segment detection process not executed in step S202 is started. On the other hand, if it is executed, the processing of the digital watermark detection device 34 is terminated.
Note that the processing in each of the above steps can also be realized by executing a program for performing these processing on a computer constituting the digital watermark detection apparatus 34.

Examples of the application of the technique for detecting tampering such as deletion, addition, and replacement of digital video content using the digital watermark of the time code according to the above-described embodiment include the following.
As an example, this embodiment can be applied when storing images of construction objects such as buildings and roads. Although the construction of the building is not yet finished, it is possible to cheat the video of another building that has been constructed and add it to the video of the target building or replace the video of the building. However, as in the present embodiment, such fraud can be detected depending on whether or not the time code embedded by the digital watermark is detected.

  As another example, application to a medical field can be considered. For example, a time code is embedded in a video image of a surgical pattern and stored. When an operation mistake is detected at a later date, even if an attempt is made to delete the image of the error portion, an illegal manipulation of the image can be detected by detecting the time code.

Further, as another example, the present invention can be applied for detecting alteration of a video by a person concerned inside the police.
For example, the time code is embedded in a digital watermark and stored in a video of a police enforcement and tracking site. If an insider of the police falsifies the video shot of the accident site of the violating vehicle, it is possible to detect unauthorized deletion or falsification by detecting the time code embedded in the video.

10: Content distribution system, 11: Communication network, 12: Content storage / distribution server, 20: Content processing device, 21: Content generation source, 22: Digital watermark embedding device, 23: Encoder, 24: Storage device, 25: Transmission Device: 30: content verification device; 31: reception device; 32: storage device; 33: decoder; 34: digital watermark detection device.

Claims (15)

An electronic watermark detection apparatus for detecting an electronic watermark embedded in a moving image content composed of a plurality of frame data,
From a predetermined number of consecutive frame data, detection means for detecting regular information embedded in a digital watermark in advance (referred to as regularity information) using a digital watermark, and from the predetermined number of continuous frame data And a means for calculating the number of undetected times when a digital watermark is not detected, and a means for determining alteration to the predetermined number of frame data based on the detected regularity information and the number of undetected times. A digital watermark detection apparatus. The digital watermark detection apparatus according to claim 1, wherein the detection unit detects sequential information or a time code having order and embedded as the regularity information. The regularity information is embedded in a predetermined first predetermined number of continuous frame data with a digital watermark, and additional information is embedded in a predetermined second predetermined number of continuous frame data with a digital watermark. And
The digital watermark detection apparatus according to claim 1, wherein the detection unit detects the regularity information and the additional information. 4. The digital watermark detection apparatus according to claim 3, wherein the detection unit detects at least one of identification information, position information, and right information of the moving image content embedded in advance as the additional information in the frame data. . In a digital watermark detection method for detecting a digital watermark embedded in a video content composed of a plurality of frame data by a content verification device,
The content verification device includes a content acquisition unit that acquires moving image content, a watermark information detection unit that detects a digital watermark, and a determination unit that determines the consistency of the detection result,
The content acquisition unit acquiring video content embedded with a digital watermark;
The watermark information detecting unit detecting regularity information embedded in advance with a digital watermark from a predetermined number of consecutive frame data;
The watermark information detecting unit calculating the number of undetected times when a digital watermark is not detected from the predetermined number of consecutive frame data;
The determination unit determining whether the predetermined number of frame data has been tampered with based on the detected regularity information and the number of undetected times;
A digital watermark detection method comprising: The watermark information detection unit
6. The digital watermark detection method according to claim 5, further comprising detecting sequential information or time code embedded in the regularity information. The watermark information detection unit
The regularity information embedded in a predetermined first predetermined number of continuous frame data with a digital watermark and the additional information embedded in a predetermined second predetermined number of continuous frame data with a digital watermark Are detected by a digital watermark. 7. The digital watermark detection method according to claim 5 or 6, wherein: The watermark information detection unit
8. The digital watermark detection method according to claim 7, wherein at least one of identification information, position information, and right information embedded in the frame data in advance as the additional information is detected. The determination unit
9. The digital watermark detection method according to claim 5, wherein deletion, addition, or replacement of the moving image content is detected by detecting a digital watermark including the regularity information. The determination unit
After dividing the obtained predetermined number N of frames into sub-segments composed of consecutive N / 2 frames, a detection window is set for the sub-segments to detect a digital watermark,
It is determined whether a digital watermark is detected from the detection window,
If the electronic watermark is not detected as a result of the determination, the watermark non-detection count nD is incremented by 1, and then the reading position of the video content is set to the reading position of the sub-segment adjacent to the sub-segment. Move on to detection process of detection window,
6. The digital watermark detection method according to claim 5, wherein, when a digital watermark is detected as a result of the determination, the regularity information is updated by setting a watermark non-detection count nD to 0. A digital watermark detection program for detecting a digital watermark embedded in a moving image content composed of a plurality of frame data by a content verification device,
The content verification device includes a content acquisition unit, a watermark information detection unit, and a determination unit,
The detection program is:
The watermark information detection unit detects regularity information embedded in advance by a digital watermark from a predetermined number of consecutive frame data,
When the watermark information detection unit does not detect a digital watermark from the predetermined number of consecutive frame data, the number of undetected times is calculated.
Causing the determination unit to determine whether the predetermined number of frame data has been tampered with based on the detected regularity information and the number of undetected times;
A digital watermark detection program. In a system including a content processing device that processes video content, and a content verification device that is connected to the content processing device via a network and detects falsification of the video content,
The content processing apparatus
Means for acquiring regularity information from the outside or the inside, and means for embedding the obtained regularity information in a predetermined number of consecutive frame data of the moving image content with a digital watermark;
The content verification device includes:
Detection means for detecting the regularity information from a predetermined number of consecutive frame data of the moving image content by digital watermark;
Means for calculating the number of undetected times when a digital watermark is not detected from the predetermined number of consecutive frame data;
A tampering verification system for moving image content, comprising: means for determining tampering with the predetermined number of frame data based on the detected regularity information and the undetected number of times. The embedding means embeds the regularity information in a predetermined first predetermined number of continuous frame data with a digital watermark, and further adds additional information to a predetermined second predetermined number of continuous frame data. 13. The falsification verification system according to claim 12, wherein the falsification verification system is embedded with a digital watermark. The embedding means embeds sequential sequential information or time code as the regularity information,
14. The falsification verification system according to claim 12, wherein the detection unit detects the sequential information or the time code. The embedding means embeds at least one of identification information, position information, and right information of the video content as the additional information,
15. The falsification verification system according to claim 12, wherein the detection unit detects at least one of identification information, position information, and right information of the moving image content embedded as the additional information.

Images