KR20090001760A - System and method for inspection of adult moving video by audio identification - Google Patents

Abstract

An adult substance video censorship system and a method through audio discrimination are provided to put censorship on adult videos based on contents of audio, thereby blocking an adult substance effectively and reducing censorship errors more with small system loads. An adult video indexing DB(180) indexes audio fingerprints of adult video files. A file input part(100) inputs a censorship target video file. An audio characteristic detecting unit(110) extracts audio fingerprints by analyzing audio data included in the inputted video file. An adult video checking unit(120) inspects whether the video file corresponds to an indexed adult substance.

Description

System and Method for inspection of adult moving video by audio identification}

The following drawings attached to this specification are illustrative of the preferred embodiments of the present invention, and together with the detailed description of the invention to serve to further understand the technical spirit of the present invention, the present invention is a matter described in such drawings It should not be construed as limited to.

1 is a block diagram showing a schematic configuration of a system for censoring adult video by audio identification according to a first embodiment of the present invention.

2 is a block diagram illustrating a schematic configuration of a system for censoring adult video by audio identification according to a second embodiment of the present invention.

3 and 4 are flowcharts illustrating a method of operating an adult video censoring system through audio identification according to a first embodiment of the present invention.

5 and 6 are flowcharts illustrating a method of operating an adult video censoring system through audio identification according to a second embodiment of the present invention.

7 is an internal block diagram of a general-purpose computer system that may be employed to perform the method of operation of the adult movie inspection system according to the present invention.

<Main reference number in drawing>

100: file input unit 110: audio feature extraction unit

120: adult video inspection unit 130: manager censor

140: video index unit 150: general video inspection unit

160: adult video index DB 170: adult video storage

180: normal video index DB 190: normal video storage

The present invention relates to an adult video censoring system, and more particularly, to a system and a method for identifying an adult video by identifying audio included in a video.

With the rapid development of information and communication technology, the speed of data transmission over the Internet is getting faster every day. Accordingly, a video streaming service for real-time education, broadcasting, movies, shopping, etc., as well as a download service of a large video file, has become possible.

The emergence of various video services has positively indicated that the internet service has been improved quantitatively, but on the other hand, illegal distribution of adult videos using file sharing services or messenger programs linked with web hard services is emerging as another social issue. .

Adult videos are legally produced, but most of them are illegally made in foreign countries or illegally shot on a video chat site or private space without the consent of the other party. In particular, in the latter case, the act of shooting a movie itself is illegal, and when a video file is distributed through the Internet, it can cause a serious mental shock to the subject, and when the subject is a celebrity, it causes a social wave. .

In recent years, digital cameras and digital camcorders that can shoot video. As camera phones are spreading, adult video in the form of User Created Contents (UCC) is being added, and the problem of adult video is becoming more serious.

It can not be overemphasized that adult video has an adverse effect on the mental health of minors. In addition, if adult video is distributed through a public website, such as an Internet portal, it severely impacts the reliability and morality of the website operator. Therefore, there is an urgent need for a technology that can effectively detect whether a video distributed through the Internet corresponds to an adult video.

In the conventional adult video censorship technology, the video censor inspects the contents of the video manually to determine whether the adult video or not, and by analyzing the color (yellow, black, white, etc.) of the video, the specific gravity of the specific color is thresholded. The color screening method (Korean Patent Laid-Open Patent No. 2004-46537) that judges the adult video when exceeding, builds a DB of hash values of various adult video and compares the hash value of the video to be inspected with the hash value recorded in the DB. Thus, if the similarity exceeds a threshold, a hash value screening method (Korean Patent No. 10-517645), etc., which is determined as an adult video.

However, the manual censoring method takes a lot of time to perform video censorship, so it is not effective for censoring a large number of video files. In addition, the color censoring method has a problem in that a censoring error occurs due to lighting of a moving picture photographing place or clothing of an actor, resulting in inferior reliability. In addition, the hash value detection method extracts a part of clips of an adult video separately and edits them secondly, or censors properly when the encoding environment of the video file, such as the type of encoding algorithm, the tag inserted in the file, the sampling rate, and the bit rate, is changed. There is a problem that is not made.

Accordingly, the present invention has been made to solve the above-mentioned problems of the prior art, and provides a reliable adult video screening system and method based on the actual content of a video without relying on the encoding environment of the video file. There is this.

Another object of the present invention is to provide an adult video screening system and method capable of censoring not only copied adult video but also adult video of various formats produced by the same audio stream.

It is still another object of the present invention to provide a system and method capable of censoring secondary adult videos.

Still another object of the present invention is to provide an adult video censoring system and method that can reduce the burden of video censorship when constructing a video database for providing various video services.

In accordance with an aspect of the present invention, there is provided a system for censoring adult video through audio identification, comprising: an adult video index DB in which an audio fingerprint of an adult video file is indexed; A file input unit for receiving a censored video file, an audio feature extractor for analyzing an audio data included in the input video file, and extracting an audio fingerprint; and an audio fingerprint corresponding to the extracted audio fingerprint. And an adult video inspection unit for searching whether the video file corresponds to an indexed adult material by inquiring whether the DB is indexed or not.

Preferably, the adult video inspection unit inspects the video file by inquiring whether the extracted audio fingerprint and the audio fingerprint having a similarity or higher than a threshold are indexed in the adult video index DB.

The system according to the present invention, if it is determined that the video file is not an indexed adult material as a result of the censorship, and displays the image contained in the video file to the censorship manager, the manager censor and the manager to receive the type of the video from the censorship manager If the moving image file is designated as an adult through the censor, the extracted audio fingerprint may further include a moving image indexing unit for newly indexing the adult index DB.

According to an aspect of the present invention, there is provided a general video index DB in which an audio fingerprint of a general video file is indexed; And if the video file is designated as a non-adult through the manager censor, check whether the audio fingerprint corresponding to the extracted audio fingerprint is indexed in the general video index DB, and the other general video indexed with the video file. It may further include a general video inspection unit for inspecting whether or not overlap. If it is determined that the video file is not an indexed general video, the video indexing unit indexes the extracted audio fingerprint into a general video index DB.

According to another aspect of the invention, the general video index DB, the audio fingerprint of the general video file is indexed; And a general video inspecting unit for inspecting whether an audio fingerprint corresponding to the extracted audio fingerprint is indexed in the general video index DB. The adult video inspecting unit inspects the general video when the video file is determined to be non-adult. The request may be output to the general video inspecting unit, and the general video inspecting unit may perform general video inspecting according to the censoring request.

In this case, if it is determined that the video file is not an indexed general video as a result of the normal video censoring, the inspector inspecting unit and the inspector inspecting the video included in the video file to the censoring manager and receiving the type of video from the censoring manager The video file may further include a video index unit for newly indexing an audio fingerprint extracted from the adult video index DB if the video file is designated as an adult content and a general video index DB if the video file is designated as an adult content.

In the present invention, the video index unit may also index the header and meta information of the video file when new indexing of the extracted audio fingerprint.

Another technical problem of the present invention can be achieved by the above-described method of operating an adult video screening system through audio identification and a computer-readable recording medium that records the program.

Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings. Prior to this, terms or words used in the specification and claims should not be construed as having a conventional or dictionary meaning, and the inventors should properly explain the concept of terms in order to best explain their own invention. Based on the principle that can be defined, it should be interpreted as meaning and concept corresponding to the technical idea of the present invention. Therefore, the embodiments described in the specification and the drawings shown in the drawings are only the most preferred embodiment of the present invention and do not represent all of the technical idea of the present invention, various modifications that can be replaced at the time of the present application It should be understood that there may be equivalents and variations.

1 is a block diagram showing a schematic configuration of a system for censoring adult video by audio identification according to a first embodiment of the present invention.

Adult video screening system according to the first embodiment, the file input unit 100, audio feature extraction unit 110, adult video inspection unit 120, manager screening unit 130, video index unit 140, general video inspection unit 150, an adult video index DB 160, an adult video storage 170, a general video index DB 180, and a general video storage 190.

The file input unit 100 receives a video file to be censored and delivers it to the audio feature extraction unit 110. Here, censorship refers to determining whether the video file corresponds to adult content. Adult content means various harmful videos such as pornography, violence, and cruelty that have harmful effects on adolescents. The video file has a video-only format such as mpeg, avi, mov, and the like. A video file is a file transmitted over a network. For example, the video file is a file transmitted through wired or wireless Internet, a file transmitted through mail reception, or a file transmitted through P2P communication. Alternatively, the movie file is a file stored within computer storage. In the latter case, the file input unit 100 preferably provides a user interface for the censorship manager to select at least one video file. The video file selected through the user interface is selected as a censored file. The computer on which the system according to the invention can be installed is a personal computer or a server computer. Server computers include content servers capable of wired and wireless Internet services, search portal servers, and mail servers.

The audio feature extractor 110 extracts an audio fingerprint by analyzing audio data included in a video file. Here, audio fingerprint refers to a kind of audio fingerprint that can exclusively and uniquely identify audio data. Preferably, the audio feature extraction unit 110 prior to the process of selectively separating only the audio data from the video file before extracting the audio fingerprint.

The audio feature extractor 110 extracts an audio fingerprint by analyzing an audio stream in the video file. That is, the audio feature extraction unit 110 obtains audio data included in the video file and divides the audio data into frame units through window analysis. Then, Fourier transform is performed to extract the frequency characteristics for each frame. When the window is analyzed, a Hamming / Hanning window may be used. Subsequently, the audio feature extractor 110 performs frequency and time-base filtering based on the energy sequence of the filter bank output of the power spectrum for each frame to which the Fourier transform is applied, and then beats each frequency band energy based on a specific threshold value. Is assigned to generate an audio fingerprint of a predetermined number of bits (eg, 32 bits).

Preferably, the filter bank uses a Mel / bark scale filter bank in order to consider human hearing characteristics. However, the present invention is not limited thereto. On the other hand, frequency and time-base filtering are performed to minimize the correlation between the frequency adjacent band and the adjacent frame. That is, frequency and time base filtering is performed to minimize the correlation by enlarging the difference between adjacent band energies and the difference between adjacent frames. Preferably, the frequency filtering is performed using a second pass FIR filter type in the form of a band pass filter (BPF) that minimizes the correlation of energies, and the time-base filtering uses a well-known RASTA filter type to reduce channel distortion. By using it, an audio fingerprint robust to external noise and channel noise can be extracted. However, the present invention is not limited by the scheme of frequency and time base filtering.

The above-described audio fingerprinting technique is disclosed in Park Man-soo's article "Frequency-Temporal Filtering for a Robust Audio Fingerprinting Scheme in Real-Noise Environments" (ETRI Journal, vol. 28, no.4, Aug. 2006, pp509-512). It is. Accordingly, all content of the article can be incorporated as part of the construction of the present invention. On the other hand, the present invention is not limited by the specific manner of extracting the audio fingerprint, any algorithm that can extract the audio fingerprint can be employed. For example, the audio feature extractor 110 may extract an audio fingerprint using a stochastic pattern model by applying general audio features.

The adult video inspection unit 120 receives an audio fingerprint extracted from the audio feature extraction unit 110. Then, by checking whether the audio fingerprint corresponding to the extracted audio fingerprint is indexed in the adult video index DB 160, whether the video file corresponds to adult content is inspected.

If there is an audio fingerprint corresponding to the extracted audio fingerprint, the video file corresponds to adult content. On the contrary, if there is no audio fingerprint corresponding to the extracted audio fingerprint, the video file corresponds to at least indexed adult content. Do not.

The case where the audio fingerprints correspond to each other means that the similarity between the audio fingerprints exceeds a threshold. Therefore, when the video file is determined to be adult, it means that an audio fingerprint of which similarity with the extracted audio fingerprint exceeds a threshold is already indexed in the adult video index DB 160.

Here, the similarity determination between the audio fingerprints may be based on a Hamming Distance (HD) value of the audio fingerprint. As an example, it may be determined whether the files are identical based on the ratio of the fingerprints in which the HD value is '0' between the audio fingerprint sequences extracted between the two file sections. Alternatively, a full match between audio fingerprint sequences extracted between two file regions can be used to calculate the overall average bit error rate (BER) based on the HD value to determine similarity between audio fingerprints based on that value. It may be. Of course, the temporal order of the two files must be taken into account because the fingerprints extracted for each frame are compared. On the other hand, the similarity determination between audio fingerprints in the stochastic pattern modeling technique may be based on distance scale functions such as ED (Euclidean Distance), CAD (Cosine Angle Distance), and SRE (Symmetric Relative Entropy).

As described above, if an adult fingerprint of a video file is discriminated using an audio fingerprint, an effective adult substance censor is possible. If the audio fingerprint extracted from the video file is the same as the already-indexed audio fingerprint, then the video file has already been identified as adult and the content of the audio stream is the same. This is because files can be considered adult.

The adult video inspection unit 120 skips the indexing of the video file when the video file input through the file input unit 100 is determined to be the indexed adult material. Instead, the extracted audio fingerprint and the video file may be stored in separate storage spaces of the adult video index DB 160 and the adult video storage 170. It is apparent that the separate storage space may be set in a separate DB and storage from the adult video index DB 160 and the adult video storage 170.

When the adult video inspection unit 120 determines that the video file is not a pre-indexed adult material, as a result of the censoring of the video file based on the audio fingerprint, the manual censoring process by the censorship manager 130 is performed. You can optionally proceed.

That is, the manager censor 130 receives a manual censor for the video file from the adult video inspection unit 120. Then, the manager censor 130 visually displays the video file determined to be not the indexed adult material to the censor, and receives the type of the video from the censor.

The visual display of the moving image file may be performed by playing the moving image file or capturing and continuously arranging still images of the moving image at regular frame intervals from the moving image file. The latter display method has the advantage that the video censoring time required is small because the entire contents of the video can be checked at a glance. However, the present invention is not limited to the visual display method of the video file.

The inspector inspecting unit 130 may perform the following process according to how the inspector designates the type of the video file. That is, when the censorship manager designates the video file as adult content, the manager censor 130 requests a new indexing of the video file to the video indexer 140. Then, the video index unit 140 newly indexes the audio fingerprint extracted from the video file into the adult video index DB 160. Furthermore, the indexed video file is stored in the adult video storage 170. Meanwhile, although not essential, the video indexer 140 may also index the header and meta information extracted from the video file when the audio fingerprint is indexed.

On the contrary, if the censorship manager designates a video file as a normal video, the manager censor 130 requests a new indexing as the normal video to the video indexer 140. Then, the video index unit 140 newly indexes the audio fingerprint extracted from the video file to the general video index DB 180. Furthermore, the indexed video file is stored in the general video storage 190. Meanwhile, although not essential, the video indexer 140 may also index the header and meta information extracted from the video file when the audio fingerprint is indexed.

Unlike the above description, when the video file is designated as a normal video, the manager inspecting unit 130 may perform a file redundancy inspection procedure before DB indexing and storage of the video file. That is, if the inspector inspector designates the video file as a general video, the manager inspector 130 requests the file to be overlaid to the general video inspector 150. Then, the general video inspecting unit 150 searches whether the audio fingerprint corresponding to the audio fingerprint extracted from the video file is indexed in the general video index DB 180 and performs file duplication inspection. Here, the concept of correspondence between audio fingerprints has been described above.

As a result of file duplication inspection, if it is determined that the extracted audio fingerprint is not indexed in the general video index DB 180, the general video inspecting unit 150 requests a new indexing of the video file to the video indexing unit 140. . Accordingly, the video indexer 140 newly indexes the extracted audio fingerprint in the general video index DB 180 and stores the video file in the general video storage 190.

On the contrary, if it is determined that the extracted audio fingerprint is already indexed in the general video index DB 180, the general video inspecting unit 150 includes a video file whose content is substantially duplicated with the video file. It is assumed to be stored in. In this case, skipping DB indexing of audio fingerprints and general video storage storage. Instead, the extracted audio fingerprint and the video file may be stored in separate storage spaces of the general video index DB 180 and the general video storage 190, respectively. It is apparent that the separate storage space may be set in a separate DB and storage from the general video index DB 180 and the adult video storage 190.

2 is a block diagram illustrating a schematic configuration of a system for censoring adult video by audio identification according to a second embodiment of the present invention.

The adult video screening system according to the second embodiment includes the same configuration as the system according to the first embodiment described above, the arrangement order of the adult video screening unit 120, the manager screening unit 130 and the general video screening unit 150 is Are different.

That is, in the first embodiment, the components are arranged in order of the adult video inspection unit 120, the manager inspector 130, and the general video inspection unit 150, but in the second embodiment, the adult video inspection unit 120 and the general video are provided. Components are arranged in the order of the inspector 150 and the manager inspector 130. Therefore, the adult video screening system according to the second embodiment continuously performs adult video screening and file duplication screening, and finally performs manual screening by the censorship manager.

Specifically, when the adult video inspection unit 120 determines that the audio fingerprint extracted from the video file is not a pre-indexed audio fingerprint as a result of searching the adult video index DB 160, the general video inspection unit 150 is on the side. Requests duplicate file detection based on normal video.

Then, the general video inspecting unit 150 searches whether the audio fingerprint corresponding to the audio fingerprint extracted from the video file is indexed in the general video index DB 180 and performs file duplication inspection.

As a result of file duplication inspection, if it is determined that the extracted audio fingerprint is already indexed in the general video index DB 180, the general video inspecting unit 150 stores a video file whose content is substantially duplicated with the video file. It is assumed to be stored at 190. In this case, DB indexing of the audio fingerprint and storage storage of the video file are omitted. Instead, the extracted audio fingerprint and the video file may be stored in separate storage spaces of the general video index DB 180 and the general video storage 190, respectively. It is apparent that the separate storage space may be set in a separate DB and storage from the normal video index DB 180 and the normal video storage 190.

On the other hand, if it is determined that the extracted audio fingerprint is not indexed in the general video index DB 180, the general video inspecting unit 150 requests manual inspection by the censoring manager 130. Then, the manager censor 130 visually displays the video file to the censorship manager and receives a video type from the censorship manager. Here, the visual display method of the video file has already been described above.

If the censorship manager designates the video file as adult content, the manager censor 130 requests a new indexing of the video file to the video indexer 140. Then, the video index unit 140 newly indexes the audio fingerprint extracted from the video file into the adult video index DB 160. Furthermore, the indexed video file is stored in the adult video storage 170. Meanwhile, although not essential, the video indexer 140 may also index the header and meta information extracted from the video file when the audio fingerprint is indexed.

On the contrary, if the censorship manager designates a video file as a normal video, the manager censor 130 requests a new indexing as a normal video from the video indexing unit 140. Then, the video index unit 140 newly indexes the audio fingerprint extracted from the video file to the general video index DB 180. Furthermore, the indexed video file is stored in the general video storage 190. Meanwhile, although not essential, the video indexer 140 may also index the header and meta information extracted from the video file when the audio fingerprint is indexed.

The adult video censoring system according to the second embodiment described above performs the manual censoring by the censorship manager after continuously performing adult video censorship and file duplication censoring, and thus reduces the burden on the censorship manager compared to the first embodiment. There is this.

3 and 4 are flowcharts illustrating a method of operating an adult video censoring system through audio identification according to a first embodiment of the present invention. The execution subject of each step is the adult video screening system shown in FIG.

First, an adult video index DB in which audio fingerprints of a large amount of adult video files are indexed and a general video index DB in which audio fingerprints of a plurality of general video files are indexed are provided (step S101).

Subsequently, in response to the adult video index DB being provided, the video file to be inspected is input (step S102). Video files can be input via the network or from storage inside the computer.

Then, the audio stream is separated from the input video file (step S103). Then, the separated audio stream is analyzed to extract an audio fingerprint (step S104). The extraction method of the audio fingerprint has already been described above.

Next, the adult video index DB is queried to determine whether an audio fingerprint corresponding to the extracted audio fingerprint is indexed in the adult video index DB (S105).

As a result of the determination in step S105, if the corresponding audio fingerprint is indexed, the moving image file is identified as adult material, and new indexing of the extracted audio fingerprint and storage storage of the moving image file are omitted (step S106). Then, the adult video censoring procedure for the video file ends. Alternatively, the extracted audio fingerprint and the video file are moved to separate storage spaces allocated to the adult video index DB and the adult video storage, and the adult video censoring procedure for the video file is terminated.

In contrast, if the corresponding audio fingerprint is not indexed, an image included in the moving image file is displayed to the censoring manager (step S107). The display of an image may be based on a method of playing a moving image file or a method of capturing and continuously arranging still images of a moving image at regular frame intervals.

Subsequently, the type of the moving image is determined by the type of the moving image displayed by the censorship manager (step S108). As a result of the determination, if the type of the video is adult, the extracted audio fingerprint is newly indexed in the adult video index DB, and the video file is newly stored in the storage (step S109). When indexing an audio fingerprint, the header and meta information of the video file can be indexed together. On the contrary, if the type of the video is not adult content, the normal video index DB is searched to determine whether the audio fingerprint corresponding to the extracted audio fingerprint is indexed in the general video index DB and then duplicate file inspection is performed (step S110). .

As a result of the duplicate file inspection in step S110, if the corresponding audio fingerprint is indexed, the moving image file is identified as a duplicate general moving image, and new indexing of the extracted audio fingerprint and storage storage of the moving image file are omitted (step S111). Then, the duplicate file inspection process is terminated. Alternatively, the extracted audio fingerprint and the video file may be moved to separate storage spaces allocated to the general video index DB and the general video storage, and then the duplicate file screening process may be terminated.

On the contrary, if the corresponding audio fingerprint is not indexed, the video file is identified as a new general video, the extracted audio fingerprint is newly indexed in the general video index DB, and the video file is newly stored in the general video storage (S112). . When indexing an audio fingerprint, the header and meta information of the video file can be indexed together.

As described above, when the adult video censorship, manual censoring by the censorship manager and the file duplication screening process based on the general video are finished, the entire process of the adult video censoring method according to the first embodiment of the present invention is completed.

5 and 6 are flowcharts illustrating a method of operating an adult video censoring system through audio identification according to a second embodiment of the present invention. The execution subject of each step is the adult video screening system shown in FIG.

As shown in the figure, providing an adult video index DB and a general video index DB (step S101), receiving a censored video file (step S102), and separating an audio stream from the video file (step S103). ), Extracting the audio fingerprint from the separated audio stream (step S104), censoring the video using the adult video index DB (step S105), and extracting the audio fingerprint when the video file is determined to be adult Omitting the new indexing and storage storage of the video file (step S106) is substantially the same as the adult video clip detection method according to the first embodiment described above. However, if the video file is not determined to be adult, the subsequent process is different from that of the first embodiment.

That is, if the video file is not determined as adult in step S105, the general video index DB is searched to determine whether the audio fingerprint corresponding to the extracted audio fingerprint is indexed in the general video index DB, and then duplicate file inspection is performed. (Step S113).

As a result of the duplicate file inspection in step S113, if the corresponding audio fingerprint is indexed, the moving image file is identified as a duplicate general moving image, and new indexing of the extracted audio fingerprint and storage storage of the moving image file are omitted (step S114). Alternatively, the extracted audio fingerprint and the video file may be moved to separate storage spaces allocated to the general video index DB and the general video storage.

On the contrary, if the corresponding audio fingerprint is not indexed, an image included in the video file is displayed to the censoring manager (step S115). The display of an image may be based on a method of playing a moving image file or a method of capturing and continuously arranging still images of a moving image at regular frame intervals.

Subsequently, the type of the moving image is determined by the type of the moving image displayed by the censorship manager (step S116). As a result, if the type of the video is adult, the extracted audio fingerprint is newly indexed in the adult video index DB, and the video file is newly stored in the adult video storage (step S117). On the contrary, if the type of video is a normal video, the extracted audio fingerprint is newly indexed in the general video index DB and the video file is newly stored in the general video storage (S118). When indexing an audio fingerprint, the header and meta information of the video file can be indexed together.

As described above, when censoring adult video, duplicate file censoring based on a general video, and manual censoring by the censorship manager are completed, the entire process of censoring adult video according to the second embodiment of the present invention is completed.

The operation method of the adult movie screening system according to the present invention is implemented in the form of program instructions that can be executed by various computer means can be recorded in a computer-readable medium. The computer readable medium may include program instructions, data files, data structures, etc. alone or in combination. Program instructions recorded on the media may be those specially designed and constructed for the purposes of the present invention, or they may be of the kind well-known and available to those skilled in the computer program arts. Examples of computer-readable recording media include magnetic media such as hard disks, floppy disks, and magnetic tape, optical media such as CD-ROMs, DVDs, and magnetic disks, such as floppy disks. Hardware devices specially configured to store and execute program instructions such as magneto-optical media and ROM, RAM, flash memory and the like. The medium may be a transmission medium such as an optical or metal wire, a waveguide, or the like including a carrier wave for transmitting a signal specifying a program command, a data structure, or the like. Examples of program instructions include machine code, such as produced by a compiler, as well as high-level language code that can be executed by a computer using an interpreter. Such hardware devices may be configured to operate as one or more software modules to perform the operations of the present invention, and vice versa.

7 is an internal block diagram of a general-purpose computer system that may be employed to perform the method of operation of the adult movie inspection system according to the present invention.

The general purpose computer system 400 includes one or more processors 410 connected to a main memory including a RAM 420 and a ROM 430. The processor 410 is also called a central processing unit (CPU). As is well known in the art, the ROM 430 serves to transmit data and instructions to the processor 410 unidirectionally, and the RAM 420 typically transfers data and instructions to the processor 410 bidirectionally. Used to transmit RAM 420 and ROM 430 may include any suitable form of computer readable media. The mass storage device 440 is bidirectionally coupled to the processor 410 to provide additional data storage capability, and may be any of the computer readable recording media described above. The mass storage device 440 is used to store programs, data, and the like, and is a secondary memory device such as a hard disk which is generally slower than the main memory device. Certain mass storage devices such as CD ROM 460 may be used. The processor 410 may include one or more input / output interfaces 450 such as video monitors, trackball mice, keyboards, microphones, touchscreen displays, card readers, magnetic or paper tape readers, voice or handwriting readers, joysticks, or other known computer input / output devices. ). Finally, the processor 410 may be connected to a wired or wireless communication network through the network interface 470. Through this network connection, the procedure of the method described above can be performed. The apparatus and tools described above are well known to those skilled in the computer hardware and software arts. On the other hand, the hardware device described above may be configured to operate as one or more software modules to perform the operations of the present invention.

As described above, although the present invention has been described by way of limited embodiments and drawings, the present invention is not limited thereto and is intended by those skilled in the art to which the present invention pertains. Of course, various modifications and variations are possible within the scope of equivalents of the claims to be described.

According to the present invention, since adult content videos are censored based on the content of audio, adult content can be effectively blocked. In particular, the load of the system is smaller and the detection error can be further reduced compared to the conventional technology of analyzing the color of the image and the like to censor adult material. In addition, if the contents of the audio stream are the same, the adult video can be inspected even if the encoding environment of the video file is changed or the video file is secondarily edited. Furthermore, since the indexing of the adult video is continuously performed, the burden of the censorship manager can be continuously reduced in proportion to the number of the adult video indexed in the construction of the video database for providing the video service. In addition, the use of storage space can be achieved by integrating adult video screening and file duplication screening.

Claims (28)

An adult video index DB in which an audio fingerprint of an adult video file is indexed; File input unit for receiving censored video files An audio feature extraction unit for extracting an audio fingerprint by analyzing audio data included in the input video file; And an adult video inspection unit for searching whether the audio fingerprint corresponding to the extracted audio fingerprint is indexed in the adult video index DB and inspecting whether the video file corresponds to the indexed adult video. Censorship system for adult video with audio identification. The method of claim 1, The adult video inspection unit checks whether the extracted audio fingerprint and the audio fingerprint having a similarity or higher than a threshold value are indexed in the adult video index DB and inspects the video file. Screening system. The method of claim 1, And the adult video inspection unit omits the indexing operation of the video file when the video file is determined to be the indexed adult material. The method of claim 1, If it is determined that the video file is not an indexed adult material, the inspector inspecting unit displays an image included in the video file to the censorship manager and receives a type of video from the censorship manager. And a video indexing unit for newly indexing the extracted audio fingerprint into an adult index database when the video file is designated as an adult through the inspector inspecting unit. The method of claim 4, wherein And the manager inspecting unit displays a video to the inspecting manager in a manner of capturing and continuously arranging the still images of the videos at regular frame intervals from the video file. The method of claim 4, wherein And the video indexing unit indexes the extracted audio fingerprint and the header and meta information of the video file together. The method of claim 4, wherein A general video index DB in which an audio fingerprint of the general video file is indexed; If the video file is designated as a non-adult through the manager censor, it is checked whether the audio fingerprint corresponding to the extracted audio fingerprint is indexed in the general video index DB, and the other video is indexed with the general video index. And a general video inspecting unit for inspecting whether or not there is a duplication. The method of claim 7, wherein And if it is determined that the video file is not an indexed normal video, the video indexing unit newly indexes the extracted audio fingerprint into the general video index DB. The method of claim 8, And the video indexing unit indexes the extracted audio fingerprint and the header and meta information of the video file together. The method of claim 1, A general video index DB in which an audio fingerprint of a general video file is indexed; And And a general video inspecting unit for inspecting whether an audio fingerprint corresponding to the extracted audio fingerprint is indexed in the general video index DB. The adult video inspection unit outputs a general video censoring request to the general video inspection unit when the video file is determined to be non-adult; The general video inspection unit according to the censoring request, the adult video detection system through audio identification, characterized in that for performing a general video censor. The method of claim 10, And the general video inspecting unit omits the indexing operation of the video file if it is determined that the video file is a pre-indexed general video. The method of claim 10, If it is determined that the video file is not an indexed general video as a result of the normal video censoring, the manager censor for displaying the video included in the video file to the censoring manager and receiving the type of the video from the censoring manager; And a video indexing unit for newly indexing the audio fingerprint extracted from the adult video index DB if the video file is designated as adult through the manager censor, and extracted from the general video index DB if the video file is designated as adult. Adult movie screening system through audio identification. The method of claim 12, And the manager inspecting unit displays a video to the inspecting manager in a manner of capturing and continuously arranging the still images of the videos at regular frame intervals from the video file. The method of claim 12, And the video indexing unit indexes the extracted audio fingerprint and the header and meta information of the video file together. (a) providing an adult video index DB in which an audio fingerprint of an adult video file is indexed; (b) receiving a video file to be inspected (c) extracting an audio fingerprint by analyzing audio data included in the input video file; and (d) checking whether the audio fingerprint corresponding to the extracted audio fingerprint is indexed in the adult video index DB and inspecting whether the video file corresponds to adult content How to censor adult video by identification. The method of claim 15, The step (d) is a step of inspecting whether the extracted audio fingerprint and the audio fingerprint having a similarity or more than a threshold value is indexed in the adult video index DB, and censoring the video file. How to censor adult video. The method of claim 15, In the step (d), if it is determined that the video file is a pre-indexed adult content, the indexing of the video file is skipped, characterized in that the adult video detection method through audio identification. The method of claim 15, If it is determined that the video file is not an indexed adult material, the image included in the video file is displayed to the censorship manager; Receiving a type of a video from the censorship manager; and And if the type of the designated video is adult, further extracting the extracted audio fingerprint into an adult index DB. 2. The method of claim 18, The image display step of the adult video detection method through the audio identification, characterized in that the step of displaying the video to the inspection manager in a way that captures a continuous image of the video at a predetermined frame interval from the video file in a continuous list. The method of claim 18, And indexing headers and meta-information of the video file when new indexing of the audio fingerprint is performed. The method of claim 18, Providing a generic video index database in which the audio fingerprint of the generic video file is indexed; and As a result of the inspecting of the manager, if the video file is designated as a non-adult, it is checked whether the audio fingerprint corresponding to the extracted audio fingerprint is indexed in the general video index DB, and the other video is indexed with the general video index. And censoring whether or not there is a duplication. The method of claim 21, And if it is determined that the video file is not an indexed general video, the extracted audio fingerprint is newly indexed in the general video index DB. The method of claim 22, And indexing headers and meta information of a video file when new indexing of the extracted audio fingerprint is performed. The method of claim 15, Providing a general video index DB in which the audio fingerprint of the general video file is indexed If it is determined that the video file is non-adult, as a result of the adult video censorship, the process shifts to a general video censorship process. And censoring whether the audio fingerprint corresponding to the extracted audio fingerprint is indexed in the general video index DB. The method of claim 24, And if it is determined that the video file is an indexed general video as a result of the general video censoring, the indexing operation of the video file is omitted. The method of claim 24, If it is determined that the video file is not an indexed general video, the image included in the video file is displayed to the censorship manager; Receiving the type of video from the censorship manager; and And if the video file is designated as an adult, newly indexing an audio fingerprint extracted from the adult video index DB and if the video file is designated as a non-adult, the video fingerprint DB. How to censor a video. The method of claim 26, The image display step, characterized in that the step of capturing a still image of the video at a predetermined frame interval from the video file and displaying the video to the censor manager in a continuous way to display the adult video through audio identification. The method of claim 26, And the audio fingerprint extracted in the indexing step and the header and meta information of the video file are indexed together.

Images