JP4483049B2 - Multimedia content registration system - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention relates to a registration apparatus for various types of multimedia information (hereinafter referred to as contents) such as sound, image, text, and the like.
[0002]
[Prior art]
Due to the spread of the Internet and the improvement of the capability of terminals connected to the Internet, music distribution companies and image distribution companies are distributing contents to each terminal free of charge or for a fee. In addition, it has become possible to exchange contents between terminals or register contents in a specific server, and a user who accesses the contents can download the contents for free or for a fee.
[0003]
For example, in a system in which content is distributed to a user from a music distribution company, content for each singer is registered on a website, and a user who accesses this site clicks on the desired singer's content for a fee. Or you can download it for free. In addition, the content created by the user can be registered in a specific Web site, and the user who accesses this site can download it to his / her terminal by clicking the title of the content for a fee or free of charge.
[0004]
[Problems to be solved by the invention]
However, in a system that distributes content from a music distribution company or image distribution company according to the user's wishes, since royalty payments are usually made to the copyright holder, there is a problem with copyright law. Although not considered, there are many cases where the copyright law does not provide for the exchange of contents among general users.
[0005]
However, if such a situation is left unattended, it will not be possible to prevent the occurrence of troubles under the Copyright Act, and if content registration is prohibited, content distribution will be hindered. It will hinder the development of industry and culture through the network.
[0006]
In addition, as methods for guaranteeing and authenticating the copyright of content, methods using digital watermark technology and encryption technology have been proposed, but the identity and similarity of content using these technologies is also proposed. In order to discriminate and eliminate something very similar to the original content, a complicated algorithm for decoding digital watermarks and encryption is required, and the number of judgment steps tends to be enormous.
[0007]
Therefore, an object of the present invention is to automatically determine not to register the same or similar contents with respect to already registered contents, and to reduce the number of man-hours for the determination and determine relatively easily. It is an object of the present invention to provide a multimedia content registration apparatus that realizes the above.
[0008]
[Means for Solving the Problems]
The present invention is configured as follows to solve the above-described problems.
(1) a content database in which multimedia content including music composed of a plurality of parts is registered;
A characteristic parameter database storing extracts motif data different parts music of each multimedia content registered in the content database as a characteristic parameter,
Extracts each part of the motif data of the music of registered candidate multimedia content to be newly registered in the content database as a characteristic parameter, wherein the parameter to calculate the degree of similarity between the characteristic parameters of the feature parameter database similar Degree calculation means,
The similarity is divided into high, medium and low according to a predetermined threshold,
In each similarity calculated by the feature parameter similarity calculation means,
If there is a high similarity, the registration candidate multimedia content registration is rejected,
If there is no high similarity and there is medium similarity, the registration of the registration candidate multimedia content is suspended,
If there is no medium similarity, registration determination means for permitting registration of the registration candidate multimedia content is provided.
[0009]
According to the present invention, a server is provided with a feature parameter database in which multimedia data including a piece of music composed of a plurality of parts is registered , and motif data for each part of the piece of music of the content is extracted and stored as a feature parameter. ing. The detection of a musical motif is typically performed by extracting the melody data of the first four bars or the melody data from the first bar until the next stop form appears, A data string from which rests are deleted can be used as a musical motif. From a user, and extracted the content (registered candidate content) wishing to register it is transmitted via the Internet or the like, the part-specific motif data of the music of the registered candidate content at the server as a characteristic parameter, wherein the parameter database All the similarities with each feature parameter stored in the table are calculated , and divided into high, medium and low according to a predetermined threshold value . Among all the similarities obtained in this way, it is automatically determined whether there is a high similarity that is considered to infringe the copyright under the copyright law. if so, we reject the registration of the registered candidate content.
[0010]
Also, if the similarity is neither high nor low and is in a gray zone (a range of similarities that does not make it clear whether the infringing rights are infringed), registration is suspended, the user is warned, and registration is performed. Or registration permission or registration refusal processing based on the judgment of the administrator.
[0011]
(2) In the present invention, when the registration determination unit permits registration, the registration candidate multimedia content is registered in the content database, and the feature parameter of the registration candidate multimedia content is stored in the feature parameter database. Registration means for storing is further provided.
[0013]
DETAILED DESCRIPTION OF THE INVENTION
FIG. 1 is a configuration diagram of a content registration system according to an embodiment of the present invention.
[0014]
The registration system includes a server 1 provided with a database 2 in which contents and feature parameters are registered, and a terminal connected to the server 1. Although the terminal shows the portable terminals 1 and 2 in the figure, the terminal is not limited to this and includes a terminal composed of a personal computer or the like. Each terminal is connected to the server 1, and the server 1 includes a database server, a Web server, and the like. The server and the terminal can be connected via a computer network such as the Internet or a public line by wire or wireless.
[0015]
In FIG. 1, the database 2 includes a content database 20 and a feature parameter database 21. The content database 20 stores the content itself, the user ID that created the content, and the like, and the feature parameter database 21 stores the feature parameters extracted from each of the contents corresponding to each content.
[0016]
In FIG. 1, each mobile terminal (mobile terminal 1, mobile terminal 2) can create content on the terminal. Now, when the portable terminal 1 creates content and tries to register it in the content database 20, it issues a registration request to the server 1. When the server 1 receives the registration request, the server 1 performs the following processing using the content as the registration candidate content.
[0017]
(A) Extract feature parameters of registration candidate contents.
[0018]
(B) Each feature parameter stored in the feature parameter database 21 is compared with the feature parameter of the registration candidate content, and the degree of similarity between them is calculated.
[0019]
(C) Whether or not registration candidate contents are registered is determined based on each similarity.
[0020]
In (c), if there is no high similarity, registration permission is issued for the registration candidate content. As a result, the server 1 registers the content in the content database 20 and the feature parameter in the feature parameter database 21. Further, the mobile terminal 1 is notified that registration is permitted.
[0021]
Similarly, when the mobile terminal 2 creates content and issues a registration request to the server 1, the server 1 performs the processes (a) to (c). If there is a high similarity between the feature parameter of the registration candidate content and each feature parameter stored in the feature parameter database 21, registration of the content is rejected. That is, the content is not registered in the content database 20 and the mobile terminal 2 is notified that registration is rejected.
[0022]
The server 1 automatically performs the above processing. As a result, there is no content database 20 that has a high degree of similarity.
[0023]
Next, feature parameters extracted from the content will be described. In this embodiment, the feature parameters extracted from the content are all related to the musical sound data.
[0024]
FIG. 2 shows a first embodiment. This figure shows the configuration of the database 2, but in this embodiment, the file size and hash value of the content are extracted as feature parameters and stored in the feature parameter database 21. By using the feature parameter as the file size and hash value, the calculation of the similarity becomes extremely simple. That is, if the file size and hash value are the same as the file size and hash value of the registration candidate content, it can be estimated that both contents are exactly the same.
[0025]
FIG. 3 shows an embodiment in which a file size for each part and a hash value for each part are extracted as feature parameters and stored in the feature parameter database 21. In this embodiment, since the degree of similarity for each part can be calculated, registration can be rejected even if any part such as a melody part or a bass part is imitated.
[0026]
Note that a CRC used for error code detection can be used instead of the hash value.
[0027]
FIG. 4 shows an embodiment in which a motif is extracted as a feature parameter and stored in the feature parameter database 21. Music motifs can usually be extracted from the melody part. There are various ways to extract motifs. For example, the melody data of the first four bars of a song can be extracted as motif data, or the melody data from the first stop bar to the first bar can be used as a motif. It can also be extracted as data. In addition, data obtained by deleting short notes and rests below a specified value from the data extracted in this way can be used as motif data. Furthermore, it is possible to discard the length information and use only the pitch + rest sequence as motif data. It is also possible to ignore the pitch and the order of appearance of rests, total the appearance frequency of each note, and use this as motif data. Further, from this data, it is possible to delete the short notes and rests below the specified value and finally calculate the appearance frequency of each note as the motif data. In addition, any method may be used as long as it can express the motif of the music in a data string.
[0028]
FIG. 5 shows the configuration of a database when part-specific motif data is extracted as a feature parameter and stored in the feature parameter database 21. Similar to the embodiment of FIG. 3, in order to store the motif data for each part, it is possible to further determine the presence or absence of imitation.
[0029]
FIG. 6 shows the configuration of a database when data based on lyrics data is extracted as a feature parameter and stored in the feature parameter database 21.
[0030]
As the data based on the lyric data, for example, a data string in which all the lyric data is hiragana, or data obtained by dividing the lyric into words and totaling the appearance frequency of the appearing words can be used as the feature parameter. The embodiment shown in FIG. 6 is not used to calculate the similarity of a melody or the like of a song, but is used to calculate the similarity of lyrics.
[0031]
The feature parameter database 21 as shown in FIGS. 2 to 6 is prepared in advance, and in this state, the feature parameters of the registration candidate content requested for registration from the terminal are stored in the feature parameter database 21. Compare with the parameters one by one. When calculating the similarity, if the feature parameter stored in the feature parameter database 21 is a reference parameter, the spatial distance of the registration candidate content feature parameter with respect to the reference parameter is calculated, and this value is output as the similarity. Various known methods such as a dynamic programming method for obtaining a spatial distance by absorbing fluctuations in the time axis direction and a hidden Markov method can be employed. The calculated similarity is preferably classified by using a predetermined threshold value. For example, if the similarity is classified into A to C, A is a classification with a high similarity, B is a classification with an intermediate degree of similarity, and C is a classification with a low similarity, for example, as shown in FIG. A list of similarities to the content is displayed. In the figure, No at the left end indicates the content number. In this example, the first and second contents have a low similarity (C), the third content has a high similarity (A), and the fourth content has an intermediate similarity. Yes. Note that the similarity being intermediate (B) indicates that the similarity is in a gray zone that is neither high nor low.
[0032]
In the server 1, if there is a similarity A on the similarity list obtained in this way, the registration candidate content is not registered. On the other hand, if there is a similarity B, a hold process described later is performed. Only when all are similarities C, the registration candidate content is registered.
[0033]
If there is a similarity B, the registration candidate content is suspended. The hold processing in the present embodiment means processing that does not register the content immediately, but performs registration after giving a warning or notification to the user, or leaves it to the administrator's judgment. The content of the hold process depends on the accuracy of the calculated similarity, the system operation rules, and the like. For example, if the accuracy of the calculated similarity is low, it can be said that it is desirable to leave the decision of registration to the administrator in order to eliminate the risk of registering very similar content. On the other hand, if the accuracy is high, registration is performed, but one method is to notify the user that the similarity is in the gray zone.
[0034]
FIG. 8 is a flowchart showing the above processing in time series. This operation is performed in the server 1.
[0035]
When there is a content registration request from a certain terminal (step ST1), the requested content is extracted as a registration candidate content (ST2). Next, the similarity between the extracted feature parameter and all the feature parameters stored in the feature parameter database 21 is calculated, and a list as shown in FIG. 7 is created. Next, the list is referenced to determine whether there is a similarity A (ST4) or a similarity B (ST5). If there is a thing of similarity A, registration is refused, the fact is notified to the terminal, and the process ends (ST6). If there is no similarity A and there is similarity B, a hold process is performed for the content (ST7), and the process ends. If none of the similarities A and B exist, the registration candidate content is registered in the content database 20, and the feature parameter is stored in the feature parameter database 21, and registration of the terminal is permitted. Notification is performed (ST8).
[0036]
In the above operation, the feature parameter extraction of the registration candidate content is performed in the server 1, but this process may be performed in each terminal. In that case, when making a registration request from the terminal to the server 1, the content and the extracted feature parameters are transmitted simultaneously.
[0037]
Further, the content may be an image as well as music information such as music. In this case, the feature parameter of the image is stored in the feature parameter database, and the feature parameter of the image is extracted from the registration candidate content.
[0038]
【The invention's effect】
According to the present invention, since the feature parameter database storing the feature parameters of each content together with the content database is provided on the server side, the content requested for registration from the terminal is similar to the content stored in the content database. It is possible to easily determine whether or not For this reason, in particular, distribution including registration and use of contents via the Internet is promoted.
[Brief description of the drawings]
FIG. 1 is a schematic configuration diagram of a multimedia content registration system according to an embodiment of the present invention. FIG. 2 is a database configuration example. FIG. 3 is another database configuration example. FIG. 5 shows another configuration example of the database. FIG. 6 shows another configuration example of the database. FIG. 7 shows a similarity list. FIG. 8 shows a schematic operation of the server.

Claims (2)

A content database in which multimedia content including music composed of multiple parts is registered;
A characteristic parameter database storing extracts motif data different parts music of each multimedia content registered in the content database as a characteristic parameter,
Extracts each part of the motif data of the music of registered candidate multimedia content to be newly registered in the content database as a characteristic parameter, wherein the parameter to calculate the degree of similarity between the characteristic parameters of the feature parameter database similar Degree calculation means,
The similarity is divided into high, medium and low according to a predetermined threshold,
In each similarity calculated by the feature parameter similarity calculation means,
If there is a high similarity, the registration candidate multimedia content registration is rejected,
If there is no high similarity and there is medium similarity, the registration of the registration candidate multimedia content is suspended,
If there is no medium similarity, registration determination means for permitting registration of the registration candidate multimedia content ;
A multimedia content registration system comprising: When the registration determination unit permits registration,
The multimedia content registration according to claim 1, further comprising registration means for registering the registration candidate multimedia content in the content database and storing a feature parameter of the registration candidate multimedia content in the feature parameter database. system.

Images