HK1021588B - Method and system for managing access to data - Google Patents

Description

Method and system for managing data access

Technical Field

The present invention relates to a method and system for managing access to data by using management information embedded in the data. And more particularly, to a method and system for embedding management information in data and managing access to the data based on the management information when the data is accessed.

Background

With the development of multimedia environments, systems for processing digital information, such as Digital Video Discs (DVDs), Digital Video Cameras (DVCs), digital CS satellite broadcasting, and the internet, are becoming more popular. Also, development of devices capable of processing and transmitting very large amounts of multimedia data at high rates has been greatly advanced. But whether the aforementioned devices will be further expanded depends on the number of applications for which attractive content, such as cinema movies, for example, is available. However, no matter how good the hardware is, the expansion of the hardware will be difficult unless there is a lot of content to attract the customers' desire to purchase.

What becomes a problem in this case is the protection of the work. For example, digital video discs have sufficient hardware based methods in the distribution of theatre movies, but the problem is the protection of content, especially with illegal copies (duplication), where there is no way to provide methods that can be met by the provider providing the content. This is because, although the contents of digital data can be easily copied and changed, effective specifications for preventing such operations are not provided. Therefore, the current state is that many attractive content cannot be expected to be distributed with these media. In fact, since the problem of using copyright is not solved, the digital video camera is merely sold as a dedicated camera. Therefore, effective protection of digitized works is indispensable for further expanding the use of multimedia, and development of such protection techniques and standardized production is expected.

One conventional attempt to control the copying of digital content like music is to use a Copy Generation Management System (CGMS) signal in a digital tape recorder (DAT). The CGMS signal is composed of two bits of data and represents the following reproduction conditions.

CGMS Signal reproduction Condition

11 copy inhibition

00 has no limitation to copy

10 copy once

The method described above writes two data bits to a predetermined location of the data format. When data copying is performed, the above-described CGMS signal is detected at the receiver side. If the content of the signal indicates copy inhibition, the receiver stops copying of the content. However, the writing position can be easily specified and the content can also be easily rewritten, so it is easy for the user to rewrite the content of the signal. That is, it is difficult to effectively prevent illegal copying of contents with the conventional technique because the conventional technique has various disadvantages such that:

appending data management information separate from the data; and

the location of the management information is easily recognized.

Therefore, it is desired to develop a technology capable of effectively controlling access to digital contents and effectively preventing illegal copying.

Disclosure of Invention

In view of the above-described problems, it is an object of the present invention to provide a system capable of controlling access to data and effectively preventing illegal access to data.

To achieve the object, the present invention provides an access management information embedding system for preparing management information for managing access to data, and embedding the management information into the data by data transformation without changing an amount of data, wherein the management information embedded by data transformation without changing an amount of data is extracted in managing access to the data, the access to the data being managed based on the extracted management information. Further, management of access to the data varies depending on the type of medium in which the data is stored, such as RAM (rewritable type), R (write once type), or ROM (read only type). Further, the management information is designed to include information for controlling recording of data, reception of data, playback of data, disturbance of data output, reproduction history of data, or modification of management information.

The recording of control data means that when the management information is embedded in data contained in a medium such as a CD-ROM, the contents of the CD-ROM are permitted or prohibited from being copied to another medium such as a digital video disc or the like.

The reception of the control data means that when the management information is embedded in data input via, for example, radio waves or a network, the data is received by a receiver and output from the receiver is permitted or prohibited.

Controlling playback of data means that when the management information is embedded in data contained in one media type, such as RAM (rewritable type) or R (write once type), the data is permitted or prohibited from being reproduced.

The output interference of the control data means that when the management information is embedded in data contained in a medium such as a RAM or an R type, the data is allowed or prohibited to be output to the outside, or the output interference is controlled by superimposing noise on an output signal.

The copy history of the control data means, for example, that when the management information embedded in the data includes information that allows copying only once, additional information for prohibiting copying is added to the management information instead of the management information after copying is performed once, thereby controlling further copying or playback by means of the management information and the additional information.

Controlling the modification of the management information means, for example, that when the management information embedded in the data includes information that allows copying only once, after copying is performed once, the management information itself is modified to prohibit further copying.

Although the management information relates to recording, copying, reception, playback, copying history, output interference, and the like of data, the present invention can be suitably applied to other management such as restriction on access devices, restriction on access users, expiration of data use, and confirmation information according to the content of embedded management information in accessing data without departing from the spirit of the present invention.

"embedding" as used herein means to hide specific information by deformation of the data itself by using a data hiding technique or the like. Among them, data hiding is a general term of technology for embedding information in another medium (still picture, audio, moving picture, etc.). Unlike encryption, the goal of this technology is not how to hide information, but how to integrate information in one medium that is embedded.

An important feature of data hiding is that it is an invisible marking technique. Since the data hiding does not add a data bit when information is embedded in a medium, but embeds the information imperceptible to human vision by manipulating existing data (data transformation), there is no increase in the total amount of data caused by the embedding of additional information. For example, a storage party can process a media type by embedding text or sound information in an image. One of the other features is the inclusion of embedded information. Since the data hiding directly embeds the additional information in the data structure of the medium, not in the header and special files, the embedded information can be extracted as long as the quality of the original data is maintained, even if the operating platform or data format of the medium is changed.

The method and system of the present invention manages access to data by using management information embedded in the data with such data hiding techniques. It should be noted that this access management method according to the present invention is not limited to a specific data hiding technique, but can be used by any data hiding technique as long as the information embedded by data transformation thereof does not change the data amount.

The present invention provides a method for managing access to data by using management information embedded in the data without changing the amount of the data, wherein the management information includes a plurality of bits of information embedded in said data, comprising the steps of: defining sets of data regions in the data for the plurality of bits of information to be embedded, wherein each data region in each set has at least one pixel and the data regions of each set have a high characteristic correlation therebetween; comparing characteristic values between the data regions of each set and embedding one of the plurality of bit information in each set consisting of data regions by representing one bit information as a function of the characteristic value comparison and one bit value, and extracting a plurality of bit information of the management information embedded in the data by comparing the characteristic values between the data regions in the defined plurality of sets; access to said data is managed based on said extracted management information.

The present invention also provides a system for managing access to data by using management information embedded in the data without changing the amount of the data, wherein the management information includes a plurality of bits of information embedded in said data, comprising: defining sets of data regions in the data for the plurality of bits of information to be embedded, wherein each data region in each set has at least one pixel and the data regions of each set have a high characteristic correlation therebetween; comparing characteristic values between the data regions of each set and embedding one of the plurality of bit information in each set consisting of data regions by representing one bit information as a function of the characteristic value comparison and one bit value, and extracting a plurality of bit information of the management information embedded in the data by comparing the characteristic values between the data regions in the defined plurality of sets; means for managing access to said data based on said extracted management information.

Drawings

FIG. 1 is a schematic diagram of a process for recording media data on a storage medium;

fig. 2 is a schematic diagram for explaining concealment and extraction of data using PBC (pixel block coding);

fig. 3 is a schematic diagram for explaining data concealment using PBC in the case where one pixel includes one pixel block;

FIG. 4 is a schematic diagram showing a process for managing playback of recorded data on a medium;

FIG. 5 is a block diagram of a system for managing access to and outputting data;

FIG. 6 is a diagram illustrating an image arrangement in MPEG;

FIG. 7 is a diagram showing the state of macroblocks arranged in a B picture; and

FIG. 8 is a block diagram of a data access management system according to another embodiment.

Reference numerals indicating the main components correspond to the following:

51 DVD-ROM 52 reader

53 decoder 54 decryption unit

55 MPEG decoder 56 management information extractor

57D/a converter 58 management information modifier

59 switch 60 interference signal generator

61 data provider 62 signal processing system

63 output section

Detailed Description

Fig. 1 is a schematic diagram showing a process (authoring) for recording data on a storage medium. Data such as images, sounds, and software are recorded on a DVD-ROM medium according to the following procedure. The analog data is converted into digital data by analog/digital conversion processing (step 11). The converted digital data is compressed (step 12) with, for example, a moving picture experts group scheme (MPEG). The compressed data is further encoded and encrypted (step 13). Finally the media data is modulated (step 14) for recording on a DVD-ROM medium.

In this embodiment of the present invention, the management information is embedded in the data before, during or after compression, and then recorded on a DVD-ROM device. When a system uses the data recorded on the DVD-ROM, the management information is used to restrict access to the data and to control the system according to the management information. In the present embodiment, management information for managing the copying of data will be described as an example. The management information indicates different states with the following two bits.

Content of state management information

Copy inhibition 11

Unlimited copy 00 (i.e., no data)

Copy once capability 10

The embedding of the management information into the data is performed, for example, between the analog/digital conversion (step 11) and the MPEG compression (step 12) (or after compression). The embedding can be achieved by a technique called data hiding.

Here, as an example, a pixel block coding (hereinafter referred to as PBC) which is a method of embedding management information becoming a target of concealment into certain data and a method of inversely extracting the embedded data will be described.

When PBC is used, data is processed according to a predetermined conversion rule in data hiding and data extraction. Usually, in a certain image, its main characteristics, such as the pixel values of two neighboring pixels, have a high correlation with each other. For example, if there is a "forest" region in a portion of the image, adjacent pixels are very similar green and have a high correlation with each other. Therefore, even if these pixels are replaced with each other, the replacement is rarely visually perceived. That is, even if pixel values having high correlation are operated according to a certain conversion rule, deterioration (of image quality) does not occur in practice.

In view of this characteristic, one bit of data is hidden by defining an image area having at least one pixel as one pixel block (i.e., embedded area) and also by artificially controlling the characteristic of an adjacent pixel block according to a certain conversion rule. The data is represented by a comparison of the characteristics of adjacent blocks of pixels. Also, when data is extracted, the data is extracted according to an extraction criterion, which is determined according to a conversion rule.

The bit information is represented by mutually replacing the characteristic values (e.g., luminance values) of two adjacent pixel blocks according to the following conversion rule.

(conversion rule)

Bit 1: one of the pixel blocks (PB1) has a characteristic value greater than the other (PB2)

Bit 0: one of the pixel blocks (PB1) has a characteristic value smaller than that of the other pixel block (PB2)

Also, the bit information is extracted by comparing feature values (e.g., luminance values) of two adjacent pixel blocks according to the following extraction criteria corresponding to the above-described conversion rule.

(criteria for extraction)

The eigenvalue of one of the pixel blocks (PB1) is greater than the eigenvalues of the others (PB 2): bit 1

The eigenvalue of one of the pixel blocks (PB1) is smaller than the eigenvalues of the others (PB 2): bit 0

Fig. 2 is a schematic diagram illustrating how to perform concealment and extraction of data with the PBC described above. The pixel block PB1 or PB2 may be defined as a group of a plurality of pixels, for example 3 × 3 pixels, and it is also possible to define a single pixel as a single pixel block. Since the adjacent pixel blocks have high correlation, even in the case where the positions of the adjacent pixel blocks are replaced with each other (fig. 2(a)), the image degradation will not be so much as to be visually recognized. Consider the case of the locations of those blocks of pixels in an original image, which are represented in fig. 2 (b). First, the feature values of the two pixel blocks are compared, and it is assumed that the result is that the feature value of PB1 is larger than that of PB 2. When data 1 is hidden in the original image, the feature values of the pixel block have satisfied the condition of data 1 in the conversion rule, so the feature values of the data blocks are not replaced with each other. On the other hand, when data is extracted, the extraction criterion has defined that the case where the feature value of PB1 is greater than PB2 is data 1, so data 1 is extracted.

Further, when data 0 is hidden in the original image, the relationship between the feature values of the pixel blocks in the original image does not satisfy the condition of data 0 in the conversion rule, so the feature values of the pixel blocks are replaced with each other. However, such replacements cannot be visibly identified. When data is extracted, data 0 is extracted from the relationship of the feature values of the data blocks according to an extraction criterion.

In addition to the luminance values mentioned above, the feature value can use a value associated with a main characteristic of a pixel block (i.e. an embedded region) and a value associated with an auxiliary characteristic of the pixel block. The main characteristic is a direct parameter of a pixel value, such as luminance or chrominance of a block of pixels. Furthermore, the auxiliary characteristic is obtained by analyzing the main characteristic, like a value representing a statistical characteristic, such as an average or distribution of the above-mentioned parameters.

Further, it is noted that the pixel block targeted for PBC is not always limited to the adjacent data block. The feature value of one pixel block will be described below using a luminance value as a main characteristic and a distribution value as an auxiliary characteristic as examples.

First, a case where the feature value of one pixel block is one luminance value is described. Since most natural images have a high correlation between adjacent pixels, their substitution with each other will not cause significant deterioration in the image. Fig. 3 is a schematic diagram illustrating a case where six bits of data are embedded in six pixel blocks, one pixel corresponding to one pixel block, according to the above-described PCB.

In the case where a data block containing a relatively large number of pixels is used to improve resistance against image editing or compression, if the luminance values of the pixels between the pixel blocks are replaced with each other, noise such as streaks may occur on the image. Therefore, a method of using the luminance distribution value of the pixel as one characteristic value is considered.

When the characteristics of the luminance values of a pixel block are resolved into an average value h and a distribution value d, the influence on the image in the pixel block may be small in the case where the average value h in the pixel block remains the same and only the distribution value d is replaced as in the case where the average value is simply replaced. Therefore, according to the conversion rule described above, by using the characteristic, selecting the feature value of the pixel block as the distribution value d, and replacing the distribution value, it is possible to transform the hidden information by the data without changing the data amount. See Japanese patent application No.8-159330 (Japanese patent application No. JA9-96-044).

As shown in FIG. 2(c), consider that pixel block PB1 has an average value h₁And the distribution value d₂And a pixel block PB₂Having an average value h₂And the distribution value d₂When one bit 1 is hidden, d₁Is less than d₂So this case does not satisfy the condition of bit 1 of the conversion rule. Thus, only the distribution value d is between two pixel blocks₁And d₂Are replaced with each other. This is equivalent to exchanging the size of the distribution structure without changing the average value h of the characteristic values of the pixels in the data block.

Therefore, the data embedding method using the data hiding technique first specifies an embedding area in which management information is embedded in the data. Subsequently, a conversion rule in which the content to be embedded is made to correspond to the state (characteristic) of one embedding area is generated, and the state (characteristic) of the embedding area is directly controlled in accordance with the management information by referring to the conversion rule. By such direct control, the management information can be embedded in the data. Embedding consistent management information into the raw data is performed by directly controlling the raw data. Therefore, since it is difficult to separate and change only the management information, access management of data can be efficiently performed. Note that details of PBC can be found in japanese patent application No. 8-159330. Such techniques are also collectively referred to as data hiding, electronic watermarking.

The data embedded with management information, which is stored on a DVD-ROM device by using the data hiding technique through the above-described method, is accessed according to the following procedure. Data is first read from the DVD-ROM device. As described in connection with fig. 1, the read-out data is signal modulated (step 14) after the data embedded with the management information is compressed using the MPEG standard (step 11) and after the media data is encoded and encrypted (step 12).

Fig. 4 is a diagram showing a process for managing the data access method. First data with embedded management information is provided to the access system. The data source includes, for example, storage media such as DVD-ROM and CD-R, communication networks such as the Internet, or satellite broadcasting. The provided data is demodulated (step 41) and decoded (step 42). If the data is MPEG compressed, it is decompressed (step 43). The management information embedded in the data is extracted and, if necessary, the content of the management information is modified and only the required modifications are embedded again in the data.

In order to extract management information from data input to the recorder, the above-described extraction process, i.e., one of the data hiding methods, is used. First, an embedding area in which management information is embedded is specified from read data. Subsequently, the copy start information is extracted in accordance with the state of the embedded area by referring to the extraction criterion that makes the state of one embedded area correspond to the content of the data. At this point, if the extracted management information is data 00, the copying of the data is allowed according to the above-described criteria. And if the extracted management information is data 11, it means that copying of the data is prohibited. An access system in this case is controlled to prohibit copying of data. Data 10 means that copying of media data is allowable under a certain condition (only once). This means that copying is allowed to take place on the basis of the data in an original DVD-ROM, but further copying is prohibited on the basis of the copied data.

In this case the access system is controlled such that copying of data is allowed only once and copying of data according to the copy needs to be prevented. Since it is important to prohibit subsequent copying based on the copied data from the original data, when the original data is copied, the copying condition indicated by the management information embedded in the original data is rewritten from data 10 (copy-once margin) to data 11 (copy-prohibited) and then output as one output signal. Since the data as the output signal is embedded with management information indicating a "copy prohibition" condition, the data cannot be further copied based on the copied data. In addition, there is an option to add a bit indicating the experience of copying once. It is important in subsequent playback control to determine whether it is a legitimate copy or an illegitimate copy.

Fig. 5 is a block diagram of a playback system for data stored on a DVD-ROM device. As described above, the DVD-ROM of the DVD-ROM device 51 with respect to one data source provider 61 stores data in which management information is embedded in a data hiding technique. The data of the DVD-ROM device 51 read by the reader 52 in the data supplier 61 is processed by the demodulator 53, the cryptographic decoder 54, and the MPEG decoder 55 constituting a signal processing system 62. The result is decompressed digital data.

The management information extractor 56 specifies an embedding area in which management information output from the MPEG decoder 55 (or directly from MPEG data) is embedded, and also extracts the management information based on the state of the embedding area by referring to an extraction criterion so that the state of an embedding area corresponds to the content of data to be extracted. The D/a converter 57 converts the digitized data from which the management information is deleted into an analog signal. Also, the management information converter 58 is used to change the content of the management information as needed. (where it should be noted that the management information can be extracted either before or after the MPEG decompression.)

The D/a converter 57 in the signal processing system 62 converts the digitized data from which the management information is removed into an analog signal, and outputs an analog playback signal (e.g., NTSC signal). In addition, an interference signal (for example, an analog protection system (Δ PS), such as a micro video signal) generated by the interference signal generator 60 is selectively superimposed on the analog playback signal by the switch 59 in the output section 63. The switch 59 is controlled by a control signal from the management information extractor 56. The output section 63 outputs the analog playback signal or the superimposed signal as an analog output signal.

In this system, the management information extractor 56 outputs a control signal to turn off the switch 59 when the management information extracted by the management information extractor 56 starts the copying of data. In this case, the interference signal is not superimposed on the analog playback signal, and one analog output signal is output as it is.

On the other hand, when the management information prohibits the copying of data, the management information extractor 56 outputs a control signal to turn on the switch 59. In this case, a disturbance signal generated by the disturbance signal generator 60 is superimposed on the analog signal output from the D/a converter 57. Due to the characteristics of the monitor, when it is intended to display a movie image on a monitor based on a signal on which an interference signal is superimposed, a standard movie image can be displayed on the monitor without being affected by the interference signal. However, if it is intended to record movie images using a digital VTR having an analog input terminal, they will be affected by interference signals and standard movie images cannot be recorded. Since the movie image can be reproduced from a signal on which an interference signal is superimposed, copying of data can be effectively prevented, but it cannot be digitized for recording.

Note that when the management information on the video recorder is "copy once" so that when it is recorded, the management information changer 58 changes the management information of the read data to "copy prohibited" and embeds it in the data when written to a subsequent medium. Therefore, according to the copied data, the subsequent data copying can be effectively prevented.

Also, additional information may be used to control copying (duplication) and playback in more detail, in addition to the basic management information. This enables to flexibly provide control according to the type of the recording/playback medium (such as ROM, RAM, or R-type). Playback medium status recording medium management information additional information ROM type RAM/R playback: OK playback: NG replication: NG 110 replay: OK playback: NG replication: NG 111 playback: OK playback: and OK copying: OK 000 playback: OK playback: and OK copying: OK 001N/A playback: and OK copying: NG 100N/A playback: and OK copying: NG 101

0 in the additional information indicates that copying is not performed, and 1 indicates that copying is performed once.

Furthermore, for a playback medium of the ROM type, N/A means that this combination does not exist. That is, by using one-bit information indicating whether or not copying is performed once, in addition to the data information 11 (copy unconditionally prohibited), 10 (copy once), and 00 (copy unconditionally permitted), it is possible to flexibly perform copy/playback control in detail without erasing/modifying the basic management information. The additional information may be embedded by the same method as that used to embed data (management information) in advance, or may be embedded by another method in order to increase security. As another example, the additional information may be provided by using several bits instead of a one-bit design to provide versatile control. In conclusion, it can be similarly implemented without departing from the spirit of the present invention.

Further, the management information is not limited to information about copy restriction. The invention is of course also applicable to various management information, such as control systems. For example, the management information may be the following playback start information.

(Medium type ═ random access memory)

Content of state management information

Playback inhibit 11

Unlimited playback 00

Primary regeneration capacity 10

When the playback start information starts playback of the data, the system is controlled so that playback of the data is permitted. Also, when the playback start information prohibits playback of the data, the system is controlled so that the data is not output from the system.

In addition to the above-described information, the management information may be a useful item that determines a period that can be used for reproducing or recording media data, confirming information for playback of media content, a disc key, or a title key. When data is accessed, any piece of information is necessary.

Another data hiding technique is described as a second embodiment. Note that this embodiment is different from the above-described embodiment in that the embedding of the management information is performed at the time of MPEG encoding, and the extraction is performed at the time of MPEG decoding.

The manner of embedding the management information actually performed in MPEG will now be briefly explained by referring to the data hiding technique of japanese patent 8-272721.

In MPEG, forward prediction from past playback pictures, backward prediction from future playback pictures, and bidirectional prediction using both forward and backward predictions are used.

Fig. 6 is a schematic diagram illustrating an arrangement state of pictures in MPEG. As shown in the figure, MPEG specifies three image frames for bi-directional prediction: i picture, P picture and B picture.

Wherein the I picture is a picture processed in intra coding (intra-coding), and all macroblocks on the picture are processed by intra coding (intra prediction coding). P-pictures are pictures that are processed by forward inter-prediction coding, and sometimes some macroblocks on the picture are processed by intra-coding. Also, the B picture is a picture processed by bidirectional inter prediction encoding. Macroblocks in B pictures are basically backward predicted or bi-predicted macroblocks coded by forward prediction, but may also include intra coding. The picture screen coded over the entire surface by intra-frame coding is an I picture, and the I and P pictures are coded in the same order as the original moving picture image. On the other hand, for B pictures, the I and P pictures are processed, and then B pictures inserted between the processed pictures are encoded.

The embedding area in which the management information is embedded is a macroblock on the B picture, and an information bit can be embedded in one macroblock. Therefore, when the information data is a plurality of bits, it is necessary to perform the embedding process for the macro data blocks corresponding to the plurality of bits in total. Fig. 7 is a diagram showing the state of a macroblock placed on a B picture. The macroblock is a unit that is encoded. For each macroblock, motion compensation is performed with respect to a luminance block having 16 pixels × 16 pixels, and information compression is performed according to the correlation of one time screen as an inter prediction method using motion compensation with respect to one macroblock unit.

The macroblocks on the B picture can be divided into the following four groups according to the prediction type.

Intra macroblock (intra prediction macroblock)

An intra macroblock is a macroblock that is encoded only by information about the screen itself, without performing inter prediction.

Forward predicted macroblock

The forward prediction macroblock is a macroblock encoded in forward prediction by referring to a past I or P picture (reference frame). Specifically, a region of a square having 16 pixels × 16 pixels most similar to that in the past reference frame is retrieved, and the macroblock has information on a prediction error (Δ P) which is a difference value with respect to the retrieved square frame and also has information on a spatial relative position (motion vector). Where the prediction error ap is expressed as a difference between the luminance and the color difference obtained for 16 pixels × 16 pixels. Note that the criteria for selecting a similar square region depends on the encoder.

Backward prediction of macroblocks

A backward prediction macroblock is a macroblock encoded by backward prediction by referring to a future reference frame in display order. An area that is most similar is retrieved among future reference frames, and the macroblock has information on a prediction error (Δ N), which is a difference value associated with the retrieved area, and also has information on a relative position (motion vector) in space.

Bidirectional predictive macroblock

The bidirectional predictive macroblock is a macroblock encoded in bidirectional prediction by referring to past and future reference frames. An area most similar among the past reference frames is retrieved, and an area most similar among the future reference frames is retrieved, and the macroblock has information on a prediction error (Δ N + (P)/2), which is a difference value of the average values (per pixel) of the two areas, and also has information on a spatial relative position (two motion vectors) between them.

For the embedded information data, at least one macro block to which the embedding process is given must first be specified in the B picture. For example, may be defined as a separate macroblock (embedding area) existing between the first and third lines of the B picture, or may be defined as an entire macroblock of a certain frame. In addition to defining the macro blocks as a format in advance, it can be determined by using an algorithm that generates a positional order. Note that the algorithm for generating a positional order can use an algorithm disclosed in, for example, japanese patent application No. 8-159330. Subsequently, for a given embedded macroblock, one bit of data is embedded into one macroblock according to an embedding criterion. The embedding criterion is a predictive type embedding criterion in which bit information is made to correspond to one macro data block. For example, there are the following criteria.

(embedding criterion)

Inter prediction type of bit information macro block to be embedded

Bit "1" bi-directionally predicting a macroblock (denoted by B)

Bit "0" forward predicts the macroblock (denoted by P) or

Backward prediction macro block (denoted by N)

For example, consider a case where the bits 1010 of the management information are embedded. The four bits of data are embedded in four embedding areas of the first to fourth macro data blocks placed from the left side of the first row shown in fig. 7. First the first data bit is 1, so the prediction type of the leftmost macroblock (first embedding area) is determined to be bi-directional prediction (B) according to the aforementioned embedding criterion. The prediction error in this case becomes a prediction error that is a difference value with respect to an average value of an area most similar among the past reference frames and an area most similar among the future reference frames. The data bit is then a 0.

Thus, the prediction type of the second macroblock (second embedding area) is either a forward predicted macroblock (P) or a backward predicted macroblock (N) according to the embedding criterion. In this case, in order to suppress image degradation, a prediction error in forward prediction and a prediction error in backward prediction are compared to select a type in which the prediction error is small. In the example of fig. 3, the forward prediction (P) is selected for this second macroblock, since the prediction error in forward prediction is smaller than the prediction error in reverse prediction. A similar process is repeatedly applied for the third embedding region and the fourth embedding region.

As a result, since the prediction error in the backward prediction is small, the prediction type of the third macroblock becomes bidirectional prediction (B) and the prediction type of the fourth macroblock is determined as backward prediction (N). In this way, the inter prediction types of the first to fourth embedded areas are taken as BPBN, and four data bits 1010 (i.e., management information 1010) are embedded in these areas by data transformation without changing the data amount. An example of embedding management information in MPEG using a data hiding technique is described in japanese laid-open unexamined patent application 8-272721.

A method of extracting the management information embedded in the above-described process will now be described. In the case of extracting the management information, information for specifying one macro data block in which the management information has been embedded must first be given. The designation information may be given by an external unit. Further, it is possible to embed the specifying information in advance in the data itself. Furthermore, in the case where the position of the embedding region is standardized, or if an algorithm for generating a positional order is known, the information data can be extracted. A method of extracting information data using the positional order is disclosed in the above-mentioned japanese patent application No. 8-159330.

Then, from the prediction type of the specified embedded region, the information embedded in this region is extracted by referring to an extraction criterion. The extraction criterion is one in which the prediction type of the macro data block is made to correspond to bit information, and must be given with the information when extraction is performed. The following rules apply as the criteria. Note that the correspondence between the prediction type and the bit information in this extraction criterion is the same as that of the embedding criterion described above.

(criteria for extraction)

Bit information to be extracted for inter prediction type of macroblock

Inter prediction type bit "1" of macroblock

Inter prediction type of macroblock

The forward prediction macroblock (denoted by P) has bit "0

Or backward predicting a macro block

Description will now be made regarding a case in which management information as shown in fig. 7 has been embedded. As described previously, as a premise, the management information bits have been embedded in the embedding area from the first macroblock on the left of the first row shown in fig. 3 up to the fourth macroblock. Since the prediction type of the leftmost macroblock is bi-directional prediction (B), bit 1 is extracted by referring to the extraction criterion described above. The prediction type of the second macroblock is forward prediction (P), so bit 0 is extracted according to the extraction criterion. By repeatedly performing the same process on other macro data blocks, bit 1 and bit 0 are sequentially extracted. As a result, the management information bits 1010 are extracted from these areas.

According to the present embodiment, when a moving picture image is encoded, the prediction type of a macroblock and an embedding bit are determined so that they coincide with each other. Therefore, the management information can be embedded in the moving picture image without substantially affecting the compression efficiency of the moving picture and without substantially causing degradation in image quality. Further, it is difficult to remove the management information embedded in this manner from the moving image. Also, since much information to be embedded is almost independent of the content of the image, it is possible to efficiently embed information data.

As seen from the above description, the present invention is characterized in that: management information is embedded in data by using a data hiding technique that embeds data by data transformation without changing the amount of data, and a system for managing access to the data is controlled based on the management information when the data is accessed (reproduced or recorded). Also, the present invention is not limited to the above-described data hiding method, and it should be noted that the present invention is applicable to various methods.

Finally, a data access system shown in fig. 8 will be described. Although the above-described management information represents three states in two bits, the embodiment described therein is intended to provide three states for the system using essentially one bit of management information. To this end, the system of the present embodiment includes a controller 81 that detects what state data is supplied to the system, and controls the output of the system according to the state. The data supply unit 82 is for supplying data embedded with management information to the access system, and includes, for example, an optical system for reading out data from a DVD-ROM, and a circuit for capturing data on the internet into the system. Wherein the management information is represented by one bit of information, each bit of which indicates the following information.

(management information)

State management information content

Copy inhibition 1

Copy permission 0

The data supplied from the data supply unit 82 is processed by a signal processing system 83 which performs demodulation, decryption, D/a conversion, and the like. From the data obtained by the processing by the signal processing system 83, the management information extractor 84 specifies an embedding area in which the management information is embedded, and extracts the management information according to the state of an embedding area with reference to an extraction criterion in accordance with the content of the data for extracting the state of the embedding area. The management information extractor 84 outputs a control signal based on the extracted information. The control signal is used to control an output unit 85 to superimpose a jamming signal selectively generated by the jamming signal generator 86, depending on the content of the control signal on the output signal. That is, when the management information is bit 0 (copy permission), one switch in the output unit 85 is turned off to output the data. On the other hand, when the management information is bit 1 (copy prohibition), the switch is turned on to superimpose a disturbance signal on the data and output them. Since the operation is substantially the same as that of the constituent elements of the system shown in fig. 5, further detailed description will not be provided.

The controller 81 is connected to the data providing unit 82 and monitors the state of the data provided to the system. Subsequently, even when the management information is bit 0, that is, when copying of data is permitted, the controller 81 controls the output unit 85 so that the switch in the output unit 85 is turned on in accordance with the result of the supply state to output a superimposed signal. That is, the copying of the data is forcibly prohibited although the management information itself permits the copying of the data.

The specific data providing state is determined by monitoring, for example, "when data is captured through the recording medium in reproduction to an access system, whether the loaded recording medium is of a read-only type or a rewritable type". The type of recording medium can be easily determined because the type of recording medium in an unfilled area (lead-in area) of the recording medium can be identified by the hardware of the system. When copying of data is intended to be performed in a state where data is supplied from a read-only recording medium, the controller 81 will not force the switch to be turned off. Therefore, the access system operates after the management information is extracted, and the data can be copied as long as the management information allows such copying. On the other hand, when copying of data is intended to be performed in a state where data is supplied from one rewritable recording medium, the controller 81 turns on the switch regardless of the content of the management information. Therefore, data cannot be copied from the rewritable recording medium.

Generally, when a famous or important content is prevented or intervened from being illegally copied, a provider of the content stores and distributes the content in a ROM. The provisioning state in the ROM state means that the data can be considered as the original state rather than the copied data. On the other hand, when data is provided as one RAM or R data, the data may be considered as copied data that has been copied in the RAM or R in most cases. Therefore, as long as data stored in one ROM is regarded as original data and data in the RAM or R is regarded as copied data, further copying, i.e., illegal copying, of the copied data can be effectively prevented by providing such a function in the system. As a result, even with one 1-bit management information that can only indicate two states of "copy enable or disable", the system can determine whether or not data is original data based on its supply state, so that a third state of "copy once" can be substantially supplied for the system.

Since the management information to be embedded can be represented by one bit instead of two bits, it is not necessary to consider the order of arrangement of the bits but the size of the management information. When the management information is two bits, if the information contents for the bit arrangements of 01 and 10 are different, it is necessary to consider their order in the extraction of the information. However, if it is a bit, this order consideration is not necessary. Therefore, an important improvement is made in the accuracy in the extraction of information.

In addition to the above-described cases, the provision state to be monitored may include a case of "when data is captured by communication or broadcasting, whether an access system is connected to a network in replication" or not. In this case, if data is connected to the network in replication, the data is considered as original data. However, if the data is not connected to the network in the replication, the data to be replicated is considered to be data further replicated from the replicated data stored in a location different from the network (for example, a read only memory in its own system). Therefore, for the replication through a network, it is sufficient to allow the replication under the condition that data is connected to the network. Also, various schemes may be considered including monitoring the time at which data provision is made.

As described above, the provider embeds an access condition for a receiver that receives content to be provided in the content itself as management information, and distributes the content to the general public. An access system that receives the content includes a function of extracting management information embedded in the content and controlling access to the content based on an access condition defined by the management information. Therefore, the receiving system can effectively prohibit access to the content by the receiver, and particularly prohibit illegal copying of the content by the receiver.

Claims (10)

1. A method for managing access to data by using management information embedded in said data, wherein said management information comprises a plurality of bits of information embedded in said data, comprising the steps of:

defining sets of data regions in the data for the plurality of bits of information to be embedded, wherein each data region in each set has at least one pixel and the data regions of each set have a high characteristic correlation therebetween;

comparing characteristic values between the data regions of each set and embedding one of the plurality of bit information in each set consisting of data regions by representing one bit information as a function of the characteristic value comparison and one bit value, and extracting a plurality of bit information of the management information embedded in the data by comparing the characteristic values between the data regions in the defined plurality of sets;

access to said data is managed based on said extracted management information.

2. The method of claim 1, wherein said step for managing the access comprises: a step of changing the management of access to said data according to the type of medium used to store said data.

3. The method of claim 2, wherein said media type is ROM, RAM or R.

4. A method as claimed in any one of claims 1 to 3, wherein said management information includes information for controlling storage of said data, reception of said data, playback of said data, disturbance of output of said data, reproduction history of said data, or modification of said management information.

5. The method of claim 4, wherein said step for managing the access comprises the steps of: after managing access to said data based on said management information, additional information for managing access to said data is embedded in said data independently of said management information, and said management information and additional information manage subsequent access to said data.

6. A system for managing access to data by using management information embedded in said data, wherein said management information comprises a plurality of bits of information embedded in said data, comprising:

defining sets of data regions in the data for the plurality of bits of information to be embedded, wherein each data region in each set has at least one pixel and the data regions of each set have a high characteristic correlation therebetween;

comparing characteristic values between the data regions of each set and embedding one of the plurality of bit information in each set consisting of data regions by representing one bit information as a function of the characteristic value comparison and one bit value, and extracting a plurality of bit information of the management information embedded in the data by comparing the characteristic values between the data regions in the defined plurality of sets;

means for managing access to said data based on said extracted management information.

7. The system of claim 6, wherein said means for managing access comprises a means for changing management of access to said data based on a type of medium used to store said data.

8. The system of claim 7, wherein said medium is of the type ROM, RAM or R.

9. The system of any of claims 6-8, wherein said management information includes information for controlling storage of said data, reception of said data, playback of said data, disturbance of output of said data, duplication history of said data, or modification of said management information.

10. The system of claim 9, wherein said means for managing said access comprises: a unit for embedding additional information for managing access to said data in said data independently of said management information after managing access to said data based on said management information, and

a unit for managing subsequent access to said data based on said management information and said additional information.