Adaptive and progressive system and method for the secure distribution of wavelet-coded still images

Translation of description of equivalent WO 2004068858 A2

METHOD AND SYSTEM FOR ADAPTIVE AND PROGRESSIVE SECURE DISTRIBUTION OF IMAGES ENCODED WAVELET

The present invention relates to the field of digital image processing coded wavelet.

It is proposed in the present invention to provide a system to confuse visually and restore progressively and adaptively original content of a digital still image encoded in wavelets. The general problem is to provide a method capable of securely transmitting digital data for high quality images in any digital format from a wavelet encoding, live or deferred to a display screen and / or to be saved to the hard drive or other storage device belonging to a housing connecting the remote transmission network to a monitor type screen or TV screen, while preserving visual quality but avoiding any fraudulent use such as the possibility of making pirated copies of the images digitally encoded. Conventional encryption techniques consist generally to combine (according to type of addition or subtraction operations) with the original data of pseudo-randomly generated values ​​from an initialization key. The mere possession of this key allows the complete decryption of encrypted data, the latter containing substantially all of the original information.

In the European Patent Application EP 1011269A1 referenced entitled "System for processing an information signal," described an encryption method of an information signal that can be applied to the case of still images. The method is to add the original signal not compressed pseudo-random noise to obtain a new signal. The signal is then encrypted and compressed using suitable standard algorithms and transmitted. The key is in turn transmitted securely to the future user of the encrypted signal. This known method can be applied to the case of images coded in the JPEG standard.

It is made in this document of the prior art no reference to the case of images encoded by wavelets. Furthermore, possession of the key determines completely the decryption of the transmitted signal.

In "An integrated approach to encrypting scalable video" Eskicioglu et al., Proceedings of the 2002 IEEE International Conference on Multimedia and Expo, Lausanne, Switzerland, the authors describe a method for generating, managing and updating of key of encryption used to protect a bitstream multi broadcast encoding a video sequence and presenting scalability properties (multi-layer). The protection of the described workflow system is a system in which each layer is encrypted using a different secret key, this key being known by a group of users and can change periodically over time and depending on the number of users . The proposed system is therefore based on selective conventional encryption technologies using one or more keys all data are present in the protected stream and they condition only access to content, therefore this prior art does not solve the high security problem, the object of the present invention.

In the article "Protecting the VoD EasierWay" Griwodz et al. , Proceedings of the ACM Multimedia September 1998, describe a method for distributing, via broadband networks or temporary servers and a secure point to point connection, multimedia content protected whose access is controlled and traced. The initial stream coding the original audiovisual content is deliberately corrupted by a predetermined modification of certain bytes in the stream and a signal for its reconstruction is subsequently transmitted to the customer at the time of viewing content: a key is to initially communicated to the client that will enable it to recalculate the location of the corrupted bytes in the stream. Then a signal containing the original byte is sent after encryption to reconstruct the initial stream. The reconstruction of the stream is thus conditioned by a key and therefore the method described in this document of the prior art do not provide the high level of security offered in the present invention.

The present invention relates more particularly to a device capable of securely transmitting an entire high visual quality of digital still images to a display and / or to be stored in the memory storage device of a box connecting the network remote transmission to the display, while maintaining visual quality but avoiding the possibility that images can be copied illegally.

The invention relates to a method for the secure distribution of still digital images in a nominal format derived from the coding wavelet, represented by a bit stream comprised of at least one packet (on the organization of the binary sequence) containing the least one block consisting of simple elements (eg coefficients) digitally encoded according to a method specified in within the stream concerned and used by all decoders capable of reproducing or decoding the order to display it properly. This method comprises:

• a preparatory step of changing at least one of the simple elements,

• a step of transmission of a main stream in conformity with nominal format, consisting of blocks and packets modified during the preparatory stage and - through a separate channel of said main stream of additional digital information to reconstruct the original stream from the calculation on the destination equipment, according to said main stream and the complementary information. complementary information is defined as a set consisting of data (examples of coefficients describing the original digital stream or extracts of the original stream) and functions (eg, substitution or permutation function). A function is defined as containing at least one instruction by relating data and operators. The complementary information describes the operations required to restore the original flow from said modified stream.

In the present invention, it defines the concept of "flow" as a structured binary sequence consisting of single element and ordained representative in encoded form data and meet a standard or a given broadcast standard.

The act of removing a portion of the original data of the original stream during the generation of the modified main stream does not permit restitution of said stream of origin from only the data of the modified main stream. The modified main stream is then called "flow secured ". The "secure distribution" is a distribution of secure flow.

In the present invention, is meant by the term "scrambling" editing a digital bitstream using appropriate methods so that the flow remains consistent with the standard by which it was generated, while making decodable and displayable on 1 screen visual information encoded by this stream, but altered the point of view of human visual perception.

In the present invention, refers to as "scrambling" the restitution process by appropriate methods of the original bitstream, the bitstream returned after descrambling being identical, so no loss, the original bitstream.

We define the concept of "granular scalability" from the English expression "granular scalability". Is defined as the scalability property which characterizes an encoder capable of encoding or a decoder capable of decoding an ordered set of bit streams so as to produce or reconstitute a so-called multi-layer sequence. We define the granularity as the amount of information that can be transmitted through each layer of a stream resulting from encoding in layers, the flow is then also called "granular". We define the qualitative scalabilities, and spatial resolution. A stream has a "qualitative scalability" if it is organized in an ordered structure of successive sub-layers which enables the addition improving the visual quality of the image.

A stream has a "spatial scalability" if it is organized in an ordered structure of data blocks encoding a locally spatial information in the image. A stream has a "scalability resolution" if it is organized in an ordered structure of data blocks coding information to decode the image at a resolution level set.

We define scalability in resolution as the ability to decode the image in several resolution levels from a single bit stream representing the image encoded by wavelets. A stream has a "spectral scalability" if it is organized in an ordered structure of data blocks coding information to decode an image multi components according to a fixed component.

The present invention provides protection encoded digital image wavelet based entirely on the "bit stream (bitstream)" structure (binary sequence), protection of modifying targeted parts of the bit stream (for modeling wavelet) and its characteristics. Some true values ​​are extracted from the bit stream and stored as supplementary information, and, in their place, are being random or calculated values ​​or swapped values, and for the entire digital stream. Thus, the scrambler adds "decoys" for the decoder which receives as input a bitstream fully conforms to original digital format, but from which the decoded and displayed image is not acceptable from the point of view of human visual perception. The module performs a scrambling bit stream analysis and selects the parts of the bit stream as it introduces disturbances. A disturbance is defined as a change (eg change of value, sign change, saturation, threshold), or substitution by a random or calculated value, or a permutation. The method for scrambling - descrambling is achieved without loss of quality to the original image. Advantageously, the operation of scrambling is also performed with a decoding - encoding part of the bit stream (bitstream) representing the encoded image.

Advantageously, the present invention enables the protection of an encoded digital image into a stream having a spatial scalability property.

Advantageously, the present invention enables the protection of an encoded digital image into a stream having a resolution scalability property.

Advantageously, the present invention enables the protection of an encoded digital image into a stream having a scalability property quality. Advantageously, the present invention enables the protection of an encoded digital image into a stream having a spectrum scalability property.

Advantageously, the operation of scrambling is also using full prior decoding of the bitstream representing the encoded image and then re-encoding before changing into a stream having scalability properties.

Unlike most encryption already known by those skilled in the art systems, the principle described below makes it possible to ensure a high level of protection while reducing the amount of information necessary for the decoding performed progressive and adaptive manner.

Protection, designed so the invention is based on the principle of removal and / or replacement of information coding the original visual signal by any method, including: substitution, modification, permutation or travel information. This protection is also based on the knowledge of the structure of the flow in the output of the video encoder: 1 scrambling depends on the structure of said digital stream. The reconstitution of the original stream is performed on the destination equipment from the modified main stream already present or available (eg on a CD or DVD) or sent in real time on the destination equipment and supplementary information sent real-time view time including data and functions executed using routines (set of instructions). Knowing how are conducted modeling, compression and encoding wavelet image by wavelet encoder and / or the standard or the given standard, it is still possible to extract from the bit stream ( bitstream) the main parameters that describe and which are sent to the decoder.

Many systems scrambling have an immediate effect, that is to say what the initial stream is totally scrambled or the initial stream is not scrambled at all, and even for descrambling systems of visual content . With rigid systems of this type, it is difficult to meet the management of user rights and quality of service multi-user client-server systems, multi-application and multi-services that is to say, adapting services according to the different profiles of users and their rights.

A "profile" of the user, a digital file comprising descriptors and information specific to the user, eg its cultural preferences and social and cultural characteristics, usage patterns such as the frequency of use display means, the average length of viewing still images scrambled and / or scrambled Taklimakan Desert, viewing frequency of scrambled sequence and / or unscrambled, or other behavioral characteristic regarding the operation of still images and still images estates. This profile is formalized by a digital file or a digital table by computer and is in the server and / or client STB.

A "hardware profile" of the user, a digital file comprising descriptors and information specific to the user's display hardware, for example the resolution of its display screen, the computing power of the still image decoder or any other physical characteristic in relation to the exploitation of still images or succession of still images. This profile is formalized by a digital file or a digital table by computer and is in the server and / or client STB.

The present invention intends to overcome the disadvantages of the prior art by providing an adaptive and progressive descrambling system of the content displayed according to the profile and user rights.

In the present invention, apply an adaptive and progressive descrambling content displayed according to the profile and the rights of each user. The server sends only the parts of the complementary information, which has a structure characterized by a "granular scalability" to provide the user with a content more or less confusing depending on certain criteria, profiles and rights. Digital streams encoded wavelet possess granular scalabilities spatial properties, quality and resolution.

The granularity of the complementary information is relative to the degree of scrambling. For example, the still images are completely confused, once for all users. Then, the server sends all or part of said additional information so that the image or succession of still images to appear more or less confusing to the user. Content submitted the complementary information and content displayed on the user's display screen are based on each client and server manages and sending real-time view time per user.

In its most general sense, the present invention relates to a method for the secure distribution of still digital images via stream comprising data sequences, each containing a portion of the picture information, the method comprising a step of changing the original stream by modifying at least a portion of said data sequences, modification producing a modified stream at the same nominal size as the original stream, the method comprising a step of transmitting the modified stream and a reconstruction stage using a decoder on the addressee equipment, characterized in that the reconstruction is adaptive and progressive as a function of information from a digital profile of the addressee user.

Advantageously, said modification produces a modified main stream and complementary information enabling the reconstruction of the original stream by a decoder, the method comprising a step of transmitting the modified stream, and further comprising a step of transmitting to the addressee equipment a sub-set of said complementary modification information, said subset being determined based on information from a recipient's digital profile. Advantageously, said modification produces a modified main stream and complementary information enabling the reconstruction of the original stream by a decoder, the method comprising a step of transmitting the modified stream, and further comprising a step of transmitting to the addressee equipment a sub-set of said complementary modification information, said subset being determined based on information from a recipient of the hardware profile. In addition, said original stream is encoded according to a wavelet encoding method.

Advantageously, the original stream has a resolution scalability property. Advantageously, the original stream has a spatial scalability property.

Advantageously, the original stream has a qualitative scalability property.

Advantageously, the original stream has a spectral scalability property.

Alternatively, the modified main stream is available on the target equipment prior to the transmission of additional information on the target equipment.

Alternatively, a portion of the modified main stream is available on the target equipment prior to the transmission of additional information on the target equipment. In another variant, the modified main stream and complementary information are transmitted together in real time. Advantageously, the determination of said subset of said complementary information is based on the scalability properties of said original stream.

Advantageously, the determination of said subset of said complementary information is based on the granular scalability of said complementary information properties.

In addition, the amount of information contained in the said subset corresponds to a scalability level determined based on the recipient's profile.

Advantageously, the type of information contained in said subset corresponds to a level of scalability determined based on the recipient's profile.

Advantageously, said additional information comprises at least one digital routine capable of executing a function.

Advantageously, said functions transmitted to each addressee are personalized for each addressee as a function of the session. Alternatively, the complementary information is encrypted for each addressee beforehand as a function of the session.

According to a variant, said additional information is subdivided into at least two sub-parts. According to one embodiment, these subparts of the complementary information are distributed by various media.

According to another embodiment, said sub-parts of complementary information are distributed by the same media.

Alternatively, all or part of the complementary information is transmitted on a physical vector.

In another embodiment, the complementary information is transmitted on-line. Advantageously, the type of information contained in said subset is updated based on the behavior of the recipient while connecting to the server, or according to their habits or according to data provided by third parties.

Advantageously, the amount of information contained in said subset is updated based on the behavior of the recipient while connecting to the server, or according to their habits or according to data provided by third parties.

In addition, the method includes a prior step of analogue / digital conversion in a structured format, the method being applied to an analog signal.

Alternatively, a preliminary step transcode the digital stream from any format to a format having scalability properties.

Advantageously, said still images constitute a succession of still images over time.

According to one embodiment, said modification of said data sequences is different for at least two images of said succession of images.

In another implementation mode, said modification of said data sequences of an image in said succession of images includes the modification of said data sequences of the preceding images in the temporal order of the succession based on the properties of spatial and qualitative scalability wavelet transforms.

Advantageously, the granular scalability of said complementary information made up of said subsets is based on scalabilities qualitative, spatial and resolution of streams from a wavelet transformation of the images.

Moreover, the process is lossless. According to a particular embodiment, the invention is also applicable to the treatment of fixed images, such images compressed using the JPEG2000 standard. In this case, it concerns a method for e secure distribution of digital still images providing that when reconstructing said original stream an indelible and imperceptible trace is inserted into original stream, this trace with an unambiguous identifier. Alternatively, an indelible and imperceptible trace is inserted into the image after reconstruction and decoding of said original stream, this trace with an unambiguous identifier.

According to an exemplary embodiment, the indelible and imperceptible trace is detectable by appropriate software analyzing the reconstructed content.

Preferably, said unambiguous identifier authenticates the user.

Alternatively, said unambiguous identifier authenticates the device on which the original stream reconstruction algorithm was executed.

In another variant, said unambiguous identifier identifies the session opened by the user in which the reconstitution of the original stream is executed. Advantageously, the session scrambling and descrambling session are carried out under the control of a secure server acting as a trusted third party.

According to a particular implementation mode, the session is identified by a secure server, available into a register for each session information on the session number, the identifier of the user or the identifier of the user equipment, the identifier of the content object of the session and a date - time group. According to another mode of implementation, a digital signature is calculated from the reconstructed stream, the inserted track generates a unique and different signature for each reconstructed stream and this signature is stored on a secure server acting as a trusted third party.

Preferably, the stream restored by descrambling has the same visual quality as the original stream and are readable form only if it has said trace.

Advantageously, the flow is restored by descrambling readable form only if the digital signature extracted during an authenticity check step is identical with the signature stored on the secure server acting as a trusted third party.

Advantageously, the invention is applied to a digital broadcasting stream from a standard or standard owner.

The invention also relates to a system for the secure distribution of still digital images comprising a server comprising means for broadcasting a modified stream, and a plurality of equipment fitted with a descrambling circuit, the server further comprising means recording the digital profile of each recipient and profile analysis of an average of each of the recipients of a modified stream, said means controlling the nature of the complementary information transmitted to each of the recipients.

The invention further relates to a system for the secure distribution of digital still images the level

(Quality, quantity, type) the additional information is determined for each recipient based on the state of its profile when viewing the main stream. The invention will be better understood using the description given below for purely explanatory purposes, an embodiment of the invention, with reference to the attached figure: Figure illustrates a particular embodiment of the client-server system according to one invention.

The digital images are obtained using compression technologies based on wavelets

(Eg the still images in the JPEG-2000 standard, MPEG

4, JJ2000, JASPER, Kakadu, Moving JPEG-2000), the concept of wavelets is an iterative scheme that is to say, the repetition of the same filtering operation at resolutions increasingly weak, and generates flows characterized by a spatial scalability, quality and resolution. The original stream is reconstructed on the destination equipment from the modified main stream and the complementary information. The additional information is divided into subsets, and depending on the user profile, a subset, several subsets or all of the additional information are sent to the partial or total unscrambling images.

Advantageously, the additional information sent to the user is encrypted prior to its transmission by using a specific key for each user.

Is defined as quantity of information contained in said subset of the number of data and / or functions belonging to the complementary information sent to the recipient during the connection.

The type of information contained in said subset corresponds to a level of spatial scalability, quality and resolution determined by the recipient's profile. Is defined as "type" of nature data and / or functions belonging to the complementary information sent to the recipient for the server connection. For example, the data type is related to the recipient's habits (full subscription, partial subscription, pay per view, connection time, connection duration, regularity of the connection and payments), environment (lives in a large city, the weather at this time) and characteristics (age, sex, religion, community). Advantageously, additional information consists of a succession of sub-sets each corresponding to a scalability level set in the original stream.

The complementary information is composed of at least one function, and functions are customized for each recipient from the login session. We define a session from the login time, duration, type of said first flow visualized and connected elements (recipients, servers). The complementary information is subdivided into at least two sub-parts, each sub-parts that can be distributed by various media, or by the same media. For example, in the case of distribution of complementary information by several media, we can ensure a more complex management of rights of recipients.

Several embodiments are described below.

The wavelet transform of an image (two-dimensional spatial signal) consists in applying to the original image a succession of high-pass filters and low pass made from the characteristics of wavelet analysis. The synthesis process which comprises reconstructing the image from the set or sub all wavelet coefficients generated by the transform, obeys an inverse filtering scheme.

Applying a step wavelet transform on a digital image (which may be composed of only or several matrices of real or integer values) is equivalent to a filtering operation on the rows and columns or matrices of values followed by a dyadic reduction (halving) of the size. Therefore it generates each step 4 new matrices of wavelet coefficients, called subbands and whose width and height are equal to half the width and height of the transformed matrix (dyadic progression). Either an I picture of width L and height H and A. The application of a resolution step of the wavelet transform therefore generates 4 matrices of dimension of wavelet coefficients (L / 2, H / 2): the sub LL band_R-_IResult of a horizontal low pass filtering (lines) and vertical (columns) of the image I, the sub-band LH_R_ι result of a low-pass filtering horizontal and vertical high-pass, band-sous HL_R_ι result of a horizontal high-pass filter and low-pass vertical and sub-band HH_R_ι result of a high-pass filtering horizontally and vertically.

Considering the wavelet transform to R levels (equivalent to R stages) of an image. A wavelet transform to R levels is associated with R + the resolution levels, numbered 0 to R, R and 0 respectively corresponding to the finer levels of resolution (original image) and coarser (approximate image). Each end of the decomposition sub-band wavelet image I is identified by its orientation (LL or HL or LH or HH) and its corresponding level of resolution (between 0 and Rl).

The original image can be seen as the LL band_R. At each level i of the wavelet decomposition (Except the last i = 0), the sub-band LLi thus divided into 4 new sub-bands LLi_χ, HLi-i, LHι_ι and HHI-i and whose size is halved compared to LLi. The process is iterated until the sub-band OLA is obtained. Thus, for a wavelet transform levels R, 3R + 1 wavelet coefficients of sub-bands are generated: LL₀, HLO, LH₀HH₀, HLi, LHi, HHI, ..., HL_R-ι, LH_R_ι, HH_R_ι.

The image reconstruction (synthesis) from 3R + 1 subband coefficients comprises applying an inverse filtering operation on these wavelet coefficients followed by a dyadic increase in size. Progressive reconstruction of the image at different resolution levels can thus be made. For example, adding in the synthesis operation in the reconstructed image of rl resolution 3 sub-bands HL wavelet coefficients_r-ι, LH_r-_!HH_r-i r a new resolution image is obtained.

The only sub-band wavelet coefficients OLA is an approximation of the original image LL_R whose resolution is 2^R times lower than the original image.

3R sub-bands of wavelet coefficients HL_r_ι, LH_r-ι, HH_r_ι (re [l, R)] is for their to details in the picture, taken at rl resolution. R is more, the greater the wavelet coefficients of these subbands are details of features of finer (smaller) in the original image.

The wavelet coefficients derived from the wavelet transform of an image are spatially local characteristics of a frequency information. More r decreases, the spatial area characterized by a single wavelet coefficient increases (multiplication by a factor of 4 in each stage). A wavelet transform to R levels of an image produces a "picture" called approximation, resolution 2^R lower and 3R "images" saying details at different resolutions (0 to R). Consequently, a bitstream with a granular scalability can be represented as follows: {Bo, Bi, ..., Butot). Each Bi represents a set of bits sub, the bit stream can then be described as a series of subsets Bi binary symbols. Thus, a bit stream from a coding wavelet has the property of "qualitative granular scalability" if and only if:

- Decoding of n (n <N tot where the bit stream is described as a sequence of Ntot subsets Bi) sub-sets of bits B₀, Bi, ..., B_not involves a decoded image of the quality Q_not, Quality being measured relative to the original image I according to a predefined metric M calculated from subjective and / or objective, that is to say Q "= H (I_d(N), I). - When m (m <n) subsets are decoded Bi, B {₀, Bi, ..., B_ra}, then Q_m <Q_not.

- Where p (p> n, p <N) Bi subsets are decoded {Bo, Bi, ..., B_m, ..., B_not, ..., B_p}, then Q_not <Q_P.

- When Ntot subsets Bi are decoded, Qutot is maximum and Q_NOTtot - Qi for 0 <i ≤ Ntot.

Similarly, a bit stream from a wavelet encoding has the property of "scalability resolution" if and only if:

- Decoding of n (n <N tot) bit subsets {Bo, B_x, ..., B_not} Implies an image I_d resolution R_not.

- When m (m <n) subsets Bi are decoded {Bo, Bi, ..., B_m}, Then R_m <R_not.

- Where p (p> n, p <Ntot) subsets Bi are decoded {Bo, Bi, ..., B_m, ..., B_not,., "B_p}, Then R_not <Rp. - When Ntot subsets Bi are decoded, R_ntot is maximum and Rutot ≥ R for 0 <i ≤ Ntot.

For example, an image was blurred by changing

(Change type of add / noise substitution, thresholding, permutation) a subset Cash N wavelet coefficients for one or more spectral components of the image and / or belong to one or more regions of interest in the original image or to the entire image and / or relative to different levels of resolution in the wavelet decomposition (0 to R) and / or belonging to one or more subbands (LL among,

HL, LH and HH).

The adaptive and progressive descrambling the image is to gradually unscramble the image into several steps: first we replace No (0 <no <Ntot) wavelet coefficients modified their original values, then neither (0 <or <Ntot) and so on up to n_P (0 <n_P <Ntot) such that: no + ni + ... + n_P == Ntot. Depending on the method of scrambling used and the customer profile, the descrambler is adapted to the behavior of the client when connecting to the server.

Describe an example of progressive descrambling. In this example, the Ntot modified wavelet coefficients belong to the sub-bands HL, LH and HH corresponding to 4 different levels of resolution (ie r, r + 1, r 2, r 3). It is to first replace the number wavelet coefficients belonging to subbands HL_rLH_r and HH_rAnd then neither the wavelet coefficients belonging to subbands HL_{r +}ι, LH_{r +}ι and HH_{r + 1}And the n₂ wavelet coefficients belonging to subbands HL_{r +}2, LH_{r +}2 and HH_r2 and finally n₃ coefficients belonging to subbands HL_{r + 3}LH_{r + 3} and HH_{r + 3}. The first step of descrambling (replacement the number coefficients) mitigates resolution and extended the effects of the initial scrambling (removal of one scrambling of resolution r details) but the details belonging to higher resolution levels (r 1, r 2 and r + 3) are always degraded. The following steps will mitigate increasing the scrambling eventually reaching full descrambling. This example is merely illustrative and should not be considered limiting. The number of resolution levels assigned by one initial scrambling may be between 1 and R + 1. Based on this number, the maximum number of steps of descrambling may be between 1 and R + 1.

Another variation is to send n coefficients belonging to subbands HL_{r +}i, LH_{r +}i and HH_{r +}i into several sub steps, thus increasing the number of stages of progressive descrambling.

Another way of progressive descrambling consists in restoring the original wavelet coefficients for a C spectral components of the image, then two of the components C and so on until the complete recovery of original wavelet coefficients for the C components. We can then talk about progressive spectral descrambling. Depending on the number of components initially scrambled (1 to C), the number of steps of descrambling also varies between 1 and C.

According to an alternative embodiment, the progressive descrambling of a scrambled picture is to restore the original wavelet coefficients belonging to a predefined spatial area in the image while maintaining a total scrambling the rest of the image. Or (L, H) the size of the original image and (w, h) the size of the area of ​​interest that it is desired descrambling on the original image. In this example, we assume that this area is located in the center of the image, but this area can be defined anywhere in the original image.

R is the resolution level of the sub-band of wavelet coefficients that must be restored. Then the following formulas indicate the intervals [e, s] and [ie, I] indices of the wavelet coefficients to restore the resolution r subband considered:

is = nlr / 2 - (1 / js =^'ncr / 2 - (h / 2^{r + 1}) 2^{r + 1}), Ie = nlr / 2 + (1/2^{r + 1}) I = ncr / 2 + (h / 2^{r + 1})

where (NLR ncr) are respectively the number of rows and columns of the matrix of wavelet coefficients in the sub-band considered. The progressive descrambling of the area of ​​interest is carried out successively by restoring the original wavelet coefficients of each sub-band for each resolution: for example, OLA and HLi, LHi, HHI and HL₂LH₂HH₂, And so on until HL_R-ι, LH_R-ι, HH_R-ι.

According to another embodiment, the progressive descrambling consists in restoring the first original wavelet coefficients belonging to the LL sub-band, then the original wavelet coefficients belonging to all the sub-bands LH, then the original wavelet coefficients belonging to all sub -bandes HL and finally the original wavelet coefficients belonging to all the subbands HH. Thus, one scrambling details gradually attenuated according to their guidelines. Advantageously, the order of the types of sub-bands for which wavelet coefficients are output can be changed.

The invention will be better understood when reading an embodiment concerning a stream in JPEG-2000 format In this example, the invention is to modify the value of certain fields, including the information necessary for a decoder for reconstruction the original stream.

On the attached drawing, the figure represents a particularly preferred embodiment of the client-server system according to the invention.

The original stream (11) can be directly in digital form (111) or analog form (101). In the latter case, the analog stream (11) is converted by a non-illustrated encoder in a digital stream (111). In what follows, we denote (1) the digital input stream corresponding to the still image. The JPEG-2000 stream that is to be secure (1) is sent to a system of analysis and scrambling (121) which generates a modified main stream (122) to the same JPEG-2000 format, a format identical to flow inlet (1) outside of that the values ​​of certain elements in the feed have been replaced by different values ​​of the original ones, and is placed in an output buffer. The additional information (123) of any format and organized in granular scalability layers, contains information related to the images that have been modified, replaced, substituted or moved, and their value or location in the original stream and has sub-units relating to its granular scalability property.

The flow-in JPEG 2000 format (122) is transmitted via a telecommunications network (4) Hertzian, cable, satellite, etc., the terminal (8) of the user, and more precisely in memory or on its hard disk (85). When the user wishes to display still images in the terminal, the terminal (8) requested to display still images in memory or on its hard disk (85). The server (12) verifies the rights of this user for this request. For this, the server can use this user data contained in a server connected to the database (12) and / or use a card-based system chip (82) connected to the synthesis system (87). Two eventualities are possible.

If the user does not have all necessary rights to the image, in this case, the stream JPEG-2000 (122) generated by the system scrambling (121) in the memory or on the hard disk (85 ) is sent to the synthesis system (87) via a read buffer (83). The synthesis system (87) does not change and transmits the same as a conventional JPEG-2000 reader (81) and its contents, visually degraded by the system scrambling (121) is displayed on the screen display (6). The user terminal (8) thus sees a blurred image.

Alternatively, the user has rights to view the image. In this case, the synthesis system address to the server a viewing request (12) containing the necessary information (123) to recover the original image (101). The server (12) sends via telecommunications such analog or digital telephone networks, DSL (Digital Subscriber Line) or BLR (local radio loop) via DAB (Digital Audio Broadcasting) or via digital mobile telecommunications networks (GSM, GPRS, UMTS) (5) at least a subset of the complementary information (123) for reconstructing the image to the terminal (8) which stores said subassembly in a buffer (86) . The system Synthesis (87) then proceeds to the restoration, in the flow

JPEG-2000 tangled he reads in the read buffer (83), modified fields which he knows the positions and the original values ​​with the content of the additional information read in the buffer

(86) to descramble the image. The amount of information contained in the additional information

(123) and is sent to the descrambling system is specific, adaptive and progressive for each user and depends on its rights, for example single or multiple use, the right to one or more private copies, delay or advance payment. To determine the amount of information of the additional information

(123) to send to the terminal (8), the server (12) first consult the user rights.

According to one embodiment, a progressive descrambling an image for which the Ntot modified wavelet coefficients belong to the sub-bands HL, LH and HH corresponding to 4 different levels of resolution (that is to say, r, r + 1 , r 2, r 3) is to first replace the number wavelet coefficients belonging to subbands HL_rLH_r and HH_rAnd then neither the wavelet coefficients belonging to subbands HL_{r +}ι, LH_{r +}ι and HH_{r +}ι, then n₂ wavelet coefficients belonging to subbands HL_{r 2}LH_{r 2} and HH_{r 2} and finally n₃ coefficients belonging to subbands HL_{r + 3}LH_{r + 3} and HH_{r + 3}. The first step of descrambling (replacement No. coefficients) mitigates resolution and extended the effects of the initial scrambling (removal of scrambling resolution r details) but the details belonging to higher resolution levels (r + 1 , r and r + 2 + 3) are always degraded. The following steps mitigate increasingly 1 scrambling eventually reaching full descrambling. Depending on the number of levels of resolution R chosen to confuse the flow, the number of resolution levels assigned by the original scrambling is 1 and R + 1. Based on this number, the number of steps of descrambling is therefore between 1 and R + LIn this embodiment, the transmission of a subset of the complementary information containing the n₀ coefficients is performed when the user logs on, selects and downloads the image he wants. The selected image is displayed on the screen partially unscrambled, because it is calculated only from the No. coefficients transmitted to the user terminal. If the user decides to view the image in a higher resolution, the server offers the user to pay a predetermined amount. If the user pays immediately by conventional means remote payment (credit card ...), the server sends a second subset of the complementary information containing n_x coefficients. During the payment transaction, are sent subsets of additional information concerning the steps of unscrambling and mitigate increasingly 1 scrambling eventually reaching full descrambling, the displayed image is identical to the original image . If the client does not agree to pay immediately, the coefficients will be gradually sent depending on the arrival of the payment. For each transaction, the server records the behavior of the user and updates the profile thereof in a database based on said behavior.

Content submitted the complementary information (123) and the content displayed on the customer display screen are based on each client and server manages and sending in real time- said sub-ense mbles when viewing for each user, eg based on the price the customer is willing to pay. Consider that we have still images stored on the server 10 with different resolutions of the R = to R = 10, R = 10 is the maximum resolution. If a customer is used to control medium resolution images, the subscription corresponds to the descrambling obtained with R = 5. If he wishes to obtain a higher resolution, so for example a descrambler for R = 7, he must change the type of payment or subscription. It can then, if desired, and with such a new payment, obtain resolution R = 8 and R = R = 9 and finally 10. All these operations are managed by the server (12) depending on the behavior of each user by using a server connected to the database (12).

Similarly, another customer who needs high resolution images takes the subscription corresponding to the maximum resolution for R = 10. If such customer has a late payment, the server automatically sends images to the unscrambled by R = 6 example, to remind him to regularize his payment. As just described, the level (quality, quantity, type) the additional information is determined for each recipient, depending on the state of its profile at the time of transmission of the main flow and a party to less said profile is stored on a destination equipment. For example, the accompanying drawing, part of the profile is stored on the smart card (82) linked to the synthesis system (87), such as digital data on the frequency of connections or regularity payments. The same data and / or the rest of the profile may or may reside on the server (12). The rest of the profile can contain for example the type of images that the user prefers. In an alternative embodiment, the recipient's profile is updated. The update also depends on the server at connection time (data relating to the behavior), whether the client connects regularly (referring to his habits). Similarly, the recipient of the profile can be updated based on data retrieved from an existing consumer database on a server and on the client.

According to another embodiment, the server transmits all or part of the additional information to the user for a few seconds the display of the image, then, over time, transmits less subsets the additional information. So the descrambling of the picture is less complete, giving effect to the user that the image displayed on the screen becomes less and less understandable, therefore increasingly blurred. This feature prompts the user to purchase the rights to see the image completely unscrambled, since he saw part the content.

According to another embodiment, all or part of the complementary information (123) is transmitted to the user on a physical carrier such as a memory card or a smart card (82). According to another embodiment, only part of the modified main stream is available on the destination device: if the characteristics of the display screen (6) does allow you to view a limited number of resolutions (R = l R = 5), the user needs to recover a portion of the modified main stream (122). Depending on the profile and user rights, part of the additional information that can view the image only to resolutions R = l R = 5 will be sent by the server (12) to the user terminal (8).

The example described below shows another preferred embodiment of the progressive and adaptive for descrambling digital images from the JPEG-2000 standard.

The analysis module and scrambling (121) generates complementary information (123) and sends it to the destination device via the network (5). Advantageously, the network (5) comprises a secure server having stored additional information (123). The module (121) also generates the modified main stream (122) in the JPEG-2000 format which is transmitted to the hard disk (85) of the customer's destination equipment by the network (4). Advantageously, the network (4) includes a media server that stores the modified main stream (122). The module scrambling (121) also fits into the meta data of the modified main stream the identifier of the complementary information corresponding to said modified main stream and the physical address of the secure server (5) on which is stored the complementary information.

The additional information is characterized by the presence of said subsets corresponding to several layers of scalability, for example four in number. The first subset contains all of the additional information concerning the quality of the original image. The second subset contains a portion of the additional information relating to an acceptable quality. The third subset contains a portion of the additional information relating to a low quality, the picture remains still visible, but unusable. The fourth subset contains just the Some complementary information corresponding to a minimum quality, the picture is not visible.

Upon reconstitution of the original stream, the descrambling module (87) inserted in the stream reconstituted an indelible and imperceptible trace of the human eye, this trace with an unambiguous identifier. Track inserted in the stream is detectable by suitable software that has the ability to analyze the reconstituted content. The insertion of said trace and the descrambling are performed sequentially and progressively, so that a stream that has been scrambled by the analysis system and scrambling (121) and then descrambled by the module ( 8) are readable form only if he has said unambiguously trace. When descrambling and sequential insertion of the trace, the reconstruction is done so that the trace is gradually inserted during the progressive descrambling. At the end of descrambling, protection is provided by the trace which is substituted for the protection obtained by mbrouillage.

Advantageously, an indelible and imperceptible trace of the human eye is inserted into the image and after decoding the reconstructed stream.

Advantageously, a protected stream with the module for scrambling (12) and descrambled with this variant of the descrambling module (87) is always carrying a protection either invisible, after insertion of the identifying mark, is visible, after the adaptive and progressive descrambling. Advantageously, said unambiguous identifier carried by the trace relates to the identification of the interactive session opened by the user. As said identifier is unambiguous authentication of the user of the destination device on which the image is reconstructed and / or session opened by one user.

According to one embodiment, when the descrambling step, the flow is restored and an indelible and imperceptible trace is inserted. A digital signature of the reconstituted stream is then calculated, this signature is unique and different for each stream restored by the insertion of the track, and this signature is stored on a secure server acting as a trusted third party.

Advantageously, the flow is restored readable form only if the signature extracted in a step control of uthenticité is identical to the signature stored on the secure server during reconstitution.

The session was characterized by a number assigned by the secure server (5), which acts as a trusted third party between the user and the adaptation of parameters characterizing the type of descrambling. The secure server (5) assigns a specific number to each session, which is saved in a register. The information in the register is the session number, built from the user identifier or the identifier of its equipment (8), the identifier of the content of the image that is subject of the session and a group date - time in ISO standard.

The identification of the image content is performed by the descrambling module (87) that retrieves from the meta data of the modified main stream the identity of the additional information regarding the modified main stream and the physical address of the network (5) where is located the secure server where you stored the complementary information. The reconstruction of the original image is carried out in several stages. In the establishment of the session, the descrambling module (87) reads the metadata of the modified main stream, the identifier of the additional information and the URL of the secure server (5). The server (5) first sends said fourth subset of the additional information that produces minimal image quality, the image being unusable and poorly visible. This step serves as confirmation of the identity of the secure server (5). The second step is to send the descrambling module said third subset of complementary information that makes the visible image, but still unusable. This step is necessary for the client to decide if he wants to obtain rights to use the image, seeing a preview of its contents. Depending on the customer's wish (8) to obtain an acceptable quality or maximum image is sent to said second subset (for acceptable quality) or said first subset (corresponding to maximum quality) the complementary information on payment corresponding to the required quality. Regardless of the subset used for descrambling, unambiguous trace is always present in the reconstructed stream. Said trace is for the protection of intellectual property, in accordance with the WIPO treaty (World Intellectual Property Organization) in December 1996 stipulating that data for the protection of intellectual property may be digital data.

The embodiments described above have value examples and do not constitute a limitation of the present invention.