EP1614290A2

EP1614290A2 - Scrambling, descrambling and secure distribution of audio-visual sequences from video encoders based on wavelet processing

Info

Publication number: EP1614290A2
Application number: EP04742528A
Authority: EP
Inventors: Daniel Lecomte; Jérôme CAPOROSSI
Original assignee: Medialive SA
Current assignee: Medialive SA
Priority date: 2003-04-16
Filing date: 2004-04-16
Publication date: 2006-01-11
Also published as: US8433092B2; US20100142621A1; US7668312B2; US20060216003A1; FR2854019B1; FR2854019A1; WO2004095840A3; WO2004095840A2

Abstract

The invention relates to a method and system which are used for the visual scrambling of a video sequence and for the reconstruction (descrambling) of the original content thereof from a digital video stream obtained from encoding based on a wavelet transform. The method for the secure distribution of video sequences according to a digital stream format stems from wavelet-based coding, said format consisting of frames comprising blocks containing wavelet coefficients which describe visual elements. According to the invention, before transmission to the client equipment, the stream is analysed in order to generate (i) a modified main stream through the removal and replacement of certain pieces of information encoding the original stream and having the same format as said original stream, and (ii) a complementary piece of information with any format, comprising digital information which encode the original stream and which can be used to reconstruct the aforementioned modified frames. The modified main stream and the complementary information thus generated are then transmitted separately from the server to the destination equipment and, subsequently, a synthesis of a stream with the nominal format is calculated on the destination equipment as a function of the modified main stream and the complementary information. The invention also relates to a system comprising at least one multimedia server containing the original video sequences, a video stream analysis device, a device for separating the original video stream into a modified main stream and a complementary piece of information according to the aforementioned analysis and at least one device on the destination equipment for the reconstruction of the video stream as a function of the modified main stream and the complementary information.

Description

SCALDING, DE-SCALDING AND SECURE DISTRIBUTION OF AUDIO-VISUAL SEQUENCES FROM VIDEO ENCODERS BASED ON WAVELET PROCESSING

The present invention relates to the field of processing video sequences encoded using video coders based on wavelet technology.

It is proposed in the present invention to provide a method and a system for visually scrambling a video sequence and subsequently recomposing (descrambling) its original content from a digital video stream obtained by encoding based on a wavelet transform .

The present invention relates more particularly to a device capable of securely transmitting a set of video streams of high visual quality to a viewing screen such as a television screen and / or to be recorded on the hard disk or on any other medium. recording of a box connecting the remote transmission network to the display screen such as a television screen or a personal computer monitor, while preserving the audiovisual quality but avoiding any fraudulent use such as the possibility of making pirated copies of films or audiovisual programs recorded on the hard disk or any other recording medium of the set-top box. The invention also relates to a client-server system, the server providing the stream allowing viewing of the secure digital video film and the client reading and displaying the digital audiovisual stream.

With current solutions, it is possible to transmit films and audiovisual programs in digital form via broadcast networks of the type wireless, cable, satellite, etc. or via telecommunications networks such as DSL (Digital Subscriber Line) or BLR (radio local loop) or via DAB (Digital Audio Broadcasting) networks, etc. Furthermore, to avoid pirating of the works thus distributed, the latter are often encrypted or scrambled by various means well known to those skilled in the art.

Concerning the processing of video sequences encoded with wavelet technology, the prior art knows the patent referenced US 6370197 and entitled "Video Compression Sche e using wavelets" in which the authors detail a method of coding a video sequence based on a transformed into wavelets and generating a nested digital stream. This prior art does not propose any method for protecting the stream and / or scrambling the video sequence.

Also known is patent application EP0734164 presenting a method and a device for increasing the coding efficiency, carried out by video encoders based on classified vector quantization by optimizing the coding so as not to have to transmit the classification information. in the encoded bit stream. This prior art applies to video streams resulting from the transformation into DCT or wavelet. For this purpose, the incoming video signal is divided into a plurality of sub-bands, for example the DC coefficients are arranged in a sub-band and the AC coefficients in remaining sub-bands, followed by formatting in size blocks. identical, each block including a DC coefficient and a multitude of AC coefficients. A selection signal is then generated, representing the vector quantization class corresponding to each assembled block. This step is followed by classification for vector quantization by the generation of parameters relating to the evolution of the DC coefficients in the horizontal and vertical direction, and of the differential entropy encoding of the DC coefficients relating to the assembled blocks to generate a first encoded video signal. The AC coefficients are classified and encoded separately using entropy encoding based on the selection information to generate a second encoded video signal. The two signals thus generated are formatted for transmission. For decoding, no classification information is transmitted, it is reconstructed from the DC coefficients encoded and transmitted to the decoder. This solution relates to a process relating to digital compression and the encoding of video streams resulting from the transformation into DCT and wavelets. The description of the process indicates the steps to be applied to implement a classified vector quantization, increasing the compression and the efficiency of the encoding. Only one stream is transmitted to the receiver. The technical problem and the objective posed in this document are to optimize the digital format and it aims to obtain a digital stream formatted at the output of a digital encoder. The process described in this solution does not make it possible to secure the video stream and does not provide any protection against illicit uses of video streams originating from wavelet encoding.

As regards the protection of wavelet-coded images, the prior art knows the document EP 1 033 880 which relates to a method and a device for protection by modifications applied to the spatio-frequency coefficients. These modifications are of the type: modification of the sign bits of the coefficients, modification of the improvement bits of the coefficients, choice of appropriate coefficients belonging to a frequency sub-band to swap them, rotation of a block grouping frequency coefficients arranged in ascending order, while trying to respect as much as possible the statistical properties and the entropy of the original signal. Each type of modification is conditioned using a key. The data thus protected is then passed through an entropy encoder and a bitstream conforming to the standard is generated. This prior art represents an encryption solution using keys, therefore, a single stream is transmitted to the receiver and all the elements constituting the original stream are inside the protected stream. This document relates to a solution that does not respond satisfactorily to the protection of the transmitted video stream. In addition, following the modifications before the entropy encoding, the statistical properties are modified and the size of the stream and the bit rate increase. Consequently, this prior art does not meet the high security objectives guaranteeing a lossless process which is the subject of the present invention.

Another reference of the prior art is the document WO 00 31964 A relating to a method and an equipment for the partial encryption of images with the aim of protecting them and optimizing the storage space. A first part of the image is compressed at low quality, without encryption, and a second part of the image is encrypted. When the first and second parts are combined, the image of maximum quality is obtained. The second part is encrypted and includes two sections, encrypted differently. The decryption of the first section and its combination with the first part restores the initial image with average quality. The decryption of the second section, its combination with the first section and the first part restore the image original at maximum quality. The image can also be partitioned into multiple independent sections, each section being encrypted with its own method and its own key. The protection method in this prior art is encryption, therefore, all the original elements of the flow remain inside the protected flow, the restitution of the entire content from the only protected flow is possible in the case where a person is ill. intentioned finds or simulates the encryption keys. As with the previous document, this solution does not provide satisfactory security against pirating of the video stream. Also, the size of the protected stream is different from the size of the original stream. This prior art therefore does not solve the problem of high security, while providing fine granularity in the quality of the reconstructed video sequences treated in the present invention.

Among the prior art relating to the secure distribution of audiovisual streams organized in multilayers based on the client - server principle, US patent 2001/0053222 A1 proposes a method and a system for the protection of video streams encoded according to the MPEG-4 standard. The audiovisual stream is composed of several audio and video objects, managed by a scenic composition. One of the objects in the video stream is encrypted using a key generated in four encryption stages which is renewed periodically. Protected objects are video objects. The encrypted object is multiplexed with the other objects and the entire stream is sent to the user. The MPEG-4 stream is reconstructed on the destination equipment by the decryption module, which reconstitutes the original video stream, from the encrypted video stream, and by regenerating the encryption key from encryption information sent beforehand and from information contained in the encrypted stream. Since all the content protected from video objects is in the stream sent to the user, a malicious person who finds the encryption keys can decrypt this protected content and view or broadcast it. This prior art therefore does not entirely solve the problem of securing the video stream.

Unlike most of these systems

"Conventional" protection, the method according to the invention provides a high level of protection while reducing the volume of information necessary to access the original content from the protected content.

The protection, carried out in accordance with the invention, is based on the principle of deletion and replacement of certain information coding the original visual signal by any method, namely: substitution, modification, permutation or displacement of the information. This protection is also based on knowledge of the structure of the bit stream at the output of the wavelet-based visual encoder.

The present invention relates to the general principle of a method for securing an audiovisual stream. The objective is to authorize video-on-demand and pay-per-view services across all broadcasting networks and local recording in the user's digital set-top box, as well as live viewing of television channels . The solution consists in permanently extracting and storing outside the user's home, in fact in the broadcasting and transmission network, part of the digital audiovisual stream recorded at the customer's premises or broadcast live, this part being essential for viewing said digital audiovisual stream on a television screen or monitor type, but being of a very low volume compared to the total volume of the stream digital audiovisual recorded at the user's home or received in real time. The missing part will be transmitted via the broadcasting or transmission network when viewing said digital audiovisual stream.

The digital audiovisual stream being modified and separated into two parts, most of the modified audiovisual stream, called “modified main stream” will therefore be transmitted via a conventional broadcasting network while the missing part called “additional information” will be sent to the request via a narrowband telecommunications network such as conventional telephone networks or cellular networks of the GSM, GPRS or UMTS type or by using a small part of a DSL or BLR type network, or even by using a subset of the shared bandwidth over a wired network. The original digital audiovisual stream is reconstructed on the recipient equipment (decoder) by a synthesis module from the modified main stream and additional information. The invention provides a protection system, comprising an analysis - scrambling and descrambling module based on a digital format resulting from the encoding of a video stream based on wavelet transforms. The analysis and scrambling module proposed by the invention is based on the substitution by “decoys” or the modification of part of the coefficients resulting from the transformation into wavelets. The fact of having removed and substituted a part of the original data of the original digital stream during the generation of the modified main stream, does not allow the restitution of said original stream from only the data of said modified main stream. By the characteristics of “scalability” of the wavelet transform, the notion of “scalability” being defined at starting from the expression in English "scalability" which characterizes an encoder capable of encoding or a decoder capable of decoding an ordered set of digital streams so as to produce or reconstruct a multilayer sequence, several variants of the scrambling and descrambling are implemented and are illustrated with exemplary embodiments.

The invention relates, in its most general sense, to a method for the secure distribution of video sequences according to a digital stream format resulting from an encoding based on wavelet processing, consisting of frames comprising blocks containing coefficients d wavelets describing the visual elements, characterized in that before the transmission to the client equipment, a flow analysis is carried out to generate a modified main flow, by deleting and replacing certain information coding the original flow and presenting the format of the original stream, and additional information of any format, comprising said digital information coding the original stream capable of allowing the reconstruction of said modified frames, then separately transmitting said modified main stream and said complementary information thus generated from the server to the destination equipment.

Protection is carried out by deleting the original elements and replacing them with decoys, the extracted original elements being stored separately in the additional information. The fact of having removed and substituted part of the original data of the original video stream during the generation of the modified main stream does not allow the restitution of said original stream from only the data of said modified main stream. The video stream is fully protected (all sub-bands) and fully transmitted via the network or via physical media to the user, regardless of their rights. The partial reconstruction is carried out by sending part of the additional information, containing the original elements, either directly or in progressive mode.

The analysis and scrambling module decides how to visually degrade the video stream according to its structure and scalability properties resulting from the wavelet transform. The study focuses on the impact of the modification of different parts of the flow (coefficients, sub-bands, scalability layers, areas of interest) on visual degradation. Preferably, the scrambling is carried out by modifying wavelet coefficients belonging to at least one temporal sub-band resulting from the temporal analysis.

Advantageously, the scrambling is performed by modifying wavelet coefficients belonging to at least one spatial sub-band resulting from the spatial analysis of a temporal sub-band.

Advantageously, scrambling is carried out by modifying wavelet coefficients belonging to at least one temporal sub-band resulting from a temporal analysis of a spatial sub-band.

Advantageously, the wavelet coefficients to be modified are chosen according to random laws and / or defined a priori. According to a particular embodiment, the parameters for scrambling depend on the properties of temporal scalability and / or spatial scalability and / or qualitative scalability and / or flow scalability and / or scalability by regions of interest offered by digital streams generated by wavelet-based coders.

Advantageously, the visual intensity of the degradation of the video sequences obtained is determined by the quantity of wavelet coefficients modified in each space-time sub-band.

Advantageously, the intensity of the visual degradation of the video sequences decoded from the modified main stream is a function of the position within the original digital stream of the modified data, said data representing, according to their positions, the values quantified according to different precisions of the wavelet coefficients belonging to a space-time sub-band. Advantageously, the intensity of the visual degradation of the video sequences decoded from the modified main stream is determined according to the membership in the quality layer of the wavelet coefficients modified in each space-time sub-band. According to a particular mode of implementation, the modification of the wavelet coefficients is carried out directly in the bit stream.

According to a variant, the modification of the wavelet coefficients is carried out with partial decoding. According to another variant, the modification of the wavelet coefficients is carried out during coding or by carrying out a decoding then a complete re-encoding.

Advantageously, the size of the modified main stream is strictly identical to the size of the original digital video stream.

According to another variant, the substitution of the wavelet coefficients is carried out with random or calculated values. Preferably, the duration of the visual scrambling obtained within a group of frames is determined as a function of the time sub-band to which the modified wavelet coefficients belong. Advantageously, the visual scrambling obtained within a group of frames is spatially limited in a region of interest of each frame.

In addition, the additional information is organized into layers of temporal and / or spatial and / or qualitative and / or flow and / or scalability scalability.

In a variant, the stream is gradually descrambled with different levels of quality and / or resolution and / or frame-rate and / or according to a region of interest via the sending of certain parts of the additional information corresponding to the scalability layers respectively qualitative and / or spatial and / or temporal and / or for a region of interest.

According to another variant, the stream is partially descrambled according to different levels of quality and / or resolution and / or frame-rate and / or according to a region of interest via the sending of part of the additional information corresponding to the (aux) scalability layer (s) respectively qualitative and / or spatial and / or temporal and / or for this region of interest. According to a particular implementation mode, a summary of a digital stream in original format is calculated on the recipient equipment as a function of said modified main stream and of said additional information. According to a particular mode of implementation, the transmission of said modified main stream is carried out through a physically distributed hardware medium [CD-ROM, DVD, hard disk, flash memory card].

Advantageously, the modified main stream undergoes transcoding, rearrangement and / or extraction of frames or groups of frames during its transmission.

Advantageously, the transmission of said additional information is carried out through a physically distributed hardware medium [flash memory card, smart card].

Preferably, the modification of the wavelet coefficients is perfectly reversible (lossless process) and the digital stream reconstituted from the modified main stream and the additional information is strictly identical to the original stream.

Advantageously, the modification of the wavelet coefficients is perfectly reversible (lossless process) and the portion of the digital stream reconstructed from the modified main stream and the additional information is strictly identical to the corresponding portion in the original stream.

According to a particular variant, the reconstruction of a descrambled video stream is controlled and / or limited in terms of predefined frame-rate and / or resolution and / or bit rate and / or quality, according to the rights of

1 user.

According to another variant, the reconstruction of a descrambled video stream is limited in terms of frame rate and / or resolution and / or bit rate and / or quality depending on the display device on which it is viewed.

According to another variant, the reconstitution of the descrambled video stream is carried out gradually in stages until the reconstruction of the original video stream.

The invention also relates to a system for manufacturing a video stream, comprising at least one multimedia server containing the video sequences. original and comprising a device for analyzing the video stream, a device for separating the original video stream into a main stream modified by deleting and replacing certain information coding the original visual signal and into additional information as a function of said analysis, and at at least one device on the recipient equipment for the reconstruction of the video stream as a function of said modified main stream and of said additional information.

The present invention will be better understood on reading the description of a nonlimiting exemplary embodiment which follows, referring to the figure describing the overall architecture of a system for implementing the method according to the invention .

The protection of the visual streams described is developed based on the structure of the binary streams and their characteristics due to wavelet-based encoding. We recall this structure below. A video coder based on wavelet processing performs a temporal and spatial decomposition of an initial video sequence in order to obtain a set of spatio-temporal wavelet coefficients. These coefficients are then quantified and then coded by an entropy coder in order to generate one or more nested bit streams which have properties of temporal scalability and / or spatial scalability (or resolution) and / or qualitative scalability and / or flow scalability and / or scalability of regions of interest.

The property of temporal scalability is the possibility of decoding from a single or several nested bit streams, video sequences whose frame display frequency (number of frames per second) is variable.

In the literature well known to those skilled in the art, the concepts in English are used "frames" and "frame rate" (number of frames per second), concepts which will be used later to describe the present invention.

The spatial scalability property is the possibility of decoding from a single or from several nested bit streams video sequences whose spatial resolution (size) of the frames is variable.

The qualitative scalability property is the possibility of decoding from a single or several nested bit streams video sequences whose visual quality of the frames, measured according to objective and / or subjective criteria, is variable.

The rate scalability property is the possibility of decoding from a single or several nested bit streams video sequences according to a variable average rate (average number of bits of information per second).

The region of interest scalability property is the possibility of decoding from one or more nested bit streams one or more targeted areas within the video sequence.

When encoding, an original video sequence is segmented into groups of N successive frames called GOF (Group Of Frames), each GOF then being processed independently during encoding. We note a GOF of length N GOF = (F ₀ , F ₁ , ..., F _H _ ₁ ), F _± being the frames for i = 0,1, 2, ... Nl). The spatiotemporal wavelet coefficients are generated in two successive stages and according to a spatial and temporal analysis of the GOF frames. The first step consists in carrying out a temporal analysis of the N frames of each GOF according to the estimated direction of movement (temporal analysis with motion estimation), this in order to remove the temporal redundancies and to focus spatially in the frames resulting from the temporal analysis energy and information due to movement. This temporal analysis can be carried out according to different spatial resolutions and after decomposition into wavelets of each frame of the GOF, the motion estimation being carried out independently within each spatial sub-band (ulti-resolution motion estimation).

The second step consists in performing a spatial analysis of the N frames resulting from the temporal analysis using a wavelet decomposition in order to remove the spatial redundancies and to concentrate the energy due to the spatial discontinuities present in each frame.

The temporal analysis is carried out in several iterations.

At the first iteration, p successive frames of the original GOF are analyzed by means of wavelet filters f _L and f _H predefined of length p and after estimation and compensation of the movement for each frame and with respect to one or more frames called reference frames. Generally p = 2 and two successive frames (F _2i , F _{2i + 1} ), i = 0, ..., N / 2, are filtered and generate a frame called "low frequency" or "medium" and noted Li = f _L (F _2i , F _{2i + 1} ) and a frame called "high frequency" or "difference" and noted H _i = f _H (F _2i , F _{2i + 1} ). At the first iteration of the time analysis step, a subset tL _x of frames of type L of length N / 2 and a subset t-Hj are thus generated _. H type frames of length N / 2, such as tL- _L ≈ (L ₀ , L _x , ..., L _{N / 2} ), tH- _L ≈ (H ₀ , E _± , ..., H _{N 2} ). At each subsequent iteration k> l, the motion estimation / compensation and the time filtering are iterated over the subset of frames tL _k-1 of type L obtained in iteration k-1 and are generated two new sub- sets of frames tL _k and tH _k whose length (number of frames) is reduced by a factor of 2 compared to tL _k-1 . In some cases, the iteration also relates to the subset of frames tH ^.

The total number of iterations during the time analysis is noted n _τ . It is between 1 and N / 2. At the end of the temporal analysis, n _τ + l temporal sub-bands are generated.

For example, with N = 16 and n _τ ≈4:

The temporal analysis of a GOF of length N = 16 with n _τ = 4 thus generates 16 new frames distributed in n _τ + l = 4 + l = 5 temporal sub-bands:

"Sub-band tL ₄ : 1 frame of type L: LLLL ₀ ," Sub-band tH ₄ : 1 frame of type H: LLLH ₀ , "Sub-band tH ₃ : 2 frames of type H: LLH ₀ , LLH ^ ,

"Sub-band tH ₂ : 4 frames of type H: LH ₀ , ..., LH ₃ "Sub-band tE ^: 8 frames of type H: H ₀ , ..., H _7. The spatial analysis is then carried out on each of the frames belonging to each temporal sub-band t-Li and t-Hi: each frame is decomposed using a discrete wavelet transform at D levels, thus generating 3xD + l spatial sub-bands of wavelet coefficients for each frame. These spatial sub-bands are denoted s-LL ₀ , s- HL ^ S-LI ^, s-HH _lf s-HL ₂ , s-LH ₂ , s-HH ₂ , ..., s — HL _D , s — LH _D , S — HH _D. temporal and spatial analyzes,

(n _τ + l) x (3D + 1) space-time sub-bands of wavelet coefficients are available: tL _nT (s-LL ₀ ), tL _nT (s-HL ₁ ), tL _nT (s- HHI), ..., tL _nT (S _HL-D), tL _nT _(S-D LH), tL _nT _(S-D HH), _"Nos tH (s-LL ₀₎ tH _nT (s-HL ₁ ), tH _nT (s-LH _x ), tH _nT (s- ₁ ), ..., tH _nT (S-HL _D ), tH _nT (S-LH _D ), tH _nT ( S-HH _D ), m ... t-Hi (S-LL ₀ ), tH ^ (S-HLi), tH _x (SL ^), tH _x (S-HH), ..., The wavelet coefficients of each space-time sub-band are then progressively compressed by bit plane using an entropy coder whose aim is to remove the statistical redundancies existing within a fixed set of wavelet coefficients. The entropy coder generates, for each set of independently coded wavelet coefficients, a binary f lux which can be divided into several substreams distributed according to different quality layers.

After an analysis of the sub-band structure described above, the analysis and scrambling module according to the invention makes modifications (by permutation and / or substitution and / or threshold threshold) of a subset coef f hereents d Wavelets belonging to one or more space-time sub-bands. These modifications introduce a visually perceptible degradation (scrambling) of the video sequence decoded from these modified coefficients. Depending on the number of modified coefficients, their locations in a spatial sub-band, their membership in the space-time sub-bands, their membership in one or more quality layers, their position within the set of coefficients belonging to a single space-time sub-band and of the type of modification, a control of the spatial and / or temporal extent and / or according to the quality layers of the scrambling, as well as a control of the intensity degradation due to scrambling is possible. By modifying wavelet coefficients in spatial sub-bands of a temporal sub-band tX of length l _x (ie containing l _x frames), the N / l _x consecutive frames of the GOF are scrambled.

The choice of the type of spatial sub-band to which the Wavelet coefficients belong (s-HL or s-LH or s-HH) makes it possible to control the visual aspect of the scrambling: for the s-HL sub-band of the vertical direction artifacts appear on the frames (degradation of vertical spatial discontinuities), for the s-LH subband of horizontal artefacts appear (degradation of horizontal spatial discontinuities) and for the s-HH subband of checkerboard type artifacts »Appear (joint degradations of horizontal and vertical spatial discontinuities). The choice of the resolution level r to which the spatial sub-band belongs (s-LL _r or s-HL _r or s-LH _r or s-HH _r ) makes it possible to control the spatial extent of the scrambling generated by the modification coefficients wavelets: the closer r is to 0, the larger the spatial extent.

A modification of the wavelet coefficients belonging to a sub-band of resolution r ≥ 0 generates a scrambling which will be visible on all the decoded frames with higher spatial resolution r + 1, r + 2, ...

, R.

A modification of the wavelet coefficients belonging to a quality layer q generates a scrambling which will be visible on all the decoded frames by considering at least the first q quality layers.

The modification of the spatio-temporal wavelet coefficients is carried out after a partial decoding of the bit stream generated according to a standard or norm or an algorithm or an encoding format. Once the modification has been carried out, a re-encoding of the coefficients is carried out in order to generate a bit stream of identical size and which respects the conformity with respect to the standard or the norm or the algorithm or the encoding format. having generated the original bitstream.

Advantageously, subsets of bits inside the original bit stream representing the coefficients of coded spatio-temporal wavelets are modified without decoding and without disturbing the conformity of the stream vis-à-vis the standard or norm or of the algorithm or encoding format that generated the original bitstream.

The choice of the space-time wavelet coefficients to be modified within a space-time sub-band is conducted randomly and / or according to rules defined a priori.

Advantageously, the modified main stream has a size identical to that of the original video stream. The scrambling thus generated has properties of temporal, spatial, qualitative, flow and area of interest scalability.

Advantageously, the additional information relating to said scrambling thus generated is organized into layers of temporal, spatial, qualitative, flow rate scalability and on area of interest.

Depending on the number of GOFs and / or the number of frames scrambled within a GOF, the scrambling has a temporal scalability between: "all the frames of all the GOFs (maximum scrambling)" and "no frame of no GOF ”(sequence not scrambled).

As a function of the resolutions of the spatial sub-bands to which the modified wavelet coefficients belong, the scrambling has a spatial scalability comprised between: "all the resolutions are scrambled" (ie from the resolution r≈O to the resolution r = R) and "None of the resolutions are confused".

Depending on the number of modified wavelet coefficients and the resolutions of the spatial sub-bands to which they belong, the scrambling has a qualitative scalability ranging from: "the whole of each frame is scrambled", "certain spatial regions of each frames are scrambled ”(regions of interest) and“ no scrambling is applied to the frames ”.

Reciprocally, descrambling will also present the different stated scalability (temporal, spatial, qualitative, bit rate and area of interest).

Advantageously, said descrambling will make it possible to address the various stated scalabilities (temporal, spatial, qualitative, bit rate and area of interest), by sending certain parts of the additional information corresponding to different layers of scalability (temporal, spatial, qualitative, bit rate and area of interest), thus giving access to different levels of quality / resolution / frame-rate for the video sequence decoded from the partially descrambled stream.

Advantageously, the different levels of quality / resolution / frame rate of the video sequence are obtained from the partially descrambled flow by sending part of the additional information by scalability layer (temporal, spatial, qualitative, bit rate and area of interest).

We will better understand the principle of scrambling and descrambling based on these different scalabilities using the nonlimiting preferred embodiment which follows.

In the appended drawing, the figure represents a particular preferred embodiment of the client-server system according to the invention.

The original stream (1) is directly in digital form or in analog form. In the latter case, the analog stream is converted by a wavelet-based encoder not shown in a digital format (2). The video stream that we want to secure (2) is passed to an analysis and scrambling module (3) which will generate a modified main stream (5) in the same format as the input stream (2) outside that some of the coefficients have been replaced by values different from the original ones, and is stored in the server (6). The additional information (4), of any format, is also placed in the server (6) and contains information relating to the elements of the images which have been modified, replaced, substituted or moved, and their value or location in the original feed.

The stream (5) in the format identical to the original stream is then transmitted, via a broadband network (9) of the radio type, cable, satellite, etc., to the viewer's terminal (8), and more precisely to its hard disk ( 10). When the viewer (8) requests to view the film on his hard drive (10), two possibilities are possible: either the viewer (8) does not have all the rights necessary to see the film, in this case, the video stream (5) generated by the scrambling module (3) present on the hard disk (10) is passed to the synthesis system (13), via a read buffer memory (11), which does not modify it and transmits identically to a display reader capable of decoding it (14) and its content, visually degraded by the scrambling module (3), is displayed on the display screen (15). Advantageously, the video stream (5) generated by the scrambling module (3) is passed directly via a network (9) to the read buffer memory (11) then to the synthesis system (13).

Advantageously, the video stream (5) undergoes in the network (9) a series of operations for transcoding and rearranging its frames or groups of frames. Either the server decides that the viewer (8) has the rights to correctly view the film. In this case, the synthesis module (13) makes a viewing request to the server (6) containing the additional information necessary (4) for the reconstruction of the original video (2). The server (6) then sends via the telecommunication network (7) of analog or digital telephone line type, DSL (Digital Subscriber Line) or BLR (Local Radio Loop), via DAB (Digital Audio Broadcasting) networks or via networks. of digital mobile telecommunications (GSM, GPRS, UMTS), the additional information (4), allowing the reconstruction of the original video, so that the viewer (8) can store it in a buffer memory (12). The synthesis module (13) then restores the scrambled video stream which it reads in its playback buffer memory (11), modified fields of which it knows the positions as well as the original values are restored, thanks to the content of the additional information read in the descrambling buffer (12). The amount of information contained in the additional information (4) which is sent to the descrambling module is specific, adaptive and progressive for each spectator and depends on their rights, for example single or multiple use, right to make one or more copies private, late or anticipated payment. The level (quality, quantity, type) of the additional information is also determined according to the visual quality required by the user. Wavelet-based video coding characterized by the previously described scalability enables the video stream to be reproduced with different quality, resolution and frame rate levels.

Advantageously, the modified main stream (5) is passed directly via a network (9) to the read buffer memory (11) then to the synthesis module (13).

Advantageously, the modified main stream (5) is written (recorded) on a physical medium such as a CD-ROM, DVD, hard disk, flash memory card, etc. (9a). The modified main stream (5) will then be read from the physical medium (9bis) by the disk drive (10bis) of the box (8) to be transmitted to the read buffer memory (11) then to the synthesis module (13) . Advantageously, the additional information (4) is recorded on a physical medium (7bis) of credit card format, consisting of a chip card or a flash memory card. This card (7bis) will be read by the module (12) of the device (8) which includes a card reader (7ter).

Advantageously, the card (7bis) contains the applications and the algorithms which will be executed by the synthesis system (13). Advantageously, the device (8) is an autonomous, portable and mobile system.

We will detail the operation of the analysis and scrambling module (3) illustrating the choice of scrambling performed. The original video sequence is segmented into GOF of N = 16 frames. Temporal analysis with n _τ = 4 iterations generates n _τ + l = 5 temporal sub-bands having respectively:

"Sub-band tL ₄ : 1 frame of type L: LLLL ₀ " Sub-band tH ₄ : 1 frame of type H: LLLH ₀ "Sub-band tH ₃ : 2 frames of type H: LLH ₀ , LLH _X " Sub -band tH ₂ : 4 frames of type H: LH _{0 /} LH _{1 (} LH ₂ LH ₃

"Sub-band tH ₁ : 8 frames of type H: H ₀ , H _x , H ₂ , H ₃ , H ₄ , H ₅ , H ₆ , H _7. The decomposition into five time sub-bands offers the possibility of reconstructing the initial video sequence according to five different “frame rates” Each frame of resolution R in each time sub-band tX is then decomposed spatially by a wavelet transform at D = 4 levels, which gives the possibility of reconstructing the image with five different resolutions, thus generating for each 3 x D + 1 = 13 spatial sub-bands: LL ₀ , LH _lf HL _X , HH _l LH ₂ , HL ₂ , HH ₂ , LH ₃ , HL ₃ , HH ₃ , LH ₄ , HL ₄ , HH ₄ . Following such encoding, the video sequence can therefore be decoded according to frame rates ranging from 1/16 x fr ₀ to fr ₀ , fr ₀ being the frame rate of the original video sequence as well as according to D + l = 5 resolutions. The video sequence is scrambled for each GOF as follows:

In the time sub-band tL ₄ , the wavelet coefficients of the spatial sub-bands s-HH ₂ and s-HH ₃ resulting from the spatial decomposition into wavelets of the frame LLLL ₀ are extracted and replaced by random or calculated values.

In the time sub-band tH ₃ , the wavelet coefficients of the spatial sub-band s-HH ₃ resulting from the spatial decomposition into wavelets of the frame LLH ₀ are extracted and replaced by random or calculated values.

In the time sub-band tH _lf the wavelet coefficients of the spatial sub-band s-HH ₃ resulting from the decomposition into wavelets of the frame LH ₀ are extracted and replaced by random or calculated values.

In the time subband tH ₀ , the wavelet coefficients of the spatial subband s-HH ₃ resulting from the decomposition into wavelets of the frame H ₀ are extracted and replaced by random or calculated values.

The video sequence decoded from the modified main stream is therefore completely scrambled, unless it is decoded at frame rates equal to l / 16xfr ₀ and l / 8xfr ₀ (decoding only from time sub-bands tL ₄ and (tL _4f tH ₄ )) and at spatial resolutions r = R / 16 or r = R / 8 (decoding only from the spatial sub-bands s-LL ₀ , s-LH _x , s-HL _! and sH respectively) ^ which have not been changed). We This makes changes to all time bands, but not to all resolutions. Leaving resolutions unchanged makes it possible to be able to reconstruct the video stream from the scrambled stream, but at a quality much lower than that of the original video stream.

The flow thus scrambled is transmitted to the client

(8) at its request, descrambling is then carried out for example in five stages, corresponding to different levels of quality obtained after each descrambling stage. In this way, a descrambling of one or more scalability layers is carried out, the quality of the film viewed being controlled by the server according to the rights of the user and the quality required by him.

Advantageously, this descrambling results in a gradual attenuation of the degradation over time until the original content of high visual quality is reconstructed. For example, the first descrambling step consists in restoring the wavelet coefficients of the spatial sub-band s-HH ₂ for the time sub-band tL ₄ . The scrambling of the decoded video sequence for a maximum resolution R and at a maximum frame rate of fr ₀ is then less spatially extended and more concentrated around the spatial discontinuities of each frame. The decoded video sequence for the frame rate l / 16xfr ₀ or l / 8xfr ₀ and for the resolutions r = R / 16, R / 8, R / 4 is however not at all confused. The visual quality of the partially descrambled film is minimal, the film is visually unusable at full resolution and at its original frame rate. This step serves to identify the additional information server when establishing the connection. The second descrambling step consists in restoring the wavelet coefficients of the spatial sub-band s-HH ₃ for the temporal sub-band tL ₄ . The scrambling of the decoded video sequence for a resolution R and at a frame rate of fr ₀ is then much less spatially extended and more concentrated around the spatial discontinuities of each frame. In addition, the video sequence is now scrambled only for a duration equivalent to half a GOF (8 frames out of 16). The decoded video sequence for the frame rates l / 16xfr ₀ or l / 8xfr _Q and for the resolutions r = R / 16, R / 8, R / 4 is not scrambled. This step is used for the user (8) to view the video film partially, in order to decide whether he wants to obtain rights to watch the film. After confirmation by the client, and depending on the payment made, steps three, four or five are executed. The third descrambling step consists in restoring the wavelet coefficients of the spatial sub-band s-HH ₃ for the temporal sub-band tH ₃ . The video sequence is now scrambled only for a period equivalent to a quarter of a GOF (4 frames out of 16). The decoded video sequence for the frame rates l / 16xfr ₀ , l / 8xfr ₀ , l / 4xfr ₀ and for the resolutions r = R / 16, R / 8, R / 4 is not scrambled. The film can be viewed visually, but it is of low quality.

The fourth descrambling step consists in restoring the wavelet coefficients of the spatial sub-band s-HH ₃ for the temporal sub-band tH ₂ . The video sequence is now scrambled only for a period equivalent to one eighth of a GOF (2 frames out of 16). The decoded video sequence for frame rates l / 16xfr ₀ , l / 8xfr ₀ , l / 4xfr ₀ , l / 2xfr ₀ and for resolutions r = R / 16, R / 8, R / 4 is not scrambled . The visual quality of the film returned is average. For these 3 ^rd and 4 ^th stages of descrambling, the full resolution video film and for the original frame rate remains partially scrambled, but from these steps, we can extract video streams with lower resolutions and frame-rates than the original film. This gives the possibility of providing versions of the same video stream of lower resolution, therefore at a lower price and better controlling access.

The fifth descrambling step consists in restoring the wavelet coefficients of the spatial sub-band s-HH ₃ for the temporal sub-band tH _x . The yideo sequence is then completely descrambled, whatever the decoding frame-rate and the resolution. The reconstructed video stream is strictly identical to the original video stream.

Claims

1. Method for the secure distribution of video sequences according to a digital stream format resulting from an encoding based on wavelet processing, consisting of frames (frames) comprising blocks containing wavelet coefficients describing the visual elements, characterized in what we do, before transmission to the client equipment, an analysis of the flow to generate a main flow modified by deleting and replacing certain information coding the original flow and presenting the format of the original flow, and additional information of any format, comprising said digital information coding the original stream able to allow the reconstruction of said modified frames, then separately transmitting said modified main stream and said additional information thus generated from the server to the recipient equipment.

2. Method for the secure distribution of video sequences according to claim 1, characterized in that the scrambling is carried out by modifying coefficients of wavelets belonging to at least one temporal sub-band resulting from the temporal analysis.

3. Method for the secure distribution of video sequences according to one of the preceding claims, characterized in that the scrambling is carried out by modifying wavelet coefficients belonging to at least one spatial sub-band resulting from the spatial analysis of a time sub-band.

4. Method for the secure distribution of video sequences according to one of the preceding claims, characterized in that the scrambling is carried out by modifying wavelet coefficients belonging to at least one temporal sub-band resulting from the temporal analysis of a spatial sub-band.

5. Method for the secure distribution of video sequences according to one of the preceding claims, characterized in that the wavelet coefficients to be modified are chosen according to random laws and / or defined a priori.

6. Method for the secure distribution of video sequences according to one of the preceding claims, characterized in that the parameters for scrambling depend on the properties of temporal scalability and / or spatial scalability and / or qualitative scalability and / or temporal scalability, flow scalability and / or scalability by regions of interest offered by the digital streams generated by wavelet-based coders.

7. Method for the secure distribution of video sequences according to one of the preceding claims, characterized in that the visual intensity of the degradation of the video sequences obtained is determined by the quantity of wavelet coefficients modified in each spatial sub-band -temporelle.

8. Method for the secure distribution of video sequences according to one of the preceding claims, characterized in that the intensity of the visual degradation of the video sequences decoded at from the modified main stream is a function of the position within the original digital stream of the modified data, said data representing, according to their positions, the values quantified according to different precisions of the wavelet coefficients belonging to a space-time sub-band.

9. Method for the secure distribution of video sequences according to one of the preceding claims, characterized in that the intensity of the visual degradation of the video sequences decoded from the modified main stream is determined according to belonging to a quality layer wavelet coefficients modified in each space-time sub-band.

10. Method for the secure distribution of video sequences according to one of the preceding claims, characterized in that the modification of the wavelet coefficients is carried out directly in the bit stream.

11. Method for the secure distribution of video sequences according to one of the preceding claims, characterized in that the modification of the wavelet coefficients is carried out with partial decoding.

12. Method for the secure distribution of video sequences according to one of the preceding claims, characterized in that the modification of the wavelet coefficients is carried out during coding or by carrying out a decoding then a complete re-encoding.

13. Method for the secure distribution of video sequences according to one of the preceding claims, characterized in that the size of the modified main stream is strictly identical to the size of the original digital video stream.

14. Method for the secure distribution of video sequences according to one of the preceding claims, characterized in that the substitution of the wavelet coefficients is carried out with random or calculated values.

15. Method for the secure distribution of video sequences according to one of the preceding claims, characterized in that the duration of the visual scrambling obtained within a group of frames is determined as a function of the time sub-band at which belong the coefficients of ¹ modified wavelets.

16. Method for the secure distribution of video sequences according to one of the preceding claims, characterized in that the visual scrambling obtained within a group of frames is spatially limited in a region of interest of each frame.

17. Method for the secure distribution of video sequences according to one of the preceding claims, characterized in that the additional information is organized in layers of temporal and / or spatial and / or qualitative and / or qualitative and / or bit rate scalability and / or in a zone of interest.

18. Method for the secure distribution of video sequences according to one of the preceding claims, characterized in that the stream is gradually descrambled with different quality levels and / or resolution and / or frame-rate and / or according to a region of interest via the sending of certain parts of the additional information corresponding to the layers of scalability respectively qualitative and / or spatial and / or temporal and / or for a region of interest.

19. Method for the secure distribution of video sequences according to one of the preceding claims, characterized in that the stream is partially descrambled according to different quality levels and / or resolution and / or frame-rate and / or according to a region of interest via the sending of part of the additional information corresponding to the scalability layer (s) respectively qualitative and / or spatial and / or temporal and / or for this region of interest.

20. Method for the secure distribution of audiovisual sequences according to one of the preceding claims, characterized in that a summary of a digital stream in original format is calculated on the recipient equipment as a function of said modified main stream and of said additional information.

21. Method for the secure distribution of audiovisual sequences according to one of the preceding claims, characterized in that the transmission of said modified main stream is carried out through a physical medium physically distributed [CD-ROM, DVD, hard disk, memory card flash].

22. Method for the secure distribution of audiovisual sequences according to one of the preceding claims, characterized in that the modified main stream undergoes transcoding, rearrangement and / or extraction of frames or groups of frames during its transmission .

23. Method for the secure distribution of audiovisual sequences according to one of the preceding claims, characterized in that the transmission of said additional information is carried out through a physical medium physically distributed [flash memory card, smart card].

24. Method for the secure distribution of video sequences according to one of the preceding claims, characterized in that the modification of the wavelet coefficients is perfectly reversible (lossless method) and that the digital stream reconstituted from the modified main stream and of the additional information is strictly identical to the original feed.

25. Method for the secure distribution of video sequences according to one of the preceding claims, characterized in that the modification of the wavelet coefficients is perfectly reversible (lossless method) and that the portion of the digital stream reconstituted from the main stream modified and additional information is strictly identical to the corresponding portion in the original feed.

26. Method for the secure distribution of video sequences according to one of claims 24 or 25, characterized in that the reconstruction of a video stream descrambled is controlled and / or limited in terms of frame rate and / or resolution, and / or speed and / or quality predefined according to the rights of the user.

27. A method for the secure distribution of video sequences according to one of claims 24 or 25, characterized in that the reconstruction of a descrambled video stream is controlled and / or limited in terms of frame rate and / or resolution, and / or predefined speed and / or quality depending on the display device on which it is displayed.

28. Method for the secure distribution of video sequences according to one of claims 24, 25, 26 or 27, characterized in that the reconstruction of the descrambled video stream is carried out gradually in stages until the reconstruction of the original video stream .

29. System for manufacturing a video stream for implementing the method according to one of the preceding claims, comprising at least one multimedia server containing the original video sequences and characterized in that it comprises an analysis device of the video stream, a device for separating the original video stream into a main stream modified by deleting and replacing certain information coding the original visual stream and into additional information according to said analysis, and at least one device on the recipient equipment for the reconstruction of the video stream as a function of said modified main stream and of said additional information.