CN100581249C - Encoding and decoding method in copyright protection of streaming video content - Google Patents

Abstract

The invention relates to an encoding and decoding method in copyright protection of streaming media video content, and belongs to the technical field of copyright protection of video content. Encoding, DCT, quantization, scrambling, and entropy encoding are performed on the collected video content to generate and encrypt a 3GP file; the PDCF file is obtained by adding the encryption algorithm name, video content identifier, address information of the video content authority publisher, and the index box and SDP file and send it in RTP packet. The streaming media client parses the RTP packet to obtain and decrypt the video frame, and obtain playable RGB or YUV data through entropy decoding, anti-scrambling, anti-quantization and anti-DCT. The advantage of the method of the present invention is that by scrambling and re-encrypting the DCT coefficients, the video content is still in an information-fuzzy state after being decrypted, the copyright protection of the video content is strengthened, and the interaction between the streaming media and the application end in the system is enhanced through custom information. Improved system security.

Description

Encoding and decoding method in copyright protection of streaming video content

technical field

The invention relates to an encoding and decoding method in copyright protection of streaming media video content, and belongs to the technical field of copyright protection of video content.

Background technique

Encryption and protection of video content is an effective means of copyright protection for streaming media applications. However, the video streaming application system is composed of multiple modules, and the current video content protection method is single, and generally only provides protection for a certain module, without considering the entire system. The protection measures for each module in the video streaming application system are not perfect, and these problems are mainly caused by the following difficulties:

(1) At present, the encryption and protection of video content adopts the Advanced Encryption and Decryption Standard (hereinafter referred to as AES) encryption algorithm, mainly because the AES algorithm has a high strength against exhaustive attacks, known plaintext attacks and chosen plaintext attacks, making video encryption The system has high security. However, AES requires that the encrypted data block be an integer multiple of its basic encryption unit, and the size of the frame content is not fixed, so the existing video content protection methods all use partial encryption algorithms with lower security levels.

(2) Since the video frame data sent by the streaming media server to the application end is standard video coded data after decryption, if the application end data stored in the memory is copied by an illegal user, the illegal user will decode and play the standard video frame data that can be copied.

Due to the above reasons, the protection of video content in the streaming video application system is a systematic process, and the security of any protection module may become the bottleneck of system security, and one-sided emphasis on the security of a single module cannot provide the security of the entire system . Only when the security of each module in the system is guaranteed, the protection of video content in the streaming video application system can be regarded as successful. However, the existing solutions all consider the security of a certain module in the video content protection system, and the protection method scheme is not perfect, and there are potential safety hazards.

Contents of the invention

The purpose of the present invention is to propose a coding and decoding method in the copyright protection of streaming media video content, which can effectively control the entire process of video content from frame coding, frame encryption, RTP packet transmission, application side decryption and frame decoding, and finally achieve The purpose of protecting the original video content.

The encoding and decoding method in the stream media video content copyright protection that the present invention proposes comprises the following steps:

(1) Encode the collected video content to obtain discrete cosine transform coefficient matrix A, and perform quantization operation on matrix A to obtain matrix B, perform scrambling transformation on matrix B to obtain matrix C, perform entropy encoding on matrix C, and generate The video frame H generates a 3GP file T by successively generating a plurality of video frames according to the 3GP container standard;

(2) According to the control information in the above-mentioned 3GP file T, calculate the size s of each video frame H respectively, carry out the modulo b operation to s, b is the size of the encryption block in the Advanced Encryption and Decryption Standard Algorithm, if the calculation result w is not 0, b-w+b bytes are filled at the end of the corresponding video frame, if the calculation result w is 0, b bytes are filled, the filling bytes are initialized to 0, and the filled words are recorded in the last b bytes The number of sections b-w+b or b, the 3GP file T' is obtained, and the video frame after filling bytes is H';

(3) The video frame H' is encrypted using the advanced encryption and decryption algorithm to obtain the 3GP file T ", and the video frame H' is encrypted and converted into H";

(4) Add encryption algorithm name, video content identifier and video content rights issuer address information to above-mentioned 3GP file T ", obtain the digital rights management content format file P of packing, add self-defined instruction in file P, to form Illustrate the index box of the transmission mode of the streaming media real-time transmission protocol format data packet, and obtain the packaged digital rights management content format file P' supporting the streaming media application;

(5) The above-mentioned packaged digital rights management content format file P' is input to the streaming server, and the streaming server takes out relevant fields from the above-mentioned file P' to obtain the session description protocol file, and the streaming server receives the user's playing video After the content is requested, the above-mentioned file P' is packaged to obtain the first type of data packet conforming to the format of the real-time transport protocol, and at the same time, the above-mentioned session description protocol file is packaged to obtain the second type of data packet conforming to the format of the real-time transport protocol, and the second type of data packet is obtained One type of data packet and the second type of data packet are sent to the streaming media client;

(6) The streaming media client unpacks the above-mentioned received data packet, obtains the above-mentioned encrypted video frame H" and the above-mentioned encryption algorithm name, decrypts H" according to the encryption algorithm name, and obtains the video frame H', according to the video frame The number of padding bytes contained in the last b bytes of H', remove the padding bytes, and obtain video frame H;

(7) Carry out entropy decoding corresponding to the above-mentioned entropy coding to above-mentioned video frame H, obtain matrix C, carry out anti-scrambling transformation to matrix C, obtain matrix B, carry out inverse quantization to matrix B, obtain matrix A ', matrix A' performs an inverse discrete cosine transform to obtain video data for video playback.

The encoding and decoding method in the stream media video content copyright protection that the present invention proposes has the following advantages:

(1) Through a kind of encryption technology based on Discrete Cosine Transform (hereinafter referred to as DCT) coefficient scrambling, the video content in the streaming video application system adopting this method is still in the information fuzzy state after decryption, so the video content has been protected. Function, strengthening the copyright protection of video content. However, the traditional video encryption algorithm only uses the AES algorithm to encrypt the video content. After the video is decrypted, the data is still threatened by illegal piracy.

(2) In this method, after the content of the video frame is fully encrypted by the Advanced Data Encryption and Decryption Standard, the video content has reliable security. Since the packaged digital rights management content format (hereinafter referred to as PDCF) contains custom information that can specify multiple encryption algorithms, the selection of system encryption strategies is diversified, and the interaction between streaming media and application terminals in the system is enhanced. security.

Description of drawings

Fig. 1 is a functional block diagram of the method of the present invention.

Fig. 2 is a block flow diagram of encoding in the method of the present invention.

Fig. 3 is a flowchart of decoding in the method of the present invention.

Detailed ways

Encoding and decoding method in the stream media video content copyright protection that the present invention proposes, its block diagram as shown in Figure 1, comprises the following steps:

1. Encode the collected video content, generate discrete cosine transform coefficient matrix A, and perform quantization operation on matrix A to obtain matrix B, perform scrambling transformation on matrix B with fixed-size macroblock data to obtain matrix C, according to the standard The entropy encoding table performs entropy encoding on C to generate video frame H, and uses the open source tool Yamb MP4Tools to generate 3GP file T from multiple video frames successively generated according to the 3GP container standard.

2. According to the control information in the above-mentioned 3GP file T, calculate the size s of each video frame H respectively, carry out modulo b operation to s, and b is the size of the encryption block in the Advanced Encryption Decryption Standard (AES) algorithm, if the calculation result w If it is not 0, fill b-w+b bytes at the end of the corresponding video frame, and if the calculation result w is 0, fill b bytes. The filling byte is initialized to 0, and the number of filled bytes b-w+b or b is recorded in the first byte of the last b byte to obtain the 3GP file T', and the video frame after the filling byte is H'.

In one embodiment of the present invention, the process of calculating the size of each video frame H is as follows: sfopen opens the file T, reads 4 bytes into variable boxsize and variable boxname successively, compares the content in the variable boxname and the character string moov , if it does not match, move the file pointer forward by boxsize-8 bytes, and then read 4 bytes into the variable boxsize and the variable boxname, and repeat this process until the variable boxname matches the string moov. At this time, the file pointer points to the absolute position of the moov box (hereinafter referred to as Box) in the file. Read 4 bytes into the variable boxsize and the variable boxname successively, compare the content in the variable boxname with the string trak, if they do not match, move the file pointer forward by boxsize-8 bytes, and then read 4 bytes successively The bytes are stored in the variable boxsize and the variable boxname, and this process is repeated until the variable boxname matches the string trak. At this time, the file pointer points to the physical address of the first track (hereinafter referred to as track) in the file T, and saves it in the long integer variable track_loc, and copies the content of the variable boxsize to the variable track_size at the same time. Apply this process of finding a specified Box to finding mdia and hdlr Boxes. After this process ends, the pointer will point to the hdlr Box, and the file pointer will be moved forward by 8 bytes to skip the field version and field pre_defined in the hdlr Box, where the field version and field pre_defined occupy 4 bytes respectively. Read 4 bytes into the variable trak_type. And compare the variable trak_type with the vide string. If they are equal, it means that the video track is found, assign the current file pointer to the variable track_loc, otherwise move the file pointer to the file location specified by this track_loc, skip track_size-8 bytes, and perform the next round of comparison until Find the video track, and the file pointer will be positioned at the position of the video track.

After locating the video track, you can locate the stco Box according to the method of locating the video track, but change the nested trak, mdia, and hdlr of the Box in the process of locating the video track to mdia, minf, stbl, and stco, when the content of the variable boxname When matched against the string stco, returns a file pointer to the location of the stco Box.

Read 4 bytes into the variable conversion and the variable cocount successively, and create cocount instances of the chunkoffset class. The chunkoffset class includes the integer field chunkID and the long integer absolute position chunkloc of the first segment in the chunk (hereinafter referred to as sample) in the file. Read 4 bytes into the chunkID field of the first chunkoffset instance, and then read 4 bytes of file location information into the member variable chunkloc of the first chunkoffset instance. The above process of reading the variable chunkID and the variable chunkloc is repeated cocount times, and the information of the 8*cocount byte data block (hereinafter referred to as chunk) in the file is stored in cocount chunkoffset instances.

According to the method of locating stco Box, you can locate stsz Box. Just change the box nested mdia, minf, stbl and stco in the process of locating the video stco to mdia, minf, stbl and stsz, and when the variable boxname content matches the string stsz, return the file pointer pointing to the location of the stsz Box. Then read 4 bytes into the variables szversion, szsize and szcount respectively, and create a long integer array frame with the size of szcount. Read the content of 4 bytes into the first element of the array frame, and loop this process for szcount times. After the process ends, the frame array stores the size information of all video frames in the 3GP file in sequence.

According to the method of locating stco Box, you can locate stsd Box. Just change the box nested mdia, minf, stbl and stco in the process of locating the video stco to mdia, minf, stbl and stsd, and when the variable boxname content matches the string stsd, return the file pointer pointing to the location of the stsd Box. Read 4 bytes into the variable version and variable entry_count respectively. Create entry_count class SampleEntry objects. The SampleEntry object mainly includes the member variable width and member variable height indicating the resolution size, and the member variable frame_count indicating how many frames are in each sample. Advance the file pointer by 12 bytes, skipping reserved bytes. Then read 2 bytes into the member variable width and member variable height of the first SampleEntry object. Move the file pointer forward by 12 bytes, then read 2 bytes into the member variable frame_count of the first SampleEntry object, then move the file pointer forward by 8 bytes to skip irrelevant information, and read this information process If the entry_count is looped, the information about how many video frames are contained in each sample in the file will be stored in the SampleEntry object array.

According to the method of locating stco Box, you can locate stsc Box. Just change the box nested mdia, minf, stbl and stco in the process of locating the video stco to mdia, minf, stbl and stsc, and when the content of the variable boxname matches the string stsc, return the file pointer pointing to the location of stsc. Read 4 bytes into variable scversion and variable sccount successively and create sccount instances of chunk_sample_frame type. The type chunk_sample_frame has a field that stores the absolute position of the chunk in the file and the number of frames containing the sample in each chunk. Read 4 bytes of information successively into the member variable first_chunk and member variable samples_per_chunk of the first chunk_sample_frame object, and loop this process sccount times. After the process ends, the chunk_sample_frame instance stores the absolute position of the chunk in the 3GP file and the number of frames contained in each sample in sequence. Create a new type Chunk_frame. The Chunk_frame class contains 3 fields, which are the ID number of the chunk, the starting physical address of the chunk, and the number of frames contained in the chunk. According to the number of frames contained in each sample stored in the SampleEntry type array and the number of samples contained in each chunk stored in the chunk_sample_frame array, the number of frames in each chunk can be calculated, and the data Stored in the frameNum field of chunk_frame. So far, the chunk_frame array has saved all the information of the chunk, including the ID number of the chunk, the starting physical address of the chunk, and the number of frames contained in the chunk. The absolute position of each frame in the file can be accurately positioned through the chunk_frame array and the array frame.

3. The video frame H' is encrypted using the advanced encryption and decryption standard algorithm to obtain a 3GP file T", and the video frame H' is converted into H".

4. Add the name of the encryption algorithm, the video content identifier, and the address information of the video content rights issuer to the above-mentioned 3GP file T" to obtain a packaged digital rights management content format file P, and add custom instructions to the file P to form a description The index box of the transmission mode of the streaming media real-time transmission protocol format data packet, obtains the packaged digital rights management content format file P' that supports the streaming media application. The specific implementation process is:

Add sinf Box information to 3GP files according to ISO IEC 14496-12 standard. The sinf Box is designed in accordance with 3GPP TS 26.244700 and ISO IEC 14496-12 standards, and its content mainly includes encryption algorithm name, video content identifier and video content rights issuer address information. fopen opens T", follows the definition of the meta Box in the ISO IEC 14496-12 standard and uses the above method of locating the Box to search for the meta Box in T". If there is no meta Box in T", point the file pointer to the end of the file, and add byte information including the meta Box and its sub-Box sinf Box according to 3GPP TS 26.244 700 and ISO IEC 14496-12 standards. If T" contains meta Box , then read the meta Box information into the temporary variable t, add the custom information encryption algorithm name, video content identifier and video content authority publisher address information to the sinf Box according to the above method of locating the Box, and overwrite it with t Original meta Box information in "T". By adding the encryption algorithm name, video content identifier and video content rights issuer address information to the above-mentioned 3GP file T", the digital rights management content format (PDCF) file P of packaging is obtained; By using the open source tool QuickTime streaming function, a file P' containing an index (hint) Box is generated. The hint Box is composed of the audio and video index information in the moov Box in the file P, and is used to support the streaming application of the streaming media server.

5. The streaming media server analyzes the above-mentioned packaged digital rights management content format P' to obtain a session description protocol file (SDP). After the streaming media server receives the request to play the video content, it packs the above-mentioned packaged digital rights management content format file P' to obtain a real-time transport protocol (RTP) format data packet, and at the same time packs the session description protocol file into the real-time transport protocol format packets and send those packets to the streaming client.

6. The streaming media client parses the encrypted frame H" and the name of the encryption algorithm from the received real-time transport protocol format data packet, and uses the advanced encryption and decryption standard algorithm to decrypt H" to obtain the video frame H', according to The content of the first byte in the last b bytes of the video frame H' calculates the number of padding bytes, removes the padding bytes at the end of the frame, and obtains the video frame H.

7. Perform entropy decoding on the above video frame H to obtain matrix C, perform inverse scrambling transformation on matrix C to obtain matrix B, perform inverse quantization on matrix B to obtain matrix A', and perform inverse discrete cosine transform on matrix A', Get the RGB or YUV data that can be used for playback, and directly use it for video playback.

Claims (1)

1, the Code And Decode method in a kind of streaming media video content copyright protection is characterized in that may further comprise the steps:

(1) video content that collects is encoded, obtain dct coefficient matrix A, and matrix A is carried out quantization operation obtain matrix B, matrix B is carried out the scramble conversion obtain Matrix C, Matrix C is carried out entropy coding, generate frame of video H, a plurality of frame of video that successively generate are generated 3GP file T according to the 3GP container standard;

(2) according to the control information among the above-mentioned 3GP file T, calculate the big or small s of each frame of video H respectively, s is carried out mould b computing, b is the size of cryptographic block in the superencipherment deciphering canonical algorithm, if result of calculation w is not 0, then fill b-w+b byte,, then fill b byte if result of calculation w is 0 at the corresponding video End of Frame, byte of padding is initialized as 0, and in the end write down byte number b-w+b or the b that fills in the b byte, and obtaining 3GP file T ', the frame of video behind the byte of padding is H ';

(3) adopt superencipherment deciphering canonical algorithm that frame of video H ' is encrypted, obtain 3GP file T ", frame of video H ' is converted into H after encrypting ";

(4) to above-mentioned 3GP file T " in add cryptographic algorithm title, video content identifier and video content rights issuer address information; the drm content formatted file P that obtains packing; in file P, add custom instruction; with the index box of the transmission means that constitutes explanation media stream in real time transmission format protocol packet, the drm content formatted file P ' of the packing that supported Streaming Media is used;

(5) the drm content formatted file P ' with above-mentioned packing is input to streaming media server, streaming media server takes out relevant field from above-mentioned file P ', obtain the Session Description Protocol file, streaming media server is after receiving user's displaying video content requests, above-mentioned file P ' is packed, obtain meeting first kind of packet of RTP form, simultaneously to above-mentioned Session Description Protocol file packing, obtain meeting second kind of packet of RTP form, and first kind of packet and second kind of packet are sent to the Streaming Media client;

(6) the Streaming Media client unpacks the above-mentioned packet that receives, obtain the frame of video H of above-mentioned encryption " and above-mentioned cryptographic algorithm title; according to the cryptographic algorithm title to H " deciphering, obtain frame of video H ', the byte of padding number that comprises in last b the byte according to frame of video H ', remove the byte of padding of frame of video H ' afterbody, obtain frame of video H;

(7) above-mentioned frame of video H is carried out and the decoding of the corresponding entropy of above-mentioned entropy coding, obtain Matrix C, the Matrix C random conversion that is inverted, obtain matrix B, matrix B is carried out inverse quantization, obtain matrix A ', to matrix A ' carry out inverse discrete cosine transformation, obtain video data, be used for video playback.

Images