JP5141334B2 - Picture encryption / decryption device - Google Patents

Description

  The present invention relates to a picture encryption / decryption device, and more particularly to a video signal encryption and decryption device based on the MPEG system.

  In a video encoding apparatus based on the MPEG system, an encoded bit stream may be encrypted for the purpose of preventing outflow of raw digital video data. However, in the logical operation (such as XOR operation) of “key data” and actual data, which is the most common technique used for digital data encryption, if the “key data” itself is stolen, it is easy. The encryption method (algorithm) may be decrypted.

Therefore, there is a need for an encryption technique that cannot be easily deciphered even if the “key data” itself is stolen.
9 and 10 are diagrams for explaining a conventional video encoding / decoding device based on the MPEG system. FIG. 10A shows a block configuration diagram of a general video encoding apparatus based on the MPEG system.

  The input video signal (original image) is subjected to DCT conversion processing by a DCT (Discrete Cosine Transform) unit 10, quantized by a quantization unit 11, and then encoded by a variable length coding unit 12. Is output as a video stream. The quantized signal is subjected to inverse quantization processing by the inverse quantization unit 13 and inverse DCT conversion processing by the inverse DCT unit 14, and is temporarily stored in the frame memory 15 as a decoded image. Further, the motion prediction unit 16 obtains a difference between the reconstructed image moved by the motion vector obtained by calculation and the original image with respect to the stored decoded image (motion prediction process), and the difference signal is obtained. In some cases, DCT conversion processing, quantization processing, and encoding processing are performed.

FIG. 10B shows a block configuration diagram of a general video decoding apparatus based on the MPEG system.
The input video stream is converted back to a DCT coefficient signal by the variable length decoding unit 17, subjected to inverse quantization processing by the inverse quantization unit 18, and subjected to inverse DCT conversion processing by the inverse DCT unit 19, thereby obtaining a video signal. Restored. Further, the decoded image after the inverse DCT conversion process is temporarily stored in the frame memory 20, and the motion compensation unit 21 moves the decoded image by the motion vector and reconstructs the addition result of the decoded image. In some cases, it is output as a video signal (motion compensation processing).

  As shown in FIG. 9, in a video encoding apparatus based on the MPEG system, a video signal is divided into blocks of 8 × 8 (= 64 pixels) units, and a high-frequency region component of a video signal is obtained by DCT conversion processing and quantization processing. The amount of information is compressed by deleting. In the variable length coding unit that assigns zero-run length codes to DCT coefficients, the DCT coefficients are read in zigzag order (zigzag scanning) from the DC component position (1, 1) to the AC high frequency component position (8, 8). An EOB (end-of-block) code is assigned when there is no non-zero coefficient in the high frequency region.

  The subject of this apparatus is providing the picture encryption / decryption apparatus which employ | adopted the encryption system in which picture information is not easily deciphered.

  This picture encryption / decryption apparatus encodes or decodes an image signal in units of blocks, and performs picture encryption / decryption in an image encoding / decoding system that performs image compression / decompression in units of macroblocks including a plurality of the blocks. In the encoding device, a numerical value is calculated from the result of calculating the classification number obtained from the type of the macroblock included in the previous image and the address information of the block at the end position of the image signal encoding of the previous image Computing means for computing the exclusive OR means for computing the exclusive OR of the computed numerical value and the image signal currently being processed, and the image obtained by encrypting or decrypting the result of the exclusive OR operation Output means for outputting as a signal.

  According to the present apparatus, it is possible to provide a picture encryption / decryption apparatus that employs an encryption method in which picture information is not easily decrypted.

  In the present embodiment, in addition to the fixed “key data” of the conventional method, random information generated at the time of bitstream generation (encoding) is used to improve the encryption strength. Further, as the random information, EOB (end of block) address information at the time of encoding the DCT coefficient in the block is used. Furthermore, the result of four arithmetic operations / logical operations of “key data” and EOB address information is used as “new key data”. Also, encryption is realized by a logical operation (XOR operation) between the “new key data” and the DC component of the DCT coefficient, and a logical operation (XOR operation) is not performed on the AC component of the DCT coefficient. . The EOB address information of the Nth block is used for encryption of the (N + 1) th block. Here, what is to be encrypted is an image, the encrypted image is variable-length encoded and transmitted, and after being subjected to variable-length decoding on the receiving side, the encrypted image is encrypted using the variable-length encoding parameters. In this configuration, an image is decoded.

FIG. 1A shows a block configuration diagram of a video encoding apparatus according to the first embodiment.
In FIG. 1A, an EOB address holding unit 30 that extracts and holds EOB address information from the variable length coding unit 12, an OFFSET register unit 31 that holds key data set from the outside, and an EOB address holding unit 30. EOB address information obtained, EOB address processing unit 32 for performing specific arithmetic processing on the OFFSET register value obtained from OFFSET register unit 31, DCT coefficient after quantization and EOB address information obtained from EOB address processing unit 32 And a SEL (selector) unit 34 for selectively outputting the DCT coefficient after quantization processing and the DCT coefficient after XOR operation obtained from the XOR unit 33.

  First, key data is set in the OFFSET register unit 31 from the outside of the apparatus. When encoding is advanced and EOB address information of the Nth block is obtained, EOB address information is held. When encoding progresses and the encoding of the (N + 1) th block is reached, the OFFSET register value and the EOB address value are subjected to a specific calculation process (for example, addition process), and the value after the calculation and the quantization process XOR operation with the later DCT coefficient. If the current position of the DCT coefficient is a direct current component, the DCT coefficient after the XOR operation is output to the variable length coding unit 12. If the current position of the DCT coefficient is an AC component, the DCT coefficient after the quantization process is output to the variable length coding unit 12. The above processing is continued until encoding of all the blocks constituting the picture screen is completed.

FIG. 1B shows a block configuration diagram of the video decoding apparatus according to the first embodiment.
In FIG. 1B, from the EOB address holding unit 40 that extracts and holds EOB address information from the variable length decoding unit 17, the OFFSET register unit 41 that holds externally set key data, and the EOB address holding unit 40. The obtained EOB address information, the EOB address processing unit 42 for performing a specific arithmetic processing on the OFFSET register value obtained from the OFFSET register unit 41, the DCT coefficient after the variable length decoding process, and the EOB obtained from the EOB address processing unit 42 An XOR unit 43 that performs an XOR operation with the address information and a SEL unit 44 that selectively outputs the DCT coefficient after the variable length decoding process and the DCT coefficient after the XOR operation obtained from the XOR unit 43 are provided.

  First, the same key data as that of the transmitting device is set in the OFFSET register unit 41. When decoding progresses and the EOB address information of the Nth block is obtained, the EOB address information is held. When decoding progresses and the decoding of the (N + 1) th block is reached, the OFFSET register value and the EOB address value are subjected to a specific calculation process (for example, addition process), and the calculated value and variable-length decoding are performed. XOR operation with the DCT coefficient after conversion. If the current position of the DCT coefficient is a direct current component, the DCT coefficient after the XOR operation is output to the inverse quantization unit 18. If the current position of the DCT coefficient is an AC component, the DCT coefficient after the variable length decoding process is output to the inverse quantization unit 18.

  In the above, when decoding a picture encrypted by XOR in the video encoding apparatus, the video decoding apparatus is also XORed because it returns to the original state when the signal is XORed twice in the same way. Because there is. Therefore, the encrypted information can be decrypted by XORing the information encrypted by XOR on the transmission side in exactly the same way on the reception side.

  When the encryption method based on the first embodiment is used, random EOB address information generated at the time of encoding a bit stream is used as an element of “key data”. Even if it is stolen, it can realize strong video stream encryption that cannot be easily deciphered.

  In the first embodiment, since the fixed “key data” itself needs to be supplied from the outside of the apparatus, there is a risk that the “key data” information is leaked to the outside in the process of transmitting / receiving the “key data”. Remaining. In the second embodiment, the fixed “key data” in the first embodiment is not supplied from the outside of the apparatus, but the picture coding information generated internally at the time of bitstream generation (coding) To improve the encryption strength. As picture coding information used above, a distribution state such as a prediction type of MB (macroblock) for each picture is used. A micro block is a block of a larger unit including a plurality of blocks which are encoding units and is a unit of image compression processing. Also, the MB prediction type for each picture is extracted for the designated MB column number, and the result of these four arithmetic operations (for example, addition) is defined as picture coding information. Further, the “designated MB column number” and “type of four arithmetic operations” are selected using the picture coding information of the previous picture.

  When the encryption method according to the second embodiment is used, there is no need to supply the fixed “key data” in the first embodiment from the outside of the apparatus, and as a result, leakage of the “key data” information to the outside. Therefore, it is possible to realize a strong video stream encryption that cannot be easily deciphered.

  In the second embodiment, the MB prediction type for each picture is extracted for the designated slice column number, and the result of these four arithmetic operations (for example, addition) is defined as picture coding information. Also, the “designated slice column number” and “type of arithmetic operation” are selected using the picture coding information of the previous picture. Alternatively, the MB prediction type for each picture is extracted diagonally from the designated MB number, and the result of these four arithmetic operations (for example, addition) is defined as picture coding information. At this time, the “designated MB number” and “type of four arithmetic operations” are selected using the picture coding information of the previous picture. Further, the extraction pattern of the picture coding information is used by switching on a picture basis. The “picture coding information extraction pattern” is selected using the picture coding information of the previous picture.

FIG. 2A shows a block configuration diagram of a video encoding apparatus according to the second embodiment.
In the second embodiment, in addition to the block configuration of the encoding apparatus of the first embodiment, an MB type holding unit 50 that extracts the MB type from the motion prediction unit 16 and converts the extracted MB type into a classification number, The classification numbers held in the MB type holding unit 50 are aggregated according to a specific rule when the picture encoding process is completed, and processed according to a specific calculation method. And a picture information processing (calculation) unit 51 that stores (updates) in the OFFSET register unit 31 at the first timing of the next picture encoding process. For example, conversion from the MB type to the classification number is performed according to the example in Table 1 or Table 2.

  Table 1 is a table for converting the MB type related to the frame structure into the classification number. The MB type includes skip MB, intra MB, frame prediction, field prediction, and DualPrime prediction. Table 2 is a table for converting the MB type related to the field structure into the classification number. Examples of MB types include skip MB, intra MB, field prediction, 16 × 8 prediction, and DualPrime prediction. These MB types are defined in MPEG, and classification numbers should be appropriately set by those using this embodiment.

FIG. 2B shows a block configuration diagram of the video decoding apparatus according to the second embodiment.
* Since the configuration is almost the same just by replacing the “motion prediction unit” of the above coding device with the “variable length decoding unit” of the decoding device, the explanation will be omitted.

In the second embodiment, in addition to the block configuration of the decoding apparatus of the first embodiment, an MB type holding unit that extracts MB types from the variable length decoding unit 17 and converts them into classification numbers and holds them. 52. When the variable number decoding process of the picture is completed, the classification numbers held in the MB type holding unit 52 are aggregated according to a certain specific rule, processed according to a certain specific calculation method, and then the calculation is performed. A picture information processing (calculation) unit 53 that holds the result as picture decoding information and writes (updates) to the OFFSET register unit 41 at the first timing of the next picture decoding process is provided. As for the conversion method from the MB type to the classification number, the conversion is performed according to the example in Table 1 or Table 2. However, the conversion method from the MB type to the classification number is the same for the picture encoding device on the transmission side and the picture decoding device on the reception side.

3 to 8 are explanatory diagrams of how to perform encryption in the second embodiment.
As a specific case, assuming that a picture encoding process is performed in a frame structure, conversion from the MB type to the classification number is performed according to Table 1. As an example, FIG. 3 shows the result of mapping the distribution status of classification numbers at the time when encoding processing for one picture is completed for one picture. In FIG. 3, the vertical direction is a slice column, and the horizontal direction is an MB column.

On the other hand, the following three patterns can be considered as a method of counting “classification numbers” according to a specific rule.
(1) Totally classify all classification numbers in the vertical direction (slice column direction) in a specific MB column (see FIG. 4).
(2) Totally classify all classification numbers in the horizontal direction (MB column direction) in a specific slice column (see FIG. 5).
(3) Totally classify all classification numbers in an oblique direction starting from a specific MB number on the first slice column (see FIG. 6).

The following three types are given as examples of processing according to a specific calculation method.
(A) Addition type picture coding information =
(1st classification No.) + (2nd classification No.) + (3rd classification No.) +… + (N-1st classification No.) + (Nth classification No.)
(B) Addition / subtraction type picture coding information =
(1st classification No.)-(2nd classification No.) + (3rd classification No.)-... + (N-1st classification No.)-(Nth classification No.)
(C) Addition / subtraction multiplication type picture coding information =
(1st classification No.) x (2nd classification No.) + (3rd classification No.)-...-(N-1st classification No.) x (Nth classification No.)
As an example, when selecting aggregation method = (1) and calculation method = (a), from FIG.
Picture coding information = 2 + 0 + 1 + 2 + 0 + 2 + 2 + 1 + 3 = 13
The operation result is obtained.

  Thereafter, the value 13 is written in the OFFSET register unit 31 at the timing of starting the encoding process of the next picture, and is also fed back to the MB type holding unit 50 to extract the MB column number and slice column to be extracted. It is used as a number or calculation type selection signal. The above is the description of the encoding process. However, as described above, the decoding process may be performed in exactly the same way as the encoding process, and thus the above description also applies to the decoding process.

  As shown in FIG. 7, the present embodiment is an algorithm in which the picture encoding information of the previous picture is always used in the encryption process of the next picture, and therefore OFFSET is only used during the encryption process of the first picture. A set value for the register unit 31 cannot be prepared. Therefore, for the first picture only, for example, a fixed value of 0x0000 is set as an initial value in the OFFSET register unit 31 or used as a selection signal for an MB column number or slice column number to be extracted or an operation type. be able to. FIG. 7 shows an example in which the first I picture fixes the set value of the OFFSET register 31 to 0x0000, and the subsequent I, B, and P pictures use the picture information of the previous picture. Here, the I picture, the B picture, and the P picture are the types of pictures in MPEG. The I picture is a picture that does not refer to other pictures at the time of compression, and the P picture is the previous picture. The B picture is a picture that refers to the previous and next pictures.

  Also, as shown in FIG. 8, in the event that a stream error is mixed during stream transmission, the initial value is set when encoding an I picture in order to suppress the adverse effect of error propagation on the decoding side as much as possible. As always, 0x0000 may be used. In FIG. 8, the setting value of the OFFSET register unit 31 at the time of encoding for I picture is fixed to 0x0000, and in the case of B picture and P picture, encoding is performed using the picture information of the previous picture. Is shown.

  As described above, in the first and second embodiments, the encoding apparatus on the transmission side and the decoding apparatus on the reception side perform the encryption process performed by the encoding apparatus and the decoding apparatus. The decryption process is exactly the same. Therefore, the description regarding the encoding process is applied to the decoding process in exactly the same manner.

In addition to the above embodiment, the following supplementary notes are disclosed.
(Appendix 1)
In a picture encoding / decoding device in an image encoding / decoding system that encodes or decodes an image signal in units of blocks and performs image compression / decompression in units of macroblocks including a plurality of the blocks,
An operation for calculating a numerical value from the result of calculating the classification number obtained from the type of macroblock included in the previous image and the address information of the block at the end position of the encoding of the image signal of the previous image Means,
Exclusive OR means for calculating exclusive OR of the calculated numerical value and the image signal currently being processed;
Output means for outputting the result of the exclusive OR operation as an encrypted or decrypted image signal;
A picture encryption / decryption device comprising:
(Appendix 2)
2. The picture encryption / decryption device according to appendix 1, wherein the image encoding / decoding system is MPEG.
(Appendix 3)
The picture encryption / decryption device according to attachment 2, wherein the end position of the encoding of the image signal is an end of block position.
(Appendix 4)
The picture encryption / decryption device according to appendix 2, wherein when an I picture is encrypted or decrypted, a fixed value is used instead of the result of calculating the classification number. (Appendix 5)
The picture encryption / decryption device according to appendix 1, wherein a fixed value is used for the first picture of the image signal instead of the result of calculating the classification number.
(Appendix 6)
2. The picture encryption / decryption device according to appendix 1, wherein the calculation means holds a table for storing a macroblock type and a classification number in association with each other.
(Appendix 7)
2. The picture encryption / decryption device according to appendix 1, wherein the separation number is calculated by combining classification numbers obtained for a plurality of macroblocks by a combination of four arithmetic operations.
(Appendix 8)
The plurality of macroblocks are macroblocks belonging to a specific macroblock sequence when the macroblocks are two-dimensionally arranged in a slice example direction and a macroblock sequence direction. Picture encryption / decryption device.
(Appendix 9)
The picture according to appendix 7, wherein the plurality of macroblocks are macroblocks belonging to a specific slice column when the macroblocks are two-dimensionally arranged in a slice example direction and a macroblock column direction. Encryption / decryption device.
(Appendix 10)
The plurality of macroblocks are macroblocks acquired obliquely from one macroblock belonging to the first slice row when the macroblocks are two-dimensionally arranged in the slice example direction and the macroblock row direction. Item 8. The picture encryption / decryption device according to appendix 7, wherein
(Appendix 11)
In a picture encoding / decoding method in an image encoding / decoding system that encodes or decodes an image signal in units of blocks and performs image compression / decompression in units of macroblocks including a plurality of the blocks,
A numerical value is calculated from the result of calculating the classification number obtained from the type of macroblock included in the previous image and the address information of the block at the end position of the image signal encoding of the previous image,
Calculate the exclusive OR of the calculated numerical value and the image signal currently being processed,
Output the result of exclusive OR operation as an encrypted or decrypted image signal,
A picture encryption / decryption method characterized by the above.

It is a block block diagram of the video coding apparatus in 1st Embodiment. It is a block block diagram of the video decoding apparatus in 1st Embodiment. It is a block block diagram of the video coding apparatus in 2nd Embodiment. It is a block block diagram of the video decoding apparatus in 2nd Embodiment. It is explanatory drawing (the 1) of how to perform the encryption in 2nd Embodiment. It is explanatory drawing (the 2) of how to perform the encryption in 2nd Embodiment. It is explanatory drawing (the 3) of how to perform the encryption in 2nd Embodiment. It is explanatory drawing (the 4) of how to perform the encryption in 2nd Embodiment. It is explanatory drawing (the 5) of how to perform the encryption in 2nd Embodiment. It is explanatory drawing (the 6) of how to perform the encryption in 2nd Embodiment. It is FIG. (1) explaining the video encoding / decoding apparatus based on the conventional MPEG system. It is FIG. (2) explaining the video encoding / decoding apparatus based on the conventional MPEG system.

Explanation of symbols

DESCRIPTION OF SYMBOLS 10 DCT part 11 Quantization part 12 Variable length encoding part 13, 18 Inverse quantization part 14, 19 Inverse DCT part 15, 20 Frame memory 16 Motion prediction part 17 Variable length decoding part 21 Motion compensation part 30, 40 EOB address Holding unit 31, 41 OFFSET register 32, 42 EOB address processing (calculation) unit 33, 43 XOR unit 34, 44 SEL unit (selector unit)
50, 52 MB type holding unit 51, 53 Picture information processing (calculation) unit

Claims (5)

In a picture encoding / decoding device in an image encoding / decoding system that encodes or decodes an image signal in units of blocks and performs image compression / decompression in units of macroblocks including a plurality of the blocks,
The result of calculating the classification number obtained from the macroblock included in the picture immediately before the picture including the macroblock currently being processed, and the encoding of the block immediately before the block currently being processed Calculating means for calculating a numerical value from address information at a position where there is no non-zero coefficient after that ,
Exclusive OR means for calculating exclusive OR of the calculated numerical value and the image signal currently being processed;
Output means for outputting the result of the exclusive OR operation as an encrypted or decrypted image signal;
A picture encryption / decryption device comprising:   2. The image encoding / decoding system according to claim 1, wherein the image encoding / decoding system is MPEG, and a fixed value is used instead of the result of calculating the classification number when encrypting or decoding an I picture. The picture encryption / decryption device described.   The picture encryption / decryption device according to claim 1, wherein a fixed value is used for the first picture of the image signal instead of the result of calculating the classification number. Calculation of the classification number, picture encryption / decryption device according to classification number obtained for a plurality of macro-blocks in claim 1, characterized in that for calculating the combination of the four arithmetic operations. In a picture encoding / decoding method in an image encoding / decoding system that encodes or decodes an image signal in units of blocks and performs image compression / decompression in units of macroblocks including a plurality of the blocks,
The result of calculating the classification number obtained from the type of the macroblock included in the picture immediately before the picture containing the macroblock currently being processed, and the block immediately before the block currently being processed A numerical value is calculated from the DCT coefficient in encoding and address information at a position where no non-zero coefficient exists thereafter ,
Calculate the exclusive OR of the calculated numerical value and the image signal currently being processed,
Output the result of exclusive OR operation as an encrypted or decrypted image signal,
A picture encryption / decryption method characterized by the above.