JP2777381B2 - Data transmission method - Google Patents

Description

Description: BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a data transmission method which is preferably implemented in, for example, a videophone, and which compresses data to be transmitted to extract and transmit a plurality of types of transmission data. .

2. Description of the Related Art In recent years, the importance of a technique for transmitting image data, such as a facsimile apparatus or a videophone, has been increasing.
However, image data has a high degree of redundancy, and a data compression technique is indispensable for transmission in a short time. Further, there is an increasing demand for confidential communication in connection with the security of transmitted image data.

When using the mean-value-separated normalization vector quantization coding method, which is one of the excellent data compression techniques of vector quantization, image data is first divided into a plurality of sections. The encoded data includes average value data in each section, normalized value data representing the variance of each section, and an identification code for specifying a representative vector described later.
Different types of encoded data are created and transmitted. The representative vector is representative data representing a plurality of data modes in one section, and an identification code is assigned to each representative vector. The correspondence between the identification code and the representative vector is called a code book or the like.

When performing confidential communication, in order to add a security function, for example, the above three types of encoded data are
An encryption process such as data rearrangement called so-called scrambling is performed. The receiving side decrypts the encrypted data and further decrypts the above three types of encoded data. That is, arithmetic processing based on the average value data and the normalized value data is performed on the representative vector corresponding to the identification code, and an image close to the original image is reproduced.

Problems to be Solved by the Invention In the above-mentioned conventional data transmission method, since the encryption processing is performed on all three types of encoded data, the processing time for decrypting the three types of encoded data is long. In other words, there is a problem that it takes a long time to restore and reproduce an image.

SUMMARY OF THE INVENTION Accordingly, an object of the present invention is to provide a data transmission method in which the processing time for encrypting transmission data is reduced and the data transmission speed is improved.

Means for Solving the Problems The present invention compresses data to be transmitted, extracts and transmits a plurality of types of transmission data, and encrypts some types of data of the plurality of types of transmission data. In the method, the data to be transmitted is image data, the image data is divided into a plurality of sections, and a plurality of types of representatives selected with respect to an average value and a variance for each section and a data mode for each section. A data transmission method characterized in that a compression process is performed by selecting identification data respectively attached to data, and the average value is encrypted.

According to the present invention, in a method of compressing data to be transmitted and extracting and transmitting a plurality of types of transmission data,
A part of the plurality of types of transmission data is encrypted. Therefore, compared with the case where the encryption processing is performed on all the transmission data, the transmission data can be encrypted and decrypted in a shorter time.

The data to be transmitted is image data, and the image data is divided into a plurality of sections. As the transmission data, an average value and a variance in each section, and identification data respectively attached to a plurality of types of image data selected for the data mode of each section are selected. A compression process is performed by selecting such transmission data.

According to the present invention, encryption processing is performed on the average value. In such a case, the average value has a low degree of redundancy and a substantial amount of information, so that the effect of the encryption is great and there is no need to encrypt other types of transmission data. Therefore, it is possible to perform encryption and decryption in a short time while maintaining a sufficient security function.

Embodiment FIG. 1 is a block diagram showing a basic configuration of a digital transmission device 1 according to an embodiment of the present invention. The data transmission device 1 is, for example, a videophone that transmits and receives image data via the telephone lines 2 and 22. The data transmission device 1 encodes image data captured by the imaging device 3 and transmits the encoded data to the telephone line 2. And a receiving unit for receiving the transmission data transmitted from the transmission unit, decoding the transmission data, and displaying the decrypted transmission data on the display unit.

In the transmitting means 5, the imaging device 3 is, for example, a CCD (C
harge Coupled Device), for example, 25
Image data having a resolution of 6 × 256 pixels and having 256 levels of gradation at each pixel is output to the encoding circuit 11.

As shown in FIG. 2A, the encoding circuit 11 converts the image data into a plurality of sections By (y = (i−
1) Divide into n + j, i = 1 to m, j = 1 to n).
As shown in FIG. 2 (2), each section By is composed of 4 × 4 pixels bx (x = 1 to 16).

Also, the encoding circuit 11 calculates the average value data AVy and the variance SDy of the gradient of the pixel in each section By,
The identification code idy is selected as described later. That is,
For example, a storage circuit 9 realized by a read-only memory or the like identifies 256 representative data modes (hereinafter, referred to as “representative vectors”) Ch of 8 gradations corresponding to 4 × 4 pixels bx. It is stored in correspondence with the code h individually.

Further, the encoding circuit 11 selects, in the section By, a representative vector that minimizes an error from the representative vector by an operation described later, and outputs an identification code idy corresponding to the representative vector. The identification code idy from such an encoding circuit 11 and the normalized value data SDy
7 is output to the transmission circuit 16. The average value data AVy from the encoding circuit 11 is subjected to, for example, an encryption process called scrambling by the encryption circuit 10 and output to the transmission circuit 16 as average value data _AVpy .

The transmission circuit 16 transmits the average value data AV _py , the normalized value data SDy, and the identification code idy to the telephone line 2 as transmission data. The telephone line 2 is connected to a telephone line 22 via an exchange 12 installed at a telephone office or the like, and transmission data from the transmitting means 5 is received by the receiving circuit 17 of the receiving means 6. The average value data AV _py , the normalized value data SDy and the identification code idy are individually output from the receiving circuit 17, and the average value data AV _py is subjected to a decoding process such as descrambling by the decoding circuit 13. Performed decoding circuit 15
Is given as average value data AVy. In addition, receiving circuit 1
The normalized value data SDy and the identification code idy from 7 are output directly to the decoding circuit 15.

The memory 14 provided in association with the decoding circuit 15 stores the same contents as the memory 9 of the transmission means 5 described above.
The decoding circuit 15 reads a representative vector Cidy from the memory 14 based on the input identification code idy. Image processing based on the normalized value data SDy and the average value data AVy is performed on the representative vector Cidy, and an image close to the original image is reproduced. This image data is, for example,
The image is given to a display unit 7 realized by a CRT display device and displayed.

Hereinafter, the data compression operation in the encoding circuit 11 of the data transmission device 1 will be described. FIGS. 6 (1) to 6 (3) are histograms showing the relationship between the gradient of each pixel in each image example and the number of pixels having the gradient. Each pixel has 256 levels of gradation, and although there are some differences depending on the image, it can be seen that each pixel is distributed over a wide range of gradations.

Such image data is divided by the encoding circuit 11 for each section By, and in each section By, the average value data AVy is obtained based on the following equation.

Here, (bx) y represents the gradient of the pixel bx in the section By, and in this embodiment, K = 16.

FIGS. 7 (1) to 7 (3) show FIGS. 6 (1) to 6
It corresponds individually to Figure (3), and in each section By,
9 is a histogram showing the relationship between the number of pixels and the difference (bx) y−AVy between the gradient (bx) y and the average value data AVy at each pixel. Within each section By, the gradient of each pixel bx (b
x) y is a value close to the average value data AVy in the section By, for example, the average value data of all the pixels in the section By
It can be seen that when reproduced in AVy, a simple image reproduces an image close to the original image.

The encoding circuit 11 further calculates each section By based on the second equation.
Is obtained as normalized value data SDy which is the variance in.

FIGS. 8 (1) to 8 (3) show FIGS. 7 (1) to 7
The horizontal axis corresponds to the difference (bx) y-A between the gradation (bx) y of each pixel and the average value data AVy.
Vy is divided by the normalized value data SDy. Thus, the normalized value data SD representing the variation within each section By
By adding y, an image closer to the original image can be reproduced.

Next, the vector quantization processing in the encoding circuit 11 will be described. The memory 9 stores a code book, that is, a representative vector Ch corresponding to the identification code h. The representative vector Ch is 16 pixels bx in each section By.
And comprises, for example, 3-bit data cx. The encoding circuit 11 obtains an error (dx) y represented by the following equation based on the average value data AVy obtained by the above arithmetic processing and the normalized value data SDy.

Further, with respect to the representative vector Ch corresponding to each identification code h, the sum Δhy of each pixel having an error represented by the equation (4) is calculated.

Here, (cx) h is data cx of the representative vector Ch corresponding to the identification code h.

The encoding circuit 11 transmits the identification code idy corresponding to the representative vector Cidy having the smallest sum Δhy of the errors to the transmitting circuit 16.
Will be output.

Thus, in the encoding circuit 11, the imaging device 3
Are compressed and encoded into three types of transmission data: average value data AVy, normalized value data SDy, and identification code idy.

Next, the encryption processing in the encryption circuit 10 will be described. In the encryption circuit 10, the average value data AVy is replaced by a replacement table P shown in the following equation, and so-called scrambling is performed.

P1, P2,..., Py,... Are, for example, a predetermined random number sequence, and the encryption circuit 10 converts the average value data AVy into the average value data AV.
Replace with _py . As a result, the average value data AVy is rearranged and encrypted.

FIG. 4 is a flowchart for explaining the operation of the data transmission device 1 in data transmission. With reference to FIG. 4, when image data is input from the imaging device 3, the process proceeds from step n1 to step n2, and the image data input by the encoding circuit 11 is divided into sections By.

In step n3, in each section By, the average value data AVy of the gradient (bx) y of the pixel bx and the normalized value data SDy in the section By are calculated.

In step n4, the identification code idy is selected corresponding to each section By by referring to the code book of the memory 9.

At step n5, the average value data is
The scrambling of AVy, that is, the rearrangement of data for encryption is started, and the normalized value data SDy is transmitted in step n6. When the transmission of the normalized value data SDy is completed, the identification code idy is transmitted in step n7.

The time chart of the processing operation in steps n5 to n7 is shown in FIG. The normalized value data SD is transmitted in a period W1 as shown in FIG. 3 (1). Further, the identification code idy is transmitted in a period W2 subsequent to the period W1, as shown in FIG. 3 (2). this period
In W1 and W2, as shown in FIG. 3 (4), encryption processing of the average value data AVy is performed in parallel. The encrypted average value data _AVpy is transmitted in the period W3 following the period W2 in step n8 as shown in FIG. 3 (3).

By such a processing operation, for example, FIG.
Is transmitted, for example, when a third party receives transmission data using the receiving means 21 via the line 20 in FIG. 1, the average value data AVy is encrypted. Therefore, an image is reproduced as shown in FIG. 5 (2). According to the experiment of the present inventor, in the case of image data, it is found that the redundancy of the average value data AVy is low and the substantial information amount is large. Therefore, as can be seen from FIG. 5 (2), if the average value data AVy is subjected to encryption processing, sufficient security can be provided.

FIG. 9 is a flowchart for explaining the receiving operation of the transmission data and the image display operation of the receiving means 6.
When the receiving means 6 receives the transmission data from the transmission means 5, the process moves from step m1 to step m2, and the descrambling of the average value data _AVpy of the transmission data is started. In step m3, the corresponding representative vector Cidy is read from the memory 14 based on the identification code idy. In step m4, the average value data AVy and the representative vector C
Based on idy and the normalized value data SDy, the gradient (ex) y of each pixel bx is calculated by the following equation.

(Ex) y = {(cx) idy + AVy} × SDy (6) At step m5, an image is displayed on the display unit 7 based on the gradation (ex) y, and the processing operation is terminated.

As described above, in the present embodiment, since the encryption processing is performed only on the average value data AVy of the transmission data, for example, the normalized value data SDy and the identification code i
The encryption processing of the average value data AVy can be performed in parallel with the transmission of the dy, and substantially no time is required only for the encryption processing. Therefore, the image data to be transmitted can be transmitted after being subjected to encryption processing for security protection without increasing the transmission time.

In the present embodiment, the representative vector Ch corresponding to each identification code h is fixed in the memory 9. However, the representative vector Ch corresponding to each image may be formed and transmitted so as to create an optimal representative vector. it can. Also, as the encryption, a case has been described in which scrambling based on the replacement table P is performed. However, a plurality of random number sequences are prepared, and an identification code for selecting one random number sequence from the random number sequence is transmitted. The scrambling may be performed using a random number sequence corresponding to the identification code as a replacement table. Also, the encryption method need not be limited to scrambling, and the encryption processing may be performed by another method such as a method of adding random number data to average value data.

Effect of the Invention As described above, according to the present invention, in a method of compressing data to be transmitted, extracting a plurality of types of transmission data, and transmitting the data, the security processing is performed without increasing the transmission time. And the convenience is greatly improved.

In particular, according to the present invention, since the average value has low redundancy and a substantial amount of information, it is not necessary to perform encryption processing on this average value and to encrypt other types of transmission data. Thus, it is possible to perform encryption and decryption in a short time while maintaining a sufficient airtight protection function.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a configuration of a data transmission apparatus 1 according to one embodiment of the present invention, FIG. 2 is a view showing a division mode of a plurality of sections By of image data, and FIG. FIG. 4 is a flow chart for explaining the data transmission operation of the data transmission apparatus 1, and FIG. 5 is a view showing an example of an image before transmission and an image subjected to encryption processing. FIG. 6 is a histogram showing the relationship between the gradient of each pixel in each image and the number of pixels.
The figure shows the average value data AVy of the gradient of each pixel in each image.
Histogram showing the relationship between the difference from
FIG. 9 is a histogram showing the relationship between the number of pixels and the average value separation normalized gray level of each pixel in each image. FIG. 9 is a flowchart for explaining the operation of the data transmission apparatus 1 for reproducing an image. is there. 1 Data transmission device 2,22 Telephone line 9,14 Memory, 10 Encryption circuit 11, Encoding circuit 13, Decoding circuit 15, Decoding circuit, AVy …… Average value data, SDy
…… Normal value data, idy …… Identification code

Claims (1)

(57) [Claims] 1. A data transmission method for compressing data to be transmitted, extracting and transmitting a plurality of types of transmission data, and encrypting a part of the plurality of types of transmission data. The data to be image data is image data, which is divided into a plurality of sections, and is attached to a plurality of types of representative data selected with respect to the average value and variance of each section and the data mode of each section. A data transmission method, wherein a compression process is performed by selecting identification data, and the average value is encrypted.