KR20000011332A - Digital image coding/decoding apparatus and method using watermarking - Google Patents

Abstract

PURPOSE: A digital image coding device is provided to calibrate the distorted host image transmitted from severe noise environment using present screen. CONSTITUTION: A digital image coder is composed of: discrete wavelitt transformer for outputting M times M wavelitt coefficient if a specific positive coefficient is M by discrete wavelitt transforming the host image to be transmitted; a main coefficient extracting unit for extracting the coefficient of a specific ratio which is more important among the M times M wavelitt coefficient; a random number generating unit for generating a random number according to the rule corresponded to a code; a coefficient replacement/association unit for replacing and associating the coefficients of replacement location of N times N expressed in the replacement location information among the M times M wavelitt coefficients into N times N number of main wavelitt coefficients.

Description

Digital image coding apparatus and decoding apparatus using watermarking and its method {Digital image coding / decoding apparatus and method using watermarking}

Watermarking is a technology for transmitting an encrypted image along with the image to be transmitted for the purpose of security and right indication of the image. The receiver side receives an image to be transmitted (hereinafter referred to as a host image) together with an encrypted image (hereinafter referred to as a signature image) secretly transmitted, and decodes the signature image by a decryption apparatus.

FIG. 1 is a block diagram illustrating a concept of a digital image codec device that generates a watermarked image and extracts a signature image from the watermarked image.

In the conventional digital video codec device, at the time of encoding, a discrete cosine transform (hereinafter referred to as DCT) for a host video and a signature video to be secretly carried on the host video are respectively performed. Extract DCT coefficients. The DCT coefficients are encoded at the encoder. At this time, the DCT coefficient component of the signature image is encrypted by a separate encryption encoder that performs a control operation for encryption. As a result, the DCT coefficient of the host image and the DCT coefficient of the signature image encoded and inserted into the DCT coefficient of the host image can be transmitted. The DCT coefficient of the host image and the signature image encoded and inserted into the DCT coefficient of the host image can be transmitted. Inverse Discrete Cosine Transformation (IDCT) of the DCT coefficient as it is, a watermarked image showing only the host image without the signature image is obtained.

On the other hand, the signature image which is watermarked in the host image is reconstructed by decoding with a special decoding device. In the reconstruction process, DCT coefficients are extracted by discrete cosine transforming the watermarked images. Next, DCT coefficients extracted by discrete cosine transforming the original host image from the DCT coefficients of the watermarked image are subtracted, and the DCT coefficient component of the encrypted signature image is extracted. The decoder decodes the DCT coefficient component. At this time, the cipher decoder performs a control operation for decryption. The DCT coefficients of the signature image decoded by the decoding are inverse discrete cosine transform to reconstruct the signature image.

In the digital image codec device using the watermarking, if the current host image is severely distorted, the decoder corrects the current screen using the previous host image. However, there is a problem in that information flow is not continuous when the previous screen is used without using the current screen for screen correction. In addition, there is a problem that a screen correction error may occur when the scene changes abruptly.

An object of the present invention is to provide a digital image encoding apparatus that encodes a current image to correct a distorted host image transmitted in a severe noisy environment.

Another object of the present invention is to provide a digital image encoding apparatus for decoding an image encoded by the encoding apparatus.

Another technical problem to be achieved by the present invention is to provide a digital image codec device for encoding and decoding, respectively, to correct a distorted host image transmitted in a severe noisy environment using a current screen.

Another object of the present invention is to provide an encoding method and a decoding method performed in the encoding device and the decoding device.

FIG. 1 is a block diagram illustrating a concept of a digital image codec device that generates a watermarked image and extracts a signature image from the watermarked image.

2 is a block diagram showing the structure of a digital video codec device according to an embodiment of the present invention.

3 is a flowchart illustrating main steps of a digital image coding method according to an embodiment of the present invention.

4 is a flowchart illustrating main steps of a digital image decoding method according to an embodiment of the present invention.

In order to achieve the above object, the digital image coding apparatus according to the present invention comprises a discrete wavelet transform unit for outputting M × M wavelet coefficients when a predetermined positive integer is M by discrete wavelet transforming the host image to be transmitted; An important coefficient extracting unit configured to extract a predetermined ratio coefficient having a high importance among the M × M wavelet coefficients; A random number generation unit for generating a random number according to a rule corresponding to an encryption code; And using the random number, when the predetermined positive integer smaller than M is N, the replacement position information indicating the replacement position of N × N points is obtained, and the replacement position information is shown in the replacement position information among the M × M wavelet coefficients. And a coefficient replacement / combination unit for replacing and combining the coefficients on the N × N replaceable positions with N × N important wavelet coefficients, respectively.

The coding apparatus may further include an inverse discrete wavelet transform unit configured to generate a synthesized host image by inverse discrete wavelet transforming the replaced M × M wavelet coefficients; And a compression unit configured to output the bitstream by compressing the synthesized host image and the encryption code.

In addition, it is preferable that said N is 0.5 M or less.

In addition, in order to achieve the above object, the digital image coding apparatus according to the present invention comprises: an decompression unit for decompressing the bitstream to restore a host image and a predetermined encryption code; An image quality evaluator for evaluating the image quality of the reconstructed host image and outputting a control signal that is in a first logic state if the reconstructed host image is less than a reference quality and otherwise becomes a second logic state; A discrete wavelet transform unit for performing discrete wavelet transform of the reconstructed host image in response to a control signal of a first logic state and outputting an M × M wavelet coefficient when a predetermined positive integer is M; A random number generator for generating a random number according to a rule corresponding to the restored encryption code; When the predetermined positive integer smaller than M is N using the random number, separation position information indicating a replacement position of N × N points is obtained, and N indicated in the separation position information among the M × M wavelet coefficients is obtained. A coefficient separator for separating the coefficients on the substitute positions at the xN position; An inverse discrete wavelet transform unit which outputs a signature image by inverse discrete wavelet transforming the M × M wavelet coefficients; An image scale converter configured to enlarge the scale of the reconstructed signature image to the scale of the host image; And an image selector configured to select the signature image whose scale is enlarged in response to the control signal of the first logic state and to select and output the reconstructed host image in response to the control signal of the second logic state. do.

In addition, in order to achieve the above object, the digital image codec device according to the present invention discrete-wavelet transform the host image to be transmitted by the discrete wavelet transform unit for outputting the M × M wavelet coefficient when a predetermined positive integer M ; An important coefficient extracting unit configured to extract a predetermined ratio coefficient having a high importance among the M × M wavelet coefficients; A random number generation unit for generating a random number according to a rule corresponding to an encryption code; When the predetermined positive integer smaller than M is N using the random number, substitution position information indicating the substitution position of N × N points is obtained, and N indicated in the substitution position information among the M × M wavelet coefficients is obtained. A coefficient replacement / combination unit for replacing and combining the coefficients on the x-N places of substitution positions with N x N important wavelet coefficients, respectively; An inverse discrete wavelet transform unit for generating a synthesized host image by inverse discrete wavelet transforming the replaced M × M wavelet coefficients; A compression unit for compressing the synthesized host image and the encryption code to output a bitstream; A decompression unit for decompressing the compressed bitstream to restore a host image and a predetermined encryption code; An image quality evaluator for evaluating the image quality of the reconstructed host image and outputting a control signal that is in a first logic state if the reconstructed host image is less than a reference quality and otherwise becomes a second logic state; A discrete wavelet transform unit for performing discrete wavelet transform of the reconstructed host image in response to a control signal of a first logic state and outputting an M × M wavelet coefficient when a predetermined positive integer is M; A random number generator for generating a random number according to a rule corresponding to the restored encryption code; When the predetermined positive integer smaller than M is N using the random number, separation position information indicating a replacement position of N × N points is obtained, and N indicated in the separation position information among the M × M wavelet coefficients is obtained. A coefficient separator for separating the coefficients on the substitute positions at the xN position; An inverse discrete wavelet transform unit which outputs a signature image by inverse discrete wavelet transforming the M × M wavelet coefficients; An image scale converter configured to enlarge the scale of the reconstructed signature image to the scale of the host image; And an image selector configured to select the signature image whose scale is enlarged in response to the control signal of the first logic state and to select and output the reconstructed host image in response to the control signal of the second logic state. do.

In addition, to achieve the above object, the digital image coding method according to the present invention comprises the steps of (a) obtaining M × M wavelet coefficients when a predetermined positive integer is M by discrete wavelet transforming the host image to be transmitted; (b) extracting a predetermined ratio coefficient having a high importance among the M × M wavelet coefficients; (c) generating a random number according to a rule corresponding to an encryption code; (d) using the random number to obtain replacement location information indicating replacement locations of N × N locations when a predetermined positive integer smaller than M is N; And (e) replacing and combining the coefficients on the N × N locations indicated in the replacement position information among the M × M wavelet coefficients into N × N significant wavelet coefficients, respectively. do.

In addition, to achieve the above object, the digital image decoding method according to the present invention comprises the steps of (a) evaluating the quality of the synthesized host image to determine whether the quality is above a predetermined criterion; (b) setting the synthesized host image as the host image when it is determined in step (a) that the quality of the host image is greater than or equal to a predetermined criterion; And (c) setting the signature image secretly stored in the synthesized host image as the host image when it is determined that the quality of the host image is less than a predetermined criterion in step (a).

Hereinafter, exemplary embodiments of a digital image coding apparatus, a decoding apparatus, a codec apparatus, and methods thereof according to the present invention will be described with reference to the accompanying drawings.

2 is a block diagram illustrating a structure of a digital video codec device according to an exemplary embodiment of the present invention. 3 and 4 illustrate, as flowcharts, main steps of the digital image coding method and the digital image decoding method according to the embodiment of the present invention, which are implemented in the digital image codec device. 3 and 4 are often referenced below.

Referring to FIG. 2, the digital image codec apparatus according to the present invention includes an encoder 20 and a decoder 22. The encoding unit 20 includes a discrete wavelet transform unit 202, an important coefficient extracting unit 204, a random number generation unit 206, a coefficient substitution / combining unit 208, an inverse discrete wavelet transform unit 210, and The compression unit 212 is provided. In addition, the decoding unit 22 includes an extension unit 222, an image quality evaluation unit 224, a discrete wavelet transform unit 226, a random number generation unit 228, a coefficient separation unit 230, and an inverse discrete wavelet transform. A unit 232, an image scale converter 234, and an image selector 236 are provided.

Referring to the operation of the encoding unit 20, the discrete wavelet transform unit 202 inputs a host image to be transmitted and performs discrete wavelet transform (step 302) to output discrete wavelet coefficients. The discrete wavelet coefficients consist of a square matrix as is known in the art. In the present embodiment, it is assumed that the discrete wavelet coefficients consist of a matrix of M rows x M columns when a predetermined positive integer is M.

The significant coefficient extracting unit 204 extracts a coefficient of 25% having a high importance among the M × M wavelet coefficients (step 304). A value of 25% means that one quarter of the total M × M wavelet coefficients are chosen.

The random number generation unit 206 generates a random number according to a rule corresponding to a predetermined encryption code that can be selectively input by a user.

The coefficient substitution / combination unit 208 sets the substitution position at the N × N position when the predetermined positive integer smaller than M is N using the random number (step 306). In addition, the coefficient replacement / combination unit 208 replaces and combines the coefficients on the N × N position of the M × M wavelet coefficients with N × N significant wavelet coefficients, respectively (step 308). Here, N is set to 1/2 of M, that is, 0.5M, because a coefficient of 25% among the M × M wavelet coefficients is selected as the coefficient of high importance in step 304. In view of the nature of watermarking, N is preferably 0.5M or less in consideration of the influence on the original host image. By such substitution and combination, the significant wavelet coefficients of the host image to be transmitted are randomly distributed as wavelet coefficients of the signature image within the wavelet coefficients of the synthesized host image.

The inverse discrete wavelet transform unit 210 generates a synthesized host image by inverse discrete wavelet transforming the replaced M × M wavelet coefficients (step 310). Since the synthesized host image contains a signature image by watermarking therein, there is no significant difference from the original host image, and the signature image is not easily identified by the naked eye.

The compression unit 212 outputs a compressed bitstream by compressing the synthesized host image and the encryption code (step 312). Here, the encryption code is compressed together to set the separation position according to the same rules as the replacement position used when encoding.

As described above, the encoding unit 20 rearranges coefficients having high importance among wavelet coefficients of the host image to be transmitted in a pseudo-randomly manner within the wavelet coefficients of the host image. Thus, important portions of the host image are watermarked within the host image itself.

The host image watermarked by the encoding unit 20 is decoded by the decoding unit 22. Referring to the operation of the decoding unit 22, the decompression unit 222 decompresses the compressed bitstream to restore the host image and the encryption code (step 402). The encryption code is for determining a generation rule of pseudo random numbers used in the substitution and combination of coefficients in the encoding unit 20. In the decoding section 22, the encryption code determines the generation rule of pseudo random number used for the separation of coefficients to be described next, similarly as in the encoding section 22.

The image quality evaluator 224 evaluates the image quality of the reconstructed host image and determines whether the quality of the reconstructed host image is greater than or equal to the reference level (step 42). That is, the image quality evaluator 224 outputs a control signal that is logical "high" if the quality of the reconstructed host image is less than the reference quality and logical "low" otherwise. Evaluation of the image quality may be performed by, for example, a method of distortion evaluation.

The discrete wavelet transform unit 226 outputs M × M wavelet coefficients when a predetermined positive integer is M by performing discrete wavelet transform on the restored host image in response to a logic “high” control signal. do.

Meanwhile, the random number generation unit 228 generates a random number according to a rule corresponding to the restored encryption code. The pseudo random number generation unit 228 performs the same operation as the pseudo random number generation unit 228 of the encoding unit 20, and thus, the restored encryption code, that is, the same encryption code as that used in the encoding unit 20. If is input, a random number is generated according to the same rules as the random number generated by the encoding unit 20 (step 462).

The coefficient separating unit 230 sets the separation position of N × N locations when the predetermined positive integer smaller than M is N using the random number (step 464). In addition, the coefficient separating unit 230 separates the coefficients on the substitution positions at the N × N locations among the M × M wavelet coefficients (step 466). Here, N is determined at the time of substitution and combination of coefficients in the encoding section 20, preferably 0.5 M or less.

The inverse discrete wavelet transform unit 232 reconstructs the signature image by performing inverse discrete wavelet transform of the M × M wavelet coefficients (step 468). The reconstructed signature image is an image that is reconstructed by coefficients of high importance among wavelet coefficients corresponding to the original host image, and thus is close to the original host image.

The image scale converter 234 converts the scale of the reconstructed signature image into the scale of the host image (step 470). The scaled signature image is set as the host image.

Finally, the image selector 236 selects the signature image whose scale is enlarged in response to the logic "high" control signal, and selects and outputs the reconstructed host image in response to the logic "low" control signal.

Therefore, if the quality of the reconstructed host image is higher than or equal to a predetermined criterion, the decoding unit 22 sets (step 44) the synthesized host image as the host image and outputs the host image. On the other hand, if the quality of the reconstructed host image is less than a predetermined criterion, the signature image secretly contained in the host image is reconstructed, and the reconstructed signature image is set as a host image (step 46) and output.

According to the digital image codec device and the method of the present invention as described above, since the distorted host image transmitted in the severe noise environment is corrected using the current screen, the information flow is continuous and the scene changes abruptly. Screen calibration error does not occur.

The encoding unit 20 and the decoding unit 22 of the codec device according to the embodiment of the present invention may be implemented as a digital image coding device and a decoding device, respectively, as necessary.

In addition, the digital image coding method and the decoding method according to the present invention can be written as a program that can be executed in a computer. And it can be implemented in a general-purpose digital computer for operating the program from a medium used in the computer. The medium includes a magnetic recording medium such as a floppy disk or a hard disk, an optical recording medium such as a CD-ROM or a DVD, and a storage medium such as a carrier wave such as transmission over the Internet. In addition, these functional programs, code and code segments can be easily inferred by a programmer in the art to which the present invention belongs.

As described above, according to the present invention, since the distorted host image transmitted in the severe noisy environment is corrected using the current screen, the information flow is continuous, and the screen correction error does not occur even when the scene changes abruptly.

Claims (12)

In the digital image coding apparatus using watermarking, A discrete wavelet transform unit for performing discrete wavelet transform on the host image to be transmitted and outputting M × M wavelet coefficients when a predetermined positive integer is M; An important coefficient extracting unit configured to extract a predetermined ratio coefficient having a high importance among the M × M wavelet coefficients; A random number generation unit for generating a random number according to a rule corresponding to an encryption code; And When the predetermined positive integer smaller than M is N using the random number, substitution position information indicating the substitution position of N × N points is obtained, and N indicated in the substitution position information among the M × M wavelet coefficients is obtained. And a coefficient replacement / combination unit for replacing and combining the coefficients on the alternate positions at the × N locations into N × N significant wavelet coefficients, respectively. The method of claim 1, An inverse discrete wavelet transform unit for generating a synthesized host image by inverse discrete wavelet transforming the replaced M × M wavelet coefficients; And And a compression unit configured to output the bitstream by compressing the synthesized host image and the encryption code. The digital image coding device according to claim 1, wherein N is 0.5 M or less. A digital video decoding apparatus for decoding a coded bit stream using watermarking, An decompression unit configured to decompress the bitstream to restore a host image and a predetermined encryption code; An image quality evaluator for evaluating the image quality of the reconstructed host image and outputting a control signal that is in a first logic state if the reconstructed host image is less than a reference quality and otherwise becomes a second logic state; A discrete wavelet transform unit for performing discrete wavelet transform of the reconstructed host image in response to a control signal of a first logic state and outputting an M × M wavelet coefficient when a predetermined positive integer is M; A random number generator for generating a random number according to a rule corresponding to the restored encryption code; When the predetermined positive integer smaller than M is N using the random number, separation position information indicating a replacement position of N × N points is obtained, and N indicated in the separation position information among the M × M wavelet coefficients is obtained. A coefficient separator for separating the coefficients on the substitute positions at the xN position; An inverse discrete wavelet transform unit which outputs a signature image by inverse discrete wavelet transforming the M × M wavelet coefficients; An image scale converter configured to enlarge the scale of the reconstructed signature image to the scale of the host image; And And an image selector configured to select the signature image whose scale is enlarged in response to the control signal of the first logic state and to select and output the reconstructed host image in response to the control signal of the second logic state. Digital video decoding device. The apparatus of claim 4, wherein N is 0.5 M or less. In a digital video codec device using watermarking, A discrete wavelet transform unit for performing discrete wavelet transform on the host image to be transmitted and outputting M × M wavelet coefficients when a predetermined positive integer is M; An important coefficient extracting unit configured to extract a predetermined ratio coefficient having a high importance among the M × M wavelet coefficients; A random number generation unit for generating a random number according to a rule corresponding to an encryption code; When the predetermined positive integer smaller than M is N using the random number, substitution position information indicating the substitution position of N × N points is obtained, and N indicated in the substitution position information among the M × M wavelet coefficients is obtained. A coefficient replacement / combination unit for replacing and combining the coefficients on the x-N places of substitution positions with N x N important wavelet coefficients, respectively; An inverse discrete wavelet transform unit for generating a synthesized host image by inverse discrete wavelet transforming the replaced M × M wavelet coefficients; A compression unit for compressing the synthesized host image and the encryption code to output a bitstream; A decompression unit for decompressing the compressed bitstream to restore a host image and a predetermined encryption code; An image quality evaluator for evaluating the image quality of the reconstructed host image and outputting a control signal that is in a first logic state if the reconstructed host image is less than a reference quality and otherwise becomes a second logic state; A discrete wavelet transform unit for performing discrete wavelet transform of the reconstructed host image in response to a control signal of a first logic state and outputting an M × M wavelet coefficient when a predetermined positive integer is M; A random number generator for generating a random number according to a rule corresponding to the restored encryption code; When the predetermined positive integer smaller than M is N using the random number, separation position information indicating a replacement position of N × N points is obtained, and N indicated in the separation position information among the M × M wavelet coefficients is obtained. A coefficient separator for separating the coefficients on the substitute positions at the xN position; An inverse discrete wavelet transform unit which outputs a signature image by inverse discrete wavelet transforming the M × M wavelet coefficients; An image scale converter configured to enlarge the scale of the reconstructed signature image to the scale of the host image; And And an image selector configured to select the signature image whose scale is enlarged in response to the control signal of the first logic state and to select and output the reconstructed host image in response to the control signal of the second logic state. Digital video codec device In the digital image coding method using watermarking, (a) obtaining M × M wavelet coefficients when a predetermined positive integer is M by discrete wavelet transforming a host image to be transmitted; (b) extracting a predetermined ratio coefficient having a high importance among the M × M wavelet coefficients; (c) generating a random number according to a rule corresponding to an encryption code; (d) using the random number to obtain replacement location information indicating replacement locations of N × N locations when a predetermined positive integer smaller than M is N; And (e) replacing and combining the coefficients on the N × N locations shown in the replacement location information among the M × M wavelet coefficients into N × N significant wavelet coefficients, respectively; Digital video coding method. The method of claim 7, wherein (f) generating a synthesized host image by inverse discrete wavelet transforming the substituted M × M wavelet coefficients; And (g) compressing the synthesized host image and the encryption code. 8. The method of claim 7, wherein N is 0.5M or less. A digital video decoding method for decoding a coded bit stream using watermarking, (a) evaluating the quality of the synthesized host image to determine whether the quality is equal to or greater than a predetermined criterion; (b) setting the synthesized host image as the host image when it is determined in step (a) that the quality of the host image is greater than or equal to a predetermined criterion; And (c) if it is determined in step (a) that the quality of the host image is less than a predetermined criterion, setting the signature image secretly stored in the synthesized host image as the host image; and decoding the digital image. Way. The digital image decoding method of claim 10, further comprising reconstructing the synthesized host image and a predetermined encryption code by decompressing the bitstream. The method of claim 10, wherein step (c) comprises: Obtaining M × M wavelet coefficients when a predetermined positive integer is M by discrete wavelet transforming the reconstructed host image; Generating a random number according to a rule corresponding to the restored encryption code; Obtaining separation position information indicating a substitution position of N × N locations when the predetermined positive integer smaller than M is N using the random number; Separating the coefficients on the N × N positions of the substitution positions shown in the separation position information among the M × M wavelet coefficients; Restoring a signature image by inverse discrete wavelet transforming the M × M wavelet coefficients; Converting the scale of the reconstructed signature image to the scale of the host image; And setting the converted signature image as a host image.