WO1998058328A1 - Procede et dispositif de conversion de signal numerique, dispositif et procede de generation de matrice de transformation et support d'alimentation - Google Patents
Procede et dispositif de conversion de signal numerique, dispositif et procede de generation de matrice de transformation et support d'alimentation Download PDFInfo
- Publication number
- WO1998058328A1 WO1998058328A1 PCT/JP1998/002653 JP9802653W WO9858328A1 WO 1998058328 A1 WO1998058328 A1 WO 1998058328A1 JP 9802653 W JP9802653 W JP 9802653W WO 9858328 A1 WO9858328 A1 WO 9858328A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- matrix
- signal
- digital signal
- conversion
- processing
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Ceased
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/14—Fourier, Walsh or analogous domain transformations, e.g. Laplace, Hilbert, Karhunen-Loeve, transforms
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
- H04N19/10—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
- H04N19/102—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the element, parameter or selection affected or controlled by the adaptive coding
- H04N19/124—Quantisation
- H04N19/126—Details of normalisation or weighting functions, e.g. normalisation matrices or variable uniform quantisers
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
- H04N19/48—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using compressed domain processing techniques other than decoding, e.g. modification of transform coefficients, variable length coding [VLC] data or run-length data
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
- H04N19/50—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using predictive coding
- H04N19/59—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using predictive coding involving spatial sub-sampling or interpolation, e.g. alteration of picture size or resolution
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
- H04N19/60—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using transform coding
- H04N19/61—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using transform coding in combination with predictive coding
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
- H04N19/90—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using coding techniques not provided for in groups H04N19/10-H04N19/85, e.g. fractals
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
- H04N19/10—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
- H04N19/102—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the element, parameter or selection affected or controlled by the adaptive coding
- H04N19/124—Quantisation
Definitions
- the present invention relates to a frequency-domain digital signal subjected to orthogonal transform such as DCT (discrete cosine transform).
- the present invention relates to a digital signal conversion method and apparatus for performing resolution conversion and base conversion, a conversion matrix creation method and apparatus, and a providing medium.
- orthogonal transform coding such as DCT (Discrete Cosine Transform) coding has been known as a coding method for efficiently compressing still image data or moving image data. I have. When dealing with such orthogonally transformed digital signal, it may be necessary to change the resolution and the transformation basis.
- the first orthogonally transformed digital signal with a resolution of 720 ⁇ 480 pixels as an example of a digital video format for home use, a so-called MPEG1 format of 35 2 ⁇ 240
- the first signal is subjected to inverse orthogonal transformation to restore it to a time axis signal, and then interpolation is performed.
- Orthogonal transformation is performed again by performing transformation processing such as decimation and thinning. In this way, the orthogonally transformed digital signal is often inversely transformed to restore the original signal, and then various transformation operations are performed, and then the orthogonal transformation is performed again.
- orthogonal transformation and its inverse transformation require a lot of calculations, and when performing resolution conversion as described above, it is necessary to perform orthogonal transformation and inverse orthogonal transformation in addition to the resolution conversion processing. Can not perform good processing. In addition, there is a disadvantage that the signal is degraded because errors accumulate as the amount of calculation increases.
- the present invention has been made in view of such circumstances, and has a digital signal conversion method that is simple in processing, causes little signal deterioration, and can perform conversion processing such as resolution conversion. It is an object of the present invention to provide a conversion matrix creating method and apparatus, and a providing medium.
- the present invention generates an inverse orthogonal transform matrix T s ⁇ 1 for inverse orthogonal transform and an orthogonal transform matrix T d for orthogonal transform, and generates the generated inverse orthogonal transform.
- a transformation matrix D is generated using the matrix T s ⁇ 1 and the orthogonal transformation matrix T d, and the input digital signal in the orthogonally transformed area is transformed in the area that is orthogonally transformed using the above transformation matrix D. And obtains the output digital signal of the orthogonally transformed area.
- the input digital signal is obtained by orthogonally transforming the original signal in the time domain by the orthogonal transform matrix T s corresponding to the inverse orthogonal transform matrix T s ⁇ 1.
- the output digital signal is decoded by an inverse transform matrix T d ⁇ 1 corresponding to the orthogonal transform matrix T d and converted into a signal in the time or space domain. To be converted.
- the transformation matrix D is obtained by arranging m inverse orthogonal transformation matrices T s (R, 1 on the diagonal axis to form an L-order square matrix A, and an orthogonal transformation matrix T d of base length L (N low frequency basis vectors are taken out to create an N-by-L matrix B,
- the transformation matrix D is a matrix that transforms the transformation basis. If the orthogonal transformation matrices T s and T d are different, orthogonal transformation is performed.
- an n-dimensional digit signal is used as the input digit signal, and a conversion process is performed on the orthogonally transformed area using the transform matrix D for each dimension, and an n-dimensional output digit signal is obtained. Can be obtained.
- the conversion magnification differs for each dimension, the conversion for each dimension A matrix may be created and conversion processing may be performed at different magnifications for each dimension using these conversion matrices.
- the conversion block of the orthogonally transformed input digit signal is used as it is to compensate for the predetermined value to convert it to a block with a length of m times. What is necessary is just to obtain the output digital signal of the area that has been subjected to the orthogonal transformation.
- FIG. 1 is an explanatory diagram for explaining a digitized signal conversion method according to a first embodiment of the present invention.
- FIG. 2 is a diagram illustrating an example of input / output signals according to the first embodiment of the present invention.
- FIG. 3 is a diagram illustrating an example of signals before and after conversion in a case where the first embodiment of the present invention is applied to a two-dimensional digital signal.
- FIG. 4 is an explanatory diagram for explaining an operation of multiplying the input matrix signal by a conversion matrix.
- FIG. 5 is an explanatory diagram for explaining a process of obtaining a two-dimensional output digital signal by applying a transformation matrix D twice to a two-dimensional input digital signal.
- FIG. 6 is a flowchart illustrating an example of a processing procedure for creating a resolution conversion matrix.
- FIG. 7 is a flowchart for explaining an example of a processing procedure for performing resolution conversion in the frequency domain using a resolution conversion matrix.
- FIG. 8 is an explanatory diagram for explaining an enlargement process in the frequency domain.
- FIG. 9 is a diagram for explaining conversion between two formats having different sampling ratios of component signals of a luminance signal and a color difference signal.
- FIG. 10 is a flowchart for explaining an example of a processing procedure for creating a resolution conversion matrix in the case of a different reduction ratio for each dimension.
- FIG. 11 is a flowchart for explaining an example of a processing procedure for performing resolution conversion in the frequency domain using a resolution conversion matrix in the case of a different reduction ratio for each dimension.
- FIG. 12 is an explanatory diagram for explaining a digital signal conversion method according to the second embodiment of the present invention.
- FIG. 13 is a diagram illustrating an example of input / output signals according to the second embodiment of the present invention.
- Figure 14 shows the connection matrix A, the selection matrix W, the processing matrix X, and the partitioning matrix B. It is a figure for explaining each function.
- FIG. 15 is a diagram for explaining an example in which an input signal is divided into metablocks and only necessary parts are selected and processed.
- FIG. 16 is a flowchart for explaining the procedure for generating the connection matrix A.
- Figure 17 is a flowchart for explaining the initialization procedure of the parameter.
- FIG. 18 is a flowchart for explaining the procedure for generating the selection matrix W.
- FIG. 19 is a flowchart for explaining the procedure for generating the processing matrix X.
- FIG. 20 is a flowchart for explaining the procedure for generating the partitioning matrix B.
- FIG. 21 is a diagram for explaining orthogonal transformation of a color difference signal of the DV (Digital Video) standard.
- FIG. 22 is a diagram for explaining a resolution conversion process for a color difference signal of the DV (Digital Video) standard.
- FIG. 23 is a block diagram showing a schematic configuration of hardware used in the embodiment of the present invention.
- FIG. 24 is a system diagram showing an example of a system for converting the embodiment of the present invention from the DV standard to the MPEG1 standard.
- FIG. 25 is a block diagram illustrating an example of a signal conversion device that converts the embodiment of the present invention from the DV standard to the MPEG-1 standard.
- BEST MODE FOR CARRYING OUT THE INVENTION A digital signal conversion method and apparatus according to the present invention performs a conversion process such as a resolution conversion or a base conversion on an orthogonally transformed digital signal without performing an inverse orthogonal transformation or an orthogonal transformation again. Things.
- the transformation matrix creation method according to the present invention creates a transformation matrix for directly performing the resolution conversion, the S-bottom transformation, and the like in an orthogonally transformed area, for example, a frequency domain or a sweat area. is there.
- MPEG MPEG
- ISO / IECJTC 1 / S C29 International Organization for Standardization / International Electrotechnical Commission, Joint Technical Committee 1 / Sub Committee 29. It is an abbreviation of Moving Picture Experts Group, which considers moving picture compression and coding in 29), and MP0EG172 is the MPEG1 standard and IS013818 is the MPEG2 standard.
- IS011172-1 and IS013818-1 are for multimedia multiplexing
- IS011172-2 and IS013818-2 are for video
- IS011172-3 and IS013818-3 are for audio. Standardized.
- the image signal is compressed and coded by using the correlation in the time and space direction of the image in the unit of a picture (frame or field).
- the use of spatial correlation is realized by using DCT (Discrete Cosine Transform) coding.
- orthogonal transform such as DCT is also performed by JPEG (Joint Photogra (Pic Coding Expert Group) is widely used for various image information compression coding.
- JPEG Joint Photogra (Pic Coding Expert Group) is widely used for various image information compression coding.
- orthogonal transform enables compression coding with high compression efficiency and excellent reproducibility by transforming an original signal in the time domain or the space domain into an orthogonally transformed domain such as the frequency domain ( hereinafter, referred to as the orthogonal transform).
- Preferred embodiments of a digital signal conversion method and apparatus, a conversion matrix creation method and apparatus, and a providing medium according to the present invention will be described with reference to the drawings.
- FIG. 1 is an explanatory diagram for explaining a digital signal conversion method and a conversion matrix creation method according to the first embodiment of the present invention.
- a human-powered orthogonal transformation matrix generation unit 11 generates an inverse matrix T s ( k ) 1 of an orthogonal transformation matrix T s representing an orthogonal transformation performed in advance on an input digital signal, and generates a transformation matrix. It is sent to the generator 13.
- the output orthogonal transformation matrix generator 12 generates an inverse transformation matrix T d (orthogonal transformation matrix T d corresponding to 1 ) indicating the inverse orthogonal transformation to be performed on the output digital signal.
- Sent to 3 The conversion matrix generation unit 13 generates a conversion matrix D for performing conversion processing such as resolution conversion in the frequency domain, and sends it to the signal conversion unit 14.
- the signal conversion unit 14 converts the input digital signal 15 converted into, for example, the frequency domain by the orthogonal conversion in an orthogonally converted area, for example, the frequency domain, and outputs the output digital signal.
- Signal 16 shall be used.
- an original signal in the time domain (or the spatial domain) (original signal A) is transformed into, for example, the frequency domain by the orthogonal transform matrix T s ( k ), and the frequency signal B i (corresponding to the above-mentioned input digital signal 15), which is reduced (or expanded) to, for example, N / L by the above-mentioned signal conversion section 14 to obtain a frequency signal B 2 (the above-mentioned output digital signal). It corresponds to No. 16.
- the frequency signal B 2 is inversely orthogonally transformed by the inverse transformation matrix Td (L 1) to obtain a time-domain signal C.
- the matrix (orthogonal transform matrix) in which the orthogonal transform base vector of length n is arranged in each row (orthogonal transform matrix) is T ( n ), and its inverse transform matrix is ⁇ Write like 1 .
- T ( n ) the matrix (orthogonal transform matrix) in which the orthogonal transform base vector of length n is arranged in each row (orthogonal transform matrix)
- ⁇ Write like 1 Denotes a vector representation of X.
- both matrices are ⁇ -order square matrices.
- the following equation (1) shows the one-dimensional DC ⁇ conversion matrix, 8) for ⁇ 28.
- the inverse orthogonal transform matrix Ts ( k ) 1 generated by the input orthogonal transform matrix generator 11 corresponds to the orthogonal transform process (reverse process) when the input digital signal 15 is generated.
- the orthogonal transformation matrix Td generated by the output orthogonal transformation matrix generation unit 12 (or the inverse of decoding the output digital signal converted by the signal conversion unit 14, that is, converting it to the time domain). It is assumed that these orthogonal transformation matrix generators 11 and 12 can generate base vectors of any length, corresponding to (or inverse of) the orthogonal transformation processing.
- these orthogonal transformation matrix generators 11 and 12 may be the same orthogonal transformation matrix generator, in which case the orthogonal transformation matrices Ts ( k ) and Td ( L ) are It becomes the same kind of orthogonal transformation matrix that differs only in length.
- the orthogonal transformation matrix generator exists for each different orthogonal transformation scheme.
- the transformation matrix generation unit 13 converts the inverse orthogonal transformation matrix Ts (k) -1 generated by the input orthogonal transformation matrix generation unit 11 into a diagonal as shown in the following equation (2). Arrange m above, and create an L-order square matrix A.
- the base length of the output digital signal 16 is N
- the low-frequency base vectors N of the orthogonal transformation matrix Td are extracted, and a matrix B composed of N rows and L columns is created.
- ⁇ are N low-frequency components when Td (is represented by the basis vector as follows. ⁇
- This matrix D is a conversion matrix for converting the resolution into the above-mentioned reduction ratio (or enlargement ratio) N / L.
- Hi is a scalar value or a vector value, and is a coefficient for level correction and the like.
- each metablock Mi is multiplied by a conversion matrix D.
- the vector C i of length N obtained by this becomes a frequency vector signal in which each metablock is reduced to N / L.
- a signal B 2 obtained by reducing the whole to N / L is obtained (the obtained output digital signal B 2 is an inverse orthogonal transform matrix Td ( N ) 1 can be converted to a time axis signal (C in Fig. 2).
- the conversion matrix D shown in the above equation (5) is a conversion matrix for performing resolution conversion and base length conversion at a magnification of N / L.
- N ⁇ L A matrix that transforms the transformation basis.
- the output digit signal in the frequency domain uses the same decoder as the input digit signal, In other words, the same inverse orthogonal transform matrix can be used.
- the transform matrix D does not transform the basis. Becomes a resolution conversion matrix that reduces the resolution to 1 / m.
- the above operations are separable to the dimensions of the original signal, can be easily extended to more than two dimensions, and can be applied to various media such as audio, still images, and moving images .
- the above-described conversion is first performed for each row, and then the same is applied to each column.
- FIG. 3 schematically illustrates an example of each signal before and after conversion in a time domain and a frequency domain when the first embodiment of the present invention is applied to a two-dimensional digital signal such as an image signal.
- FIG. The example of FIG. 3 shows a case where an original signal 21 which is a two-dimensional digital signal such as an image signal is reduced to 1/2 in each of the vertical and horizontal directions and converted into a two-dimensional digital signal 22. .
- the original signal 21 is divided into blocks of, for example, 8 ⁇ 8 pixels in length and width, and an 8 ⁇ 8 matrix is used by an orthogonal transformation matrix T s ( 8 ) such as a DCT (discrete cosine transform) with a base length of 8.
- T s orthogonal transformation matrix
- the signal is converted into a signal 23 in the frequency domain in units of the blocks, and this is used as an input digital signal in the frequency domain.
- This one block converts a base of length 8 into a base of length 16 for the two-dimensional input digital signal 23 composed of 8 ⁇ 8 frequency domain data by the base conversion matrix T as described above.
- the digital signal 24 on the frequency axis with 16 x 16 blocks as one block, and the low-frequency portion of this 1 block 16 x 16 blocks indicated by oblique lines in the figure By extracting only 8 ⁇ 8 from each block and combining them, an output digital signal 25 in the frequency domain can be obtained.
- the output digital signal 25 is converted into a time-domain digital signal 22 by an inverse orthogonal transformation matrix T d ( 8 ) 1 having a base length of 8. If it is decoded (decoded), this is a signal whose resolution has been converted so that the original signal 21 in the time domain is reduced to 1/4.
- the transformation matrix D is an 8 ⁇ 16 matrix from the above equation (5). As shown in FIG. 4, this 8 ⁇ 16 conversion matrix D is composed of two blocks arranged vertically in the input domain signal 23 in the frequency domain. By multiplying by 2, the shaded portion (8 x 1) of the signal 26 whose vertical length is compressed to 1/2 is obtained. By performing this for all of the input digital signal 23, that is, by multiplying the matrix D for every two blocks of each column of the input digital signal 23, the vertical length becomes 1 / With 2, a signal 26 with the same horizontal length as the original is obtained.
- a conversion matrix D is applied to the input digital signal 23 to obtain a signal (matrix data) 26 having a vertical length of 1/2.
- the signal 26 is transposed, that is, the rows and columns are interchanged, and a transposed signal 26 1 is obtained.
- the vertical length i.e. the original incoming The signal (corresponding to the horizontal length of the power signal signal 23) is reduced by half to obtain a signal (matrix data overnight) 27.
- the obtained signal 27 1 becomes the above output digital signal 25 with both the length and width of the input digital signal 23 reduced to 1/2.
- FIG. 6 is a flowchart showing a procedure of the above-described resolution conversion, in particular, a procedure of creating a conversion matrix D for reducing each dimension to 1 / m.
- the length of the base used for the input signal is set to k, and the reduction ratio m (that is, reduced to 1 / m) is determined in the next step S32.
- step S 35 As shown in the above equation (2), the m inverse transformation matrices T ( k) -1 are arranged on the diagonal to generate the L-order square matrix A.
- step S 3 a matrix T of base length L, which is the output orthogonal transformation matrix described above, is created, and in step S37, k low-frequency base vectors of the matrix T are extracted.
- Fig. 7 shows the resolution using the transformation matrix D created in the procedure of Fig. 6.
- 9 is a flowchart showing a processing procedure when performing conversion.
- the first step S41 in FIG. 7 it is determined whether or not the conversion has been completed in all dimensions of the input digital signal, and if the conversion has been performed in all dimensions, the resolution conversion is performed. The process ends.
- the next step S42 the next dimension is set to the current dimension. Immediately after the start of resolution conversion, the first dimension is set to the current dimension.
- step S43 it is determined whether or not the magnitude of the input signal is not equal to a multiple of L. If not, the process proceeds to step S44, where the input signal is supplemented with dummy data or the like to reduce L. To be a multiple of. In this way, the input signal that is originally a multiple of L or is a multiple of L by supplementation is divided into metablocks of length L in step S45.
- the resolution can be reduced to 1 / m for all dimensions without changing the basis for a one-dimensional or multi-dimensional input digital signal.
- the conversion matrix D can be used repeatedly during the resolution conversion processing of FIG. 7 once it is obtained by the procedure of FIG.
- the resolution can be increased at an arbitrary magnification by adding a high-frequency component to the input digital signal in the frequency domain.
- FIG. 8 shows an example of an enlargement method for supplementing 0 as a high-frequency component.
- the original signal A in the time domain is subjected to orthogonal transformation such as DCT as described above to obtain an input digital signal B, in the frequency domain.
- the input digital signal is expanded by m times for each block! ⁇
- the original block is used as it is and zeros are filled in Signal B 2.
- N / L reduction conversion may be performed using the digital signal after the enlargement as the input digital signal of the embodiment described above.
- supplementing 0 to the high frequency component is an example, and a more appropriate value may be supplemented.
- a plurality of conversion matrices generated at different sampling ratios with respect to the original signal such as a luminance signal and a color difference signal of a two-dimensional digital signal, are generated.
- Embodiments of the present invention can also be applied to component signals.
- FIG. 9 shows an example of conversion between two formats having different sampling ratios of components such as a luminance signal and a color difference signal.
- the 8 ⁇ 8 blocks of the luminance signal 51 of the signal before conversion are grouped into four blocks, and one block (8 ⁇ 8) of each of the color difference signals 52 and 5 3 is provided. Yes, it is.
- the luminance signal 56 block (8 x 8) of the converted signal has a total of 4 blocks each of 2 blocks vertically and horizontally.
- One block (8 x 8) of each of the color difference signals 57 and 58 corresponds to the collected data.
- the format is converted and the resolution is reduced to ⁇ each in the vertical and horizontal directions.
- two types of resolution conversion matrices of 1/2 and 1/4 without changing the basis are prepared by the above-described resolution conversion process B of the present embodiment, and the input luminance signal 51 is both vertically and horizontally.
- the output luminance signal 56 should be reduced to 1/2 and converted to the output luminance signal 56, and the input color difference signals 52 and 53 should be reduced to 1/4 only in the vertical direction.
- the first step S61 of the flowchart of FIG. 10 it is determined whether or not the creation of the transformation matrix has been completed in all dimensions, and if the creation of the transformation matrix has been completed in all dimensions, the resolution conversion is performed. End the matrix generation process.
- the length of the basis used for the input signal of the current dimension is set to k, and in the next step S64, the contraction rate ⁇ .,. / mi).
- the order L.i of the next square matrix A, L, described later, is determined by Lj2kxmi.
- the stearyl-up S 6 7 inverse transformation matrix is T, (to create a k.
- the inverse transform matrix T (k) 1 Are arranged on a diagonal of mj to create a square matrix A of order L above.
- the matrix T ( ) ) is extracted to create a matrix B consisting of k rows and L j columns.
- a transformation matrix D j (two times jB A) is obtained from the above equation (5), where “hi” is a scalar single value or a vector value for correction in the j-th dimension.
- FIG. 11 is a flowchart showing a processing procedure in the case where resolution conversion is performed for each dimension using the conversion matrices ⁇ , ⁇ created in the procedure of FIG.
- the new dimension (j) is set to the current dimension.
- step S83 it is determined whether or not the magnitude of the input signal is not equal to a multiple of Lj. To make it a multiple of L j. Thus, Lj is originally a multiple of L, or The multiplied input signal is divided into metablocks of length L j in step S85.
- step S86 it is determined whether or not all the data blocks for the current dimension have been converted. If the conversion of all the data blocks has been completed, the flow returns to step S81, and all the meta data blocks have been processed. If the conversion of the block has not been completed, the process proceeds to step S87.
- the resolution of the j-th dimension can be reduced to 1 / m j for the n-dimensional input digital signal without changing the basis.
- the conversion matrix D. once obtained by the procedure shown in FIG. 10, can be used repeatedly during the resolution conversion process shown in FIG.
- the transform basis can be arbitrarily changed for the input digitized signal transformed into the frequency domain by the orthogonal transform such as DCT,
- the orthogonal transform such as DCT
- the resolution and the like can be reduced at an arbitrary magnification.
- the orthogonal transformation method and the overall length of the output digital signal being the same as the input digital signal, it is possible to obtain an output digital signal having a different base length. Can be.
- the orthogonal transformation method, base length, and resolution Different output digital signals can be obtained.
- the decoder for the input digital signal can be used as it is. Can be used and only resolution can be converted. Also, by combining enlargement and reduction, resolution conversion can be performed at any magnification. It can be applied to not only one-dimensional but also multi-dimensional frequency domain signals. In addition, since conversion can be performed at any magnification, correspondence between components is maintained even when the size or configuration of the conversion block differs depending on the component, such as a luminance signal or color difference signal in a two-dimensional signal. The resolution conversion can be performed as it is.
- the compressed digital signal is subjected to only the compression / expansion, thereby enabling various transforms according to the first embodiment.
- a metablock in which several blocks are put together is used, and resolution conversion can be performed without changing the block size.
- the processing is performed on a part of the meta-block or the part that crosses the boundary of the meta-block. It cannot be performed, but by extending it as follows, resolution conversion can be performed at any magnification, and partial signals of any magnitude across block boundaries can be processed.
- FIG. 12 is a diagram showing a digital device according to the second embodiment of the present invention.
- FIG. 4 is an explanatory diagram for describing a signal conversion method and a conversion matrix creation method.
- a concatenated matrix generation unit 1 1 1 that is an extension of the input orthogonal transformation matrix generation unit 11 of FIG. Section 112, a selection matrix generation section 117 for generating a selection matrix for extracting a part of the metablock or an arbitrary part crossing the boundary of the data block, and a desired resolution conversion processing.
- a processing matrix generation unit 1 18 for generating a processing matrix for performing the signal processing of the above, and a conversion matrix generation unit 1 13 for obtaining a required matrix from these four matrices.
- the transformation matrix D from 3 is sent to the signal transformation unit 114, and the orthogonally transformed input digital signal 115 is transformed in the orthogonal transformed state, and the output digital signal is output.
- Each function of the division matrix B from the generation unit 112 will be described with reference to FIG.
- (a) corresponds to the above-mentioned one metablock of the above-mentioned input digital signal 115, and this input digital signal (a) is obtained by using an orthogonally transformed domain, for example, a frequency domain or DCT. Compression This is the signal of the compression area that has been encoded.
- the selection matrix W has a function of converting a part to be subjected to arbitrary processing from the signal (b) in the time or space domain.
- the signal (c) with the length M from the p-th to the q-th of the signal L (b) in the time or space domain with the length L is extracted. This is for performing the actual signal conversion processing such as the above-described resolution conversion and reduction / enlargement in the time or space domain, and the user may specify a desired one.
- the signal (c) of length M is converted to the signal (d) of length N.
- the signal (e) of the area is used, and this signal (e) corresponds to the above-mentioned output digital signal 1 16.
- the concatenated matrix generation unit 111 generates an inverse matrix Ts ( kI) -1 of an orthogonal transform matrix Ts ( kI ) 'representing an orthogonal transform previously performed on the input digital signal 1 15
- the inverse matrix Ts ( kI ) 1 is arranged diagonally as shown in the following equation (7), and an L-order square matrix, A matrix of L rows and L columns (L kz Xim) is created, and this is referred to as a connection matrix A.
- This equation (7) corresponds to the equation (2) in the first embodiment.
- the selection matrix generation unit 117 generates a selection matrix W for selecting a part to be processed from the metablock as described above.
- the selection matrix W is an element that is 1 for each column. There is one by one, and the rest is a matrix of 0.
- the following equation (9) shows an M-by-L matrix W for extracting a signal of length M from the eye to the qth.
- an M-order square matrix (a matrix of M rows and M columns) in which the M columns from the p columns to the q columns in the L columns are arranged diagonally with 1 is obtained. The remaining elements are all 0.
- the processing matrix generation unit 118 generates a processing matrix X for actually performing signal processing, for example, resolution conversion.
- the processing matrix X there is no need to be aware of the processing in the orthogonally transformed area, and the user may arbitrarily define a matrix that performs processing at the original signal level in the time or space domain.
- the processing matrix X is a matrix of N rows and M columns in the example of FIG. 14 described above, and a specific example will be described later.
- the transformation matrix generation unit 113 shown in FIG. 12 described above uses the following equation (based on the matrices generated by the matrix generation units 111, 117, 118, and 112). 1 0), an N-by-L transformation matrix D, i.e.,
- the signal C i is a metablock having the length N described above, and has a length k. M of blocks. It is divided into pieces.
- the processing matrix X may be a matrix for processing the original signal in the time or empty f?; J region without being aware that the signal to be processed is orthogonally transformed.
- Various signal processing can be used. For example, resolution conversion processing of any magnification, processing of transforming the basis of orthogonal transformation, effect processing on compressed images such as JPEG and MPEG, filtering processing, digital video signal to MPEG And the like, and a format conversion process such as.
- the processing matrix X any matrix can be used as long as resolution conversion can be performed on the original signal.
- a matrix represented by the following equation (1 2) is used. . ⁇ in this equation (12) is a coefficient for level correction.
- N LN Using these matrices, generate a transformation matrix D as described above, and perform signal conversion processing, so that the original signal is not restored, and the orthogonally transformed area remains unchanged at any magnification.
- the resolution conversion can be performed.
- the procedure of matrix generation in each of the matrix generation units 11 1, 1 17, 1 18, and 1 12 in FIG. 12 above This will be described with reference to the flow charts in FIGS. 16 to 20.
- FIG. 16 is a flowchart for explaining a procedure of generating the connection matrix A in the connection matrix generation unit 111.
- a first step S 12 parameters for matrix generation are initialized. The initialization of this parameter will be described later with reference to FIG.
- the inverse orthogonal transform matrix Ts (k) 1 is created.
- step S123 an arbitrary matrix of A and L rows and L columns is generated.
- step S124 all elements of the matrix A are initialized to 0.
- step S125 0 is set to 0. element and the matrices a 'diagonal the inverse orthogonal transform matrix on Ts (k) - 1 a and mi pieces arranged to produce a consolidated matrix a described above.
- the entrance Block length k! And output block length k Doo is if different, the inverse orthogonal transform matrix that is created by the Sutetsu flop S 1 2 2 becomes T s 1 is a matter of course.
- FIG. 17 is a flowchart showing an example of the procedure for initializing the parameters in step S121 of FIG.
- the block length of input and output is k
- the numerator a and denominator b of the above-mentioned resolution conversion magnification are set in steps S162 and S163, respectively.
- L bxk is set in step S164
- N aXk is set in step S165.
- This initialization of parameters is also used in the procedure of generating each matrix shown in FIGS. 18 to 20 described below.
- FIG. 18 is a flowchart for explaining the procedure of generating the selection matrix W in the selection matrix generation unit 117.
- the initialization of the parameters described in FIG. 17 is performed in the first step S 13 1, and the unit matrix I of L rows and L columns is generated in the next step S 13 2.
- the selection matrix W is set as this unit matrix I.
- FIG. 19 is a flow chart for explaining the procedure for generating the processing matrix X in the processing matrix generation unit 118.
- the parameters described in FIG. 17 are initialized, and in step S 142, an N ⁇ N inverse cosine transform matrix IDCT is obtained.
- an L-by-L cosine transformation matrix DCT is obtained.
- FIG. 20 is a flowchart for explaining a procedure of generating the partition matrix B in the partition matrix generation unit 112.
- initialization of the parameters described in FIG. 17 is performed in the first step S 15 1, and the inverse orthogonal transformation matrix T du 1 is created in step S 15 2.
- step S 1 5 3 generates an arbitrary matrix B ′ of N rows and N columns, initializes all elements of this matrix B ′ to 0 in step S 1 54, and in step S 1 5 5,
- the above-described inverse orthogonal transformation matrix T d ( k 1 is arranged mo on the diagonal, and a division matrix B is generated.
- the second embodiment can be easily extended to multidimensional signal processing.
- the processing may be performed first in the vertical direction, and then performed in the horizontal direction.
- any of the above-described resolution conversion processes can be extended to multidimensional signal processing, and can be used for resolution conversion of compressed image information such as so-called JPEG and MPEG.
- format conversion will be described.
- data may be stored in a partially irregular format for various reasons. For example, in the so-called DV (Digital Video) standard format, as shown in Fig. 21, a 4 x 8 pixel odd block at the right end of the chrominance signal can be created. It makes 8 8 blocks and decrypts it.
- DV Digital Video
- Such an irregular format can be processed without any problem by appropriately selecting the selection matrix W.
- FIG. 23 is a block diagram showing an example of a schematic configuration of a hardware for realizing the above-described processing, particularly when a digital image signal is handled.
- the CPU (Central Processing Unit) 101 for data calculation and signal processing is connected to the bus connected to the ROM 102, which stores programs and data, and is readable and writable.
- RAM, image interface 104, image output interface 106, recording medium 108 such as a hard disk, etc. are connected, and digital data is input from terminal 105 via image input / output interface 104.
- An image signal is input, and a digital image signal is output from the image output terminal 106 via the terminal 107.
- the program used for the processing according to the embodiment of the present invention is recorded on the recording medium 108 such as the ROM 102 or a hard disk and provided. It can also be provided via a disk, such as a CD-ROM, or via a communication line.
- video information captured by a home DV standard digital video camera 201 is converted to a digital video signal via an IEEE 1394 interface.
- the signal of the DV standard has an enormous amount of data and is not suitable for handling in the personal computer 202. Therefore, using the above-described embodiment of the present invention, this DV standard video data is converted into MPEG 1 standard video data. By doing so, the amount of data can be significantly reduced, and non-linear editing by software within the personal computer 202 can be performed, and images can be transferred to other devices through the so-called in-network 203. It is possible to provide the information to 204.
- the system conversion from the DV standard to the MPEG1 standard within the personal computer 202 is realized by the conversion processing unit 210 in FIG.
- the supplied DV video data is variable-length decoded by the variable-length decoding unit 211 in the conversion processing unit 210, and is inversely quantized by the inverse quantization unit 211. Then, it is sent to the resolution converter 213 as it is in the orthogonally transformed area or the compressed area, and is subjected to resolution conversion.
- the conversion process in the resolution conversion unit 21 is realized by the above-described embodiment. Also, after the motion prediction processing is performed on the area or the compressed area in the motion prediction section 214 which has been subjected to the orthogonal transformation, the motion is sent to the quantization section 215 and quantized.
- the data is subjected to variable-length encoding in 216 and extracted as MPEG-1 video data.
- the quantization unit 211 After the signal inversely quantized by the inverse quantization unit 211 is inversely orthogonally transformed into a time or space domain signal by the inverse DCT unit 211, The data format conversion is performed by the data conversion unit 222, the resolution conversion is performed by the resolution conversion unit 223, and the motion prediction is performed by the motion prediction unit 224 sequentially. Next, orthogonal transform is performed by the DCT unit 2 25 After transforming into an orthogonally transformed area or compressed area, the quantization unit
- the processing matrix can be determined without being affected by the signal I: intersection basis, and the design of the matrix is easy and can be applied to various signal processing.
- the inverse orthogonal transform matrix T s 1 for the inverse orthogonal transform and the orthogonal transform matrix T d for the orthogonal transform are generated, and the generated inverse orthogonal transform is generated.
- a transformation matrix D is generated using the transformation matrix T s ⁇ 1 and the orthogonal transformation matrix T d, and the input digitized signal in the orthogonally transformed area is subjected to the orthogonal transformation using the above transformation matrix D in the domain.
- the resolution and the transformation basis can be directly transformed in the orthogonally transformed area, and the decoding into the time domain and the spatial domain ( Inverse orthogonal transform) is not required, which simplifies calculations and enables high-quality conversion with small calculation errors.
- the hardware burden is reduced, and the conversion process can be realized at a sufficiently high speed by software.
- the input digital signal is a digital signal obtained by orthogonally transforming an original signal in a time domain or a space domain by an orthogonal transform matrix TS corresponding to the inverse orthogonal transform matrix T s ⁇ 1.
- output di di Yuru signals, the orthogonal transform matrix T d to the corresponding inverse orthogonal transform matrix T cT 1 by the decoded time domain and it is elevation Gerare that shall be converted into a signal in the spatial domain also,
- the original signal is orthogonally transformed by the orthogonal transformation matrix T s having a base length k for each transformation block of length k to obtain the input digital signal, and m input signals adjacent to the input digital signal are used.
- the input digital signal can be reduced to N / L in the orthogonally transformed area.
- N L
- the transformation matrix D is a matrix that transforms the transformation basis.
- orthogonal transformation matrices T s and T d are different, orthogonal transformation can be performed.
- an n-dimensional digit signal is used as the input digit signal, and the transform is performed in the orthogonally transformed region using the transform matrix D for each dimension.
- an n-dimensional output digital signal can be obtained.
- the conversion magnification is different for each dimension, a conversion matrix for each dimension is created, and the conversion processing is performed at a different magnification for each dimension using these conversion matrices.
- the conversion block of the input digital signal subjected to the orthogonal transformation is used as it is, and is converted to a block having a length of m times by filling a predetermined value.
- At least one block of a fixed number of samples of the input digitized signal in the orthogonally transformed area is connected to form a data block, and a connection matrix for inverse orthogonal transformation is generated.
- Generating a selection matrix for selecting and extracting a portion to be processed generating a processing matrix for performing a desired signal processing process on the selected portion, and orthogonalizing the signal processed signal.
- a conversion matrix to be applied is generated, and the above-mentioned conversion matrix is applied to the orthogonally converted input digital signal for each of the above-mentioned blocks to obtain an output digital signal. Since the original signal does not need to be restored, it can be processed at high speed, and the signal is not degraded due to the dissolution error. Conversion can be performed, and the signal block size can be converted to an arbitrary size. Also, only a part of the signal can be selectively processed by the selection matrix, and in that case, it is not necessary to be aware of the boundaries of the blocks of the input signal. Also, the processing matrix can be determined without being affected by the orthogonal transform basis of the signal, and the matrix design is easy and can be applied to various signal processing.
- the present invention is not limited to only the above-described embodiment.
- resolution conversion of an image signal subjected to orthogonal transformation has been mainly described.
- the present invention can be applied to digital signals in various frequency regions such as audio signals.
- the orthogonally transformed signal is not limited to a signal in the frequency domain, and a signal in the time domain or the spatial domain is used as the original signal.
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Signal Processing (AREA)
- Theoretical Computer Science (AREA)
- Mathematical Optimization (AREA)
- Mathematical Analysis (AREA)
- Pure & Applied Mathematics (AREA)
- Computational Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Software Systems (AREA)
- Databases & Information Systems (AREA)
- General Engineering & Computer Science (AREA)
- Algebra (AREA)
- Compression, Expansion, Code Conversion, And Decoders (AREA)
- Complex Calculations (AREA)
- Compression Or Coding Systems Of Tv Signals (AREA)
- Compression Of Band Width Or Redundancy In Fax (AREA)
Description
Claims
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| EP19980924648 EP0932106A1 (en) | 1997-06-16 | 1998-06-16 | Digital signal converting method and device, transformation matrix generating method and device, and supply medium |
| CA 2263454 CA2263454A1 (en) | 1997-06-16 | 1998-06-16 | Digital signal converting method and device, transformation matrix generating method and device, and supply medium |
Applications Claiming Priority (4)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| JP15904097 | 1997-06-16 | ||
| JP9/159040 | 1997-06-16 | ||
| JP9/238678 | 1997-09-03 | ||
| JP9238678A JPH1173410A (ja) | 1997-06-16 | 1997-09-03 | ディジタル信号変換方法及び装置並びに変換行列作成方法 |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| WO1998058328A1 true WO1998058328A1 (fr) | 1998-12-23 |
Family
ID=26485961
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| PCT/JP1998/002653 Ceased WO1998058328A1 (fr) | 1997-06-16 | 1998-06-16 | Procede et dispositif de conversion de signal numerique, dispositif et procede de generation de matrice de transformation et support d'alimentation |
Country Status (6)
| Country | Link |
|---|---|
| EP (1) | EP0932106A1 (ja) |
| JP (1) | JPH1173410A (ja) |
| KR (1) | KR20000068164A (ja) |
| CN (1) | CN1236456A (ja) |
| CA (1) | CA2263454A1 (ja) |
| WO (1) | WO1998058328A1 (ja) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| RU2346324C2 (ru) * | 2006-12-20 | 2009-02-10 | Николай Александрович Емельяненко | Устройство цифровой обработки полифазных ортогональных фазокодоманипулированных сигналов |
| US7630439B2 (en) | 2003-06-18 | 2009-12-08 | Kddi Corporation | Apparatus for converting conversion encoding coefficients using sampling/resolution conversion |
| WO2011009196A1 (en) * | 2009-07-20 | 2011-01-27 | Valorbec, Societe En Commandite | Finite dataset interpolation method |
Families Citing this family (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO1999023834A1 (fr) | 1997-11-05 | 1999-05-14 | Sony Corporation | Procede de conversion de signaux numeriques et appareil de conversion de signaux numeriques |
| EP1211896A3 (en) * | 2000-11-22 | 2006-04-19 | Matsushita Electric Industrial Co., Ltd. | System and method of providing data in one of a plurality of digital formats using an active disk |
| US8447121B2 (en) * | 2005-09-14 | 2013-05-21 | Microsoft Corporation | Efficient integrated digital video transcoding |
| JP2013098730A (ja) * | 2011-10-31 | 2013-05-20 | Toshiba Corp | 解像度変換装置、解像度変換方法及び動画像復号化装置 |
| JP2015061143A (ja) * | 2013-09-18 | 2015-03-30 | 沖電気工業株式会社 | 映像符号化装置、および、映像符号化プログラム |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH0447860A (ja) * | 1990-06-15 | 1992-02-18 | Nippon Telegr & Teleph Corp <Ntt> | 解像度変換処理方法 |
| JPH04229382A (ja) * | 1990-12-27 | 1992-08-18 | Ricoh Co Ltd | ディジタル画像データの解像度交換装置 |
| JPH05316357A (ja) * | 1992-05-14 | 1993-11-26 | Nippon Telegr & Teleph Corp <Ntt> | 解像度変換方法 |
| JPH0898173A (ja) * | 1994-09-21 | 1996-04-12 | Hitachi Ltd | 解像度変換装置およびそれを用いた解像度変換システム |
-
1997
- 1997-09-03 JP JP9238678A patent/JPH1173410A/ja not_active Withdrawn
-
1998
- 1998-06-16 WO PCT/JP1998/002653 patent/WO1998058328A1/ja not_active Ceased
- 1998-06-16 CN CN98801133A patent/CN1236456A/zh active Pending
- 1998-06-16 CA CA 2263454 patent/CA2263454A1/en not_active Abandoned
- 1998-06-16 KR KR1019997001259A patent/KR20000068164A/ko not_active Withdrawn
- 1998-06-16 EP EP19980924648 patent/EP0932106A1/en not_active Withdrawn
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH0447860A (ja) * | 1990-06-15 | 1992-02-18 | Nippon Telegr & Teleph Corp <Ntt> | 解像度変換処理方法 |
| JPH04229382A (ja) * | 1990-12-27 | 1992-08-18 | Ricoh Co Ltd | ディジタル画像データの解像度交換装置 |
| JPH05316357A (ja) * | 1992-05-14 | 1993-11-26 | Nippon Telegr & Teleph Corp <Ntt> | 解像度変換方法 |
| JPH0898173A (ja) * | 1994-09-21 | 1996-04-12 | Hitachi Ltd | 解像度変換装置およびそれを用いた解像度変換システム |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7630439B2 (en) | 2003-06-18 | 2009-12-08 | Kddi Corporation | Apparatus for converting conversion encoding coefficients using sampling/resolution conversion |
| RU2346324C2 (ru) * | 2006-12-20 | 2009-02-10 | Николай Александрович Емельяненко | Устройство цифровой обработки полифазных ортогональных фазокодоманипулированных сигналов |
| WO2011009196A1 (en) * | 2009-07-20 | 2011-01-27 | Valorbec, Societe En Commandite | Finite dataset interpolation method |
| US8582869B2 (en) | 2009-07-20 | 2013-11-12 | Concordia University | Finite dataset interpolation method |
Also Published As
| Publication number | Publication date |
|---|---|
| CA2263454A1 (en) | 1998-12-23 |
| KR20000068164A (ko) | 2000-11-25 |
| JPH1173410A (ja) | 1999-03-16 |
| EP0932106A1 (en) | 1999-07-28 |
| CN1236456A (zh) | 1999-11-24 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| JP4183205B2 (ja) | 高速アスペクト比ズーミング及び圧縮及びシーリングを用いるサムネイル操作 | |
| JP4002502B2 (ja) | 座標インタポレータの符号化/復号化装置及びその方法 | |
| US6693645B2 (en) | Optimized BIFS encoder | |
| KR20160045613A (ko) | 서브밴드 인코딩 된 이미지들로부터 전체 해상도 이미지를 저장, 처리 및 복원하는 방법 및 장치 | |
| JP2002500455A (ja) | 高速idct/ダウンサンプリング複合演算方法および装置 | |
| JP4688988B2 (ja) | ビデオデータの圧縮方法並びに装置、及び伸張方法並びに装置 | |
| JPH09322165A (ja) | 画像復号化装置とその方法、および、画像再生装置 | |
| WO1998058328A1 (fr) | Procede et dispositif de conversion de signal numerique, dispositif et procede de generation de matrice de transformation et support d'alimentation | |
| JP3701824B2 (ja) | データ処理方法、システム、装置、コンピュータ読取り可能媒体及びプログラム記憶装置 | |
| JP2001136527A (ja) | 直交変換画像の解像度変換装置及び方法 | |
| JPH08294119A (ja) | 画像符号化/復号化装置 | |
| JP3932244B2 (ja) | 画像符号/復号方法及びその装置並びにそのプログラムを記録した記録媒体 | |
| WO1993021731A1 (fr) | Appareil reproducteur d'images | |
| TW440804B (en) | Method and apparatus for resizing image information | |
| KR20000029715A (ko) | 화상매핑장치및방법과화상생성장치및방법 | |
| JP3403724B2 (ja) | 画像再生装置及び方法 | |
| JP4127465B2 (ja) | 画像伸張装置及び方法、並びに記録媒体 | |
| JP3198996B2 (ja) | 直交変換符号化画像の画像サイズ変換方法 | |
| JP2522357B2 (ja) | 画像の拡大方式 | |
| US7099523B2 (en) | Method and system for scaling a signal sample rate | |
| JP2005012460A (ja) | 変換符号化係数のサンプリング変換・解像度変換装置 | |
| KR100848816B1 (ko) | 정수 dct를 이용한 영상 이미지의 크기 변환 방법 | |
| JP3709106B2 (ja) | 画像圧縮および伸張装置 | |
| JP2006340300A (ja) | 信号処理方法及び信号処理装置、並びに信号処理プログラム及び情報記録媒体 | |
| JP3417761B2 (ja) | 画像処理装置 |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| WWE | Wipo information: entry into national phase |
Ref document number: 98801133.6 Country of ref document: CN |
|
| AK | Designated states |
Kind code of ref document: A1 Designated state(s): CA CN KR US |
|
| AL | Designated countries for regional patents |
Kind code of ref document: A1 Designated state(s): AT BE CH CY DE DK ES FI FR GB GR IE IT LU MC NL PT SE |
|
| ENP | Entry into the national phase |
Ref document number: 2263454 Country of ref document: CA Ref document number: 2263454 Country of ref document: CA Kind code of ref document: A |
|
| WWE | Wipo information: entry into national phase |
Ref document number: 1019997001259 Country of ref document: KR |
|
| WWE | Wipo information: entry into national phase |
Ref document number: 1998924648 Country of ref document: EP |
|
| 121 | Ep: the epo has been informed by wipo that ep was designated in this application | ||
| WWE | Wipo information: entry into national phase |
Ref document number: 09242300 Country of ref document: US |
|
| WWP | Wipo information: published in national office |
Ref document number: 1998924648 Country of ref document: EP |
|
| WWP | Wipo information: published in national office |
Ref document number: 1019997001259 Country of ref document: KR |
|
| WWW | Wipo information: withdrawn in national office |
Ref document number: 1998924648 Country of ref document: EP |
|
| WWW | Wipo information: withdrawn in national office |
Ref document number: 1019997001259 Country of ref document: KR |






