WO2006065078A1 - Apparatus for encoding and decoding image and method thereof - Google Patents
Apparatus for encoding and decoding image and method thereof Download PDFInfo
- Publication number
- WO2006065078A1 WO2006065078A1 PCT/KR2005/004310 KR2005004310W WO2006065078A1 WO 2006065078 A1 WO2006065078 A1 WO 2006065078A1 KR 2005004310 W KR2005004310 W KR 2005004310W WO 2006065078 A1 WO2006065078 A1 WO 2006065078A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- image
- unit
- encoding
- allowable noise
- quantization parameter
- Prior art date
Links
Classifications
-
- 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
- H04N19/30—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using hierarchical techniques, e.g. scalability
- H04N19/36—Scalability techniques involving formatting the layers as a function of picture distortion after decoding, e.g. signal-to-noise [SNR] scalability
-
- 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
-
- 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
- H04N19/129—Scanning of coding units, e.g. zig-zag scan of transform coefficients or flexible macroblock ordering [FMO]
-
- 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
- H04N19/134—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the element, parameter or criterion affecting or controlling the adaptive 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
- H04N19/10—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
- H04N19/134—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the element, parameter or criterion affecting or controlling the adaptive coding
- H04N19/154—Measured or subjectively estimated visual quality after decoding, e.g. measurement of distortion
-
- 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
- H04N19/169—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the coding unit, i.e. the structural portion or semantic portion of the video signal being the object or the subject of the adaptive coding
- H04N19/17—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the coding unit, i.e. the structural portion or semantic portion of the video signal being the object or the subject of the adaptive coding the unit being an image region, e.g. an object
- H04N19/176—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the coding unit, i.e. the structural portion or semantic portion of the video signal being the object or the subject of the adaptive coding the unit being an image region, e.g. an object the region being a block, e.g. a macroblock
-
- 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
- H04N19/169—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the coding unit, i.e. the structural portion or semantic portion of the video signal being the object or the subject of the adaptive coding
- H04N19/184—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the coding unit, i.e. the structural portion or semantic portion of the video signal being the object or the subject of the adaptive coding the unit being bits, e.g. of the compressed video stream
-
- 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
- H04N19/63—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using transform coding using sub-band based transform, e.g. wavelets
-
- 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
- H04N19/63—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using transform coding using sub-band based transform, e.g. wavelets
- H04N19/64—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using transform coding using sub-band based transform, e.g. wavelets characterised by ordering of coefficients or of bits for transmission
- H04N19/647—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using transform coding using sub-band based transform, e.g. wavelets characterised by ordering of coefficients or of bits for transmission using significance based coding, e.g. Embedded Zerotrees of Wavelets [EZW] or Set Partitioning in Hierarchical Trees [SPIHT]
Definitions
- the present invention relates to image encoding and decoding, and more particularly, to an image encoding and/or decoding apparatus for encoding an image with a quantization parameter determined by using an allowable noise having no subjective loss, and decoding the encoded bitstream, and a method thereof.
- the JND value is calculated by considering a relative change degree of neighboring pixel values for one pixel value, and models a characteristic that a human being cannot recognize a big noise in a complicated background.
- a JND value obtained in each pixel is used to additionally adjust a quantization parameter in a quantization process. Through this process, a region to which the human visual characteristic is sensitive is made to generate less noise, and a region to which the human visual characteristic is insensitive is made to generate more noise so that the compression ratio can be increased.
- the human visual characteristic in frequency domain can be considered. For this, a modulation transfer function of the human visual characteristic is empirically obtained from a variety of test images, and by using this, a quantization weight value of each frequency band for a discrete cosine transform (DCT) coefficient is extracted.
- DCT discrete cosine transform
- FIG. 1 is a block diagram showing a structure of an image encoding apparatus according to an embodiment of the present invention
- FIG. 2 illustrates a result of color quantization of leaves according to an embodiment of the present invention
- FIG. 3 illustrates error images between the original image and the color quantization results shown in FIG. 2 according to an embodiment of the present invention
- FIG. 4 illustrates allowable noises obtained in coding block units with respect to the original image in an allowable noise obtaining unit of FIG. 1 according to an embodiment of the present invention
- FIG. 5 illustrates an operation of a quantization parameter determination unit of FIG. 1 according to an embodiment of the present invention
- FIG. 6 illustrates a 3x3 window used in an entropy encoding unit of FIG. 1 in order to set a context according to an embodiment of the present invention
- FIG. 7 illustrates a lazy bit used in encoding of a bit plane in the entropy encoding unit of FIG. 1 according to an embodiment of the present invention
- FIG. 8 illustrates an operation of a scanning order determination unit of FIG. 1 according to an embodiment of the present invention
- FIG. 9 illustrates a scanning order depending on quality layers determined in FIG. 8 according to an embodiment of the present invention.
- FIG. 10 is a block diagram of a structure of an image decoding apparatus according to an embodiment of the present invention.
- FIGS. 1 1 (a) through (d) illustrate test images used to evaluate the performance of an image encoding and decoding algorithm according to an embodiment of the present invention
- FIGS. 12 (a) through (d) illustrate comparison of the image encoding and decoding algorithm according to an embodiment of the present invention with that of JPEG2000 in relation to the test image of FIG. 10 (a);
- FIGS. 13 (a) through (d) illustrate comparison of the image encoding and decoding algorithm according to an embodiment of the present invention with that of JPEG2000 in relation to the test image of FIG. 10 (b);
- FIGS. 14 (a) through (d) illustrate comparison of the image encoding and decoding algorithm according to an embodiment of the present invention with that of JPEG2000 in relation to the test image of FIG. 10 (c); and
- FIGS. 15 (a) through (d) illustrate comparison of the image encoding and decoding algorithm according to an embodiment of the present invention with that of JPEG2000 in relation to the test image of FIG. 10 (d).
- the present invention provides an image encoding and decoding method to encode an image with a quantization parameter determined by using an allowable noise without any subjective loss, and an image decoding apparatus and method thereof.
- the present invention also provides an image encoding and decoding method by which an image is quantized with a quantization parameter determined by using an allowable noise without any subjective loss, each coding block is divided into quality layers, and the quantized image is encoded in a scanning order of each quality layer, and an image decoding apparatus and method thereof.
- an image encoding apparatus including: a transform unit removing spatial redundancy by transforming an original image being input; an allowable noise obtaining unit obtaining an allowable noise from the original image; a quantization parameter determination unit determining a quantization parameter by using the allowable noise; a quantization unit generating a quantized coefficient, by quantizing a transform coefficient provided from the transform unit by using the quantization parameter; and an entropy encoding unit entropy encoding the quantized coefficient to remove statistical redundancy.
- an image encoding apparatus including: a transform unit removing spatial redundancy by transforming an original image being input; an allowable noise obtaining unit obtaining an allowable noise from the original image; a quantization parameter determination unit determining a quantization parameter by using the allowable noise; a quantization unit generating a quantized coefficient, by quantizing a transform coefficient provided from the transform unit by using the quantization parameter; a quality layer determination unit determining a quality layer by using the allowable noise; and an entropy encoding unit entropy encoding the quantized coefficient in the scanning order with respect to the quality layers to remove statistical redundancy.
- an image encoding method including: removing spatial redundancy by transforming an original image being input; obtaining an allowable noise from the original image; determining a quantization parameter by using the allowable noise; generating a quantized coefficient, by quantizing a transform coefficient provided from the transform unit by using the quantization parameter; and entropy encoding the quantized coefficient to remove statistical redundancy.
- an image encoding method including: removing spatial redundancy by transforming an original image being input; obtaining an allowable noise from the original image; determining a quantization parameter by using the allowable noise; generating a quantized coefficient, by quantizing a transform coefficient provided from the transform unit by using the quantization parameter; determining a scanning order with respect to the quality layers determined using the allowable noise; and entropy encoding the quantized coefficient in the scanning order with respect to the quality layers to remove statistical redundancy.
- an image decoding apparatus including: an entropy decoding unit generating a quantized coefficient, by entropy decoding an input bitstream; an inverse quantization unit generating a transform coefficient, by inverse quantizing the quantized coefficient; and an inverse transform unit generating a restored image, by inverse transforming the transform coefficient.
- an image decoding method including: generating a quantized coefficient, by entropy decoding an input bitstream; generating a transform coefficient, by inverse quantizing the quantized coefficient; and generating a restored image, by inverse transforming the transform coefficient.
- a computer readable recording medium having embodied thereon a computer program for executing the methods.
- an image is encoded and decoded by using a quantization parameter determined using an allowable noise without any subjective loss such that the encoding efficiency and the subjective picture quality of a restored image can be greatly improved. Furthermore, an image is quantized with a quantization parameter determined by using an allowable noise without any subjective loss, each coding block is divided into quality layers, and the quantized image is encoded in a scanning order of each quality layer. By doing so, a scalable function is provided and the encoding efficiency and the subjective picture quality of a restored image can be greatly improved.
- FIG. 1 is a block diagram showing a structure of an image encoding apparatus according to an embodiment of the present invention.
- the image encoding apparatus includes a transform unit 110, an allowable noise obtaining unit 130, a quantization parameter determination unit 150, a quantization unit 170, and an entropy encoding unit 190. Also, a scanning order determination unit 180 may be included optionally.
- the transform unit 110 is arranged to remove spatiotemporal redundancy in a frame image being input, and can use DCT, wavelet transform, prediction, or integer transform that is used in an image or video compression standard commercialized or set up.
- the transform method is not limited to these.
- the wavelet transform may be employed preferably.
- the allowable noise obtaining unit 130 is arranged to obtain an allowable noise from the input original image, and provides the obtained allowable noise to the quantization parameter determination unit 150.
- the allowable noise plays a role similar to that of a masking threshold value that is used in perception-based audio encoding.
- the allowable noise is obtained.
- the quantization parameter determination unit 150 is arranged to determine one quantization parameter for each coding block unit, by using the allowable noise obtained in the allowable noise obtaining unit 130, and provides the quantization parameter to the quantization unit 170. That is, the quantization parameter determination unit 150 determines the quantization parameter of each coding block so that the quantization noise generated as a result of quantization of the coding block can be equal to or less than the allowable noise.
- the quantization unit 170 is arranged to quantize the coding block by using the quantization parameter of the coding block determined in the quantization parameter determination unit 150, and provides the quantized transform coefficient of the coding block to the entropy encoding .unit 190.
- the scanning order determination unit 180 is arranged to determine a scanning order by using an arbitrary number of quality layers, each composed of bit planes determined by using an allowable noise in relation to the layer.
- the quality layer indicates a boundary that can influence the picture quality of a restored image, that is, the quality of the image, when a loss occurs in transmission of a generated by an embedded coding.
- the noise ( x ⁇ ) is determined by the color number (k) that can satisfy a first cost function guaranteeing a first quality.
- the first quality layer is determined, by using a second
- the second quality layer is determined, by using a third
- the number of quality layers and the cost function corresponding to each quality layer can be input in the scanning order determination unit 180 in advance in the initial stage of encoding. That is, the scanning order is from an upper quality layer towards a lower quality layer.
- the entropy encoding unit 190 generates a bitstream, by removing statistical redundancy in the quantized transform coefficient of each coding block provided by the quantization unit 170. At this time, at the top bitrate, with scanning data in units of coding block, entropy encoding is performed. Meanwhile, at a sub bitrate, entropy encoding is performed by scanning data in units of quality layers according to the scanning order determined in the scanning order determination unit 180.
- the entropy encoding unit 190 may perform embedded coding in order to provide a scalable function. At this time, a bit sliced arithmetic coding (BSAC) algorithm used in an MPEG-4 audio codec providing a fine granular scalability (FGS) function is applied.
- BSAC bit sliced arithmetic coding
- FGS fine granular scalability
- FIG. 2 illustrates a result of color quantization of leaves according to an embodiment of the present invention.
- FIG. 2 (a) shows an original image
- FIG. 2 (b) shows the result of color quantization with color number 10
- FIG. 2 (c) shows the result of color quantization with color number 30,
- FIG. 2 (d) shows the result of color quantization with color number 41.
- FIG. 3 illustrates error images between the original image and the color quantization results shown in FIG.
- FIG. 3 (a) shows the original image
- FIG. 3 (b) shows an error image between the original image and the image color quantized with color number 10
- FIG. 3 (c) shows an error image between the original image and the image color quantized with color number 30
- FIG. 3 (d) shows an error image between the original image and the image color quantized with color number 40.
- each error image is obtained by adding an offset of 128 to each of R, G, and B colors.
- a pixel In an image that is generally widely used, a pixel is composed of three components, R, G, and B, each expressed by 8 bits. In this case, the total number of colors expressing an image is 2 24 , which means the number of colors that can be expressed is very large.
- human perception cannot distinguish each of this number of colors, and this perception changes in a variety of ways according to the color distribution of neighboring pixels as well as the color value itself.
- the color quantization algorithm there is one using Fibonacci Lattice and the characteristic that a human visual distance in relation to color relatively matches a numerical distance in a Lab space (Aleksandra Mojsilovic and Emina Soljanin, "Color Quantization and Processing by Fibonacci Lattices", IEEE Trans.
- mapping table for mapping quantized values to actual pixel values, and an index image composed of quantized pixels are compressed through entropy coding (Zhu Li and Aggelos K. Katsaggelos, "Color Vector Quantization based Video Coder", IEEE International Conference on Image Processing, vol. 3, pp.673-676, Oct. 2002).
- the allowable noise obtaining unit 130 obtains an allowable noise, by using the result of color quantization of the original image in units of frames or in units of predetermined blocks. This will now be explained in more detail. It is assumed that the original image is ⁇ (x, y) (hereinafter referred to as ⁇ ), and an image which is color quantized with color number k is ⁇ (x, y) (hereinafter referred to as ⁇ k ), wherein it is assumed that ⁇ and ⁇ k are visually identical.
- ⁇ and ⁇ k in relation to an arbitrary integer k can be expressed as the following equation 1 :
- x ⁇ is an error image of the original image ( ⁇ ) and the image ⁇ k ) color quantized with color number /c.
- equation 2 there may be a variety of ways to find integer h.
- an appropriate cost function that can satisfy equation 2 is defined and by gradually increasing the color number from 2 to n, a minimum color number that the error image of the original image and a color quantized image can satisfy the cost function is determined.
- the error image ( T/ ⁇ ) is defined as an allowable noise, and when a noise equal to or less than the allowable noise is included, the original image and the quantized image are identical. That is, there is no subjective loss between the two images.
- the allowable noise is obtained by dividing respective pixel values or their absolute
- FIG. 4 illustrates allowable noises obtained in units of pixels in relation to an original image in the allowable noise obtaining unit 130 of FIG. 1.
- absolute values of pixel values in the error image ( X ⁇ ) are taken and expressed in units of coding blocks in a graph.
- FIG. 5 illustrates an operation of the quantization parameter determination unit 150 of FIG. 1 according to an embodiment of the present invention.
- One quantization parameter is determined for each coding block.
- a coding block Bo has a quantization parameter Qpo
- a coding block Bs has a quantization parameter Qp 8
- a coding block B-i ⁇ has a quantization parameter Qpi6-
- an image (a) is quantized by using a quantization parameter [Qp)
- an image (a) is restored through inverse quantization by using the identical quantization parameter (Qp).
- the restored image (a) can be expressed as the following equation 4:
- ⁇ is a function indicating a process in which the original image ( ⁇ ) is quantized by using the quantization parameter (Qp), and then is inverse quantized by using the identical quantization parameter (Qp).
- the quantization and the inverse quantization can be performed by applying a variety of algorithms. For example, a uniform quantization algorithm having a dead zone can be applied.
- quantization and inverse quantization can be expressed as the following equations 5 and 6. Equations 5 and 6 are designed to perform quantization by dividing a value that is obtained by doubling a quantization parameter, into six equal parts:
- X 1 s ⁇ gn(X) * ⁇ X ⁇ * 2 Qp%' ⁇ ) + 0.5 ⁇ (5)
- quantization parameter (Qp) Since the quantization parameter (Qp) is information to be encoded, encoding of the quantization parameters (Qp) for all pixels respectively is very inefficient.
- the original image ( ⁇ ) is divided into block units of a predetermined size, for example, coding block units, and a quantization parameter is determined for each coding block unit.
- a quantization parameter of the coding block cr & is Qp b
- * and ⁇ b are the allowable noise of a coding block existing in an entire error
- the quantization parameter (Qp b ) of each coding block a method of gradually increasing from a smaller value to a bigger value with a predetermined variation step is employed, but the present invention is not necessarily limited to this.
- the initial value of the quantization parameter (Qpb) of a coding block and the variation steps have a trade-off relation of the amount of calculation and accuracy.
- the thick line marked for each codling block (B 0 ⁇ Bi 7 ) indicates a limit value of an allowable noise set for the coding block.
- the quantization parameter of each coding block is determined so that a quantization noise equal to or less than the allowable noise limit value set to each coding block can be generated.
- the image encoding using an allowable noise can be applied to a variety of types of encoders. For example, it is assumed that a transform process and an inverse transform process used in the encoding are defined as the following equations 11 and 12, respectively:
- equation 13 can be expressed as the following equation 14:
- equation 14 can be expressed as the following equation 15:
- the entropy encoding unit 190 performs entropy encoding through embedded coding using context-based adaptive arithmetic encoding.
- a context is set for the context-based adaptive arithmetic encoding.
- the probability value of a current symbol is predicted from the processed context of data, and is applied to the arithmetic encoding. Accordingly, it is needed to explain the context so that the probability value can be predicted appropriately.
- selection of a context a context having a high correlation with the current symbol is selected.
- the transform unit 110 may perform multiple layer wavelet filtering. Since each layer provides information on different resolution together with information on the upper layer, the multiple layer wavelet filtering can provide spatial scalability.
- the top layer is composed of only a low-low (LL) band, and the remaining lower layers are composed of low-high (LH), high-low (HL), and high-high (HH) bands, respectively.
- LH low-low
- HL high-low
- HH high-high
- a coding block is defined.
- the size of the coding block may vary with respect to the layer, and when encoding is performed, the coding block can be defined with an encoding parameter.
- the coding block can be used as a unit for determining a quantization parameter, a unit for providing information on a most significant bit (MSB) in relation to a quantized coefficient value, a unit for providing information on a lazy bit, and a unit for scanning a block.
- the entropy encoding unit 190 performs bit plane encoding in the block scanning unit, that is, in units of coding blocks. That is, by scanning bits of pixels from the MSB to a least significant bit (LSB) in a scanning line order, the encoding is performed.
- LSB least significant bit
- the context-based adaptive arithmetic encoding applied to the entropy encoding unit 190 will now be explained in more detail.
- a symbol When a smallest input unit for entropy encoding in relation to a quantized coefficient is referred to as a symbol, the symbol has any one value of 0 and 1 , because the BSAC algorithm is for bit plane coding for quantized coefficients. Accordingly, the BSAC algorithm applied to the present invention is basically binary arithmetic encoding.
- As an element that should be considered in order to select a context for each symbol there is the correlation of each context with each symbol.
- the probability value of the symbol can be predicted more accurately, and according to the accurate probability value, the compression efficiency can be increased when arithmetic encoding is performed.
- the causability of a context since a context that is used in an encoder should be composed of values identical to those of a context to be used in a decoder, the context is composed of symbols, or additional information items already processed and encoded. Meanwhile, as another element that should be considered with the correlation in order to select a context, there is complexity. Since the BSAC algorithm regards all bits from the
- the number of symbols to be processed is very large. If the types of contexts for each symbol are too many, the compression efficiency can be increased, but it can impose a heavy burden in the aspects of memory capacity and processing speed in the encoder and decoder. In particular, in case of the adaptive arithmetic encoding, excessively various contexts can decrease the adaptive speed in relation to a new bit pattern such that the compression efficiency can be lowered.
- the entropy encoding unit 190 sets, for example, a 3x3 window centered at the pixel of each symbol in one coding block. According to this, a context in relation to pixels included in one window can be built.
- FIG. 6 illustrates a 3x3 window applied to an arbitrary coding block in order to set a context in the entropy encoding unit 190 of FIG. 1.
- L is a lazy bit
- V 1 is the number of 1's occurring in a (j+1)-th bit plane
- W j is a bit pattern already processed in a j-th bit plane.
- the lazy bit (L) is determined according to the following equation
- m is the mean of pixels included in a coding block
- MSB(x) is a top bit position having value 1 in a bit plane expressing x.
- FIG. 7 illustrates a context used in encoding of a bit plane in the entropy encoding unit 190 of FIG. 1 according to an embodiment of the present invention.
- calculation of the context of a symbol 710 included in the j-th bit plane is determined by using bits 720 of the G + IHh bit plane immediately above, and 4 bit patterns 710 already processed in the j-th bit plane.
- the probability value is set to a fixed value, 1/2. Since a separate context calculation for symbols in a bit plane lower than the lazy bit (L) is not performed, the amount of calculation can be reduced as much. If there are a variety of coefficients quantized with different quantization parameter values, the using of the lazy bit provides a relative bit plane value (j-L) such that the encoding efficiency can be greatly improved.
- FIG. 8 illustrates an operation of the scanning order determination unit 180 of FIG. 1 according to an embodiment of the present invention.
- the entropy encoding unit 190 a restored image in which noise is subjectively minimized can be obtained in the top bitrate. Also, since spatial and SNR scalability is provided through the context-based adaptive arithmetic encoding, a bitstream, once generated, can be truncated as required with respect to a purpose, and can be used for reproduction. A problem of optimization of picture quality occurring when part of a bitstream is thus used will now be explained.
- the entropy encoding unit 190 at the top bitrate, data is scanned in units of coding blocks according to the context-based adaptive arithmetic encoding, and each coding block is scanned in the scan line direction in relation to each bit plane.
- the scanning order of a coding block, the scanning order of pixels in a coding block, and the scanning order in a bit plane can be considered.
- the aspect of the amount of memory use or complexity should be considered together.
- a table setting a scanning order in order to scan an important coding block and important pixels with priority can be used or the scanning order can be changed adaptively.
- a quality layer is determined in relation to the quantized coefficients.
- the quality layer is a method employed in an embedded block coding with optimized truncation algorithm (EBCOT) (David Taubman, "High Performance Scalable Image Compression with EBCOT", IEEE Trans. Image Processing, vol. 9, pp. 1158-1170, July 2000).
- EBCOT embedded block coding with optimized truncation algorithm
- an arbitrary number of quality layers for bit plane encoding are disposed.
- One quality layer is composed of bit planes determined by using an allowable noise in relation to the layer. Among the constructed quality layers, scan is performed from the top quality layer.
- V V 0 V 111 -V n- I
- images color quantized from the original image are obtained first.
- These color quantized images ⁇ o, ⁇ i, ..., ⁇ n-1 have the characteristic as the following equation 18:
- ⁇ ( ⁇ ) is the number of colors in image ⁇ .
- An allowable noise (r,) in relation to each of the color quantized images is obtained according to the following equation 19:
- FIG. 10 is a block diagram of a structure of an image decoding apparatus according to an embodiment of the present invention.
- the decoding apparatus includes an entropy decoding unit 1010, an inverse quantization unit 1030, and an inverse transform unit 1050.
- the entropy decoding unit generates a quantized coefficient, by entropy decoding an input bitstream.
- the entropy decoding process corresponds to an inverse process of the entropy encoding process of the image encoding apparatus shown in FIG. 1.
- the inverse quantization unit 1030 generates a transform coefficient, by inverse quantizing the quantized coefficient provided by the entropy decoding unit 1010.
- the inverse quantization process corresponds to an inverse process of the quantization process of the image encoding apparatus shown in FIG. 1.
- the inverse transform unit 1050 generates a restored image, by inverse transforming the transform coefficient provided by the inverse quantization unit 1030.
- the inverse transform process corresponds to an inverse process of the transform process of the image encoding apparatus shown in FIG. 1.
- the test image had a 512x512 size, and YUV444 format for color was used.
- a 9/7 filter was used for the Y component, 5/3 filters were used for U and V components.
- the coefficient values of the used filters were the same as those for JPEG2000. Meanwhile, all wavelet filters were implemented to have a lifting structure.
- the wavelet decomposition could be performed to the fifth steps and spatial scalable images of a total of 6 steps could be restored.
- coding blocks were set as the following table 1 :
- JPEG2000 software used for the comparison experiment is reference software disclosed at the JPEG2000 website (http://www.jpeg.org/jpeg2000/ index.html), and is Jasper version 1.701.0.
- the encoder option was set as a default and the experiments were performed.
- FIGS. 11 (a) through (d) illustrate examples of the test images, a baboon, clips, a parrot, and leaves, respectively.
- Each image has a 512x512 size and each of the R, G, and B components is 8 bits long.
- the R-D optimization process in the MSE concept was not performed when the encoding and decoding algorithm according to the present invention were performed, the R-D optimization process in the MSE was performed in the JPEG2000.
- the image decoded according to the JPEG2000 shows a value approximately O. ⁇ dB higher with respect to the types of the test image and the compression ratio.
- the following table 2 shows comparison of PSNR values of the images decoded by the present invention and the JPEG2000, respectively, and shows that there was a PSNR difference in the data:
- FIG. 12 (a) and (c) are images obtained through encoding and decoding at compression ratios of 20:1 and 50:1 , respectively, when the present invention was applied to a baboon
- FIG. 12 (b) and (d) are images obtained through encoding and decoding at compression ratios of 20:1 and 50:1 , respectively, when the JPEG2000 was applied to the baboon
- FIG. 13 (a) and (c) are images obtained through encoding and decoding at compression ratios of 20:1 and 50:1 , respectively, when the present invention was applied to clips
- FIG. 13 (b) and (d) are images obtained through encoding and decoding at compression ratios of 20:1 and 50:1 , respectively, when the JPEG2000 was applied to the clips.
- FIG. 14 (a) and (c) are images obtained through encoding and decoding at compression ratios of 20:1 and 50.1 , respectively, when the present invention was applied to a parrot
- FIG. 14 (b) and (d) are images obtained through encoding and decoding at compression ratios of 20:1 and 50:1 , respectively, when the JPEG2000 was applied to the parrot.
- FIG. 15(a) and (c) are images obtained through encoding and decoding at compression ratios of 20:1 and 50:1 , respectively, when the present invention was applied to leaves
- FIGS. 12 through 15 confirm that the images decoded according to the present invention and the JPEG2000 are subjectively identical or similar images.
- the present invention can also be embodied as computer readable codes on a computer readable recording medium.
- the computer readable recording medium is any data storage device that can store data which can be thereafter read by a computer system. Examples of the computer readable recording medium include read-only memory (ROM), random-access memory (RAM), CD-ROMs, magnetic tapes, floppy disks, optical data storage devices, and carrier waves (such as data transmission through the Internet).
- ROM read-only memory
- RAM random-access memory
- CD-ROMs compact discs
- magnetic tapes magnetic tapes
- floppy disks optical data storage devices
- carrier waves such as data transmission through the Internet
- carrier waves such as data transmission through the Internet
- the computer readable recording medium can also be distributed over network coupled computer systems so that the computer readable code is stored and executed in a distributed fashion. Also, functional programs, codes, and code segments for accomplishing the present invention can be easily construed by programmers skilled in the art to which the present invention pertains.
Landscapes
- Engineering & Computer Science (AREA)
- Multimedia (AREA)
- Signal Processing (AREA)
- Compression Or Coding Systems Of Tv Signals (AREA)
- Image Processing (AREA)
- Compression Of Band Width Or Redundancy In Fax (AREA)
- Compression, Expansion, Code Conversion, And Decoders (AREA)
Abstract
Description
Claims
Priority Applications (5)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
KR1020077013456A KR100928968B1 (en) | 2004-12-14 | 2005-12-14 | Image encoding and decoding apparatus and method |
EP05822193.8A EP1825687B1 (en) | 2004-12-14 | 2005-12-14 | Determining quantization parameters using color quantization |
JP2007545388A JP5001853B2 (en) | 2004-12-14 | 2005-12-14 | Video encoding and decoding apparatus and method |
US11/793,147 US9008451B2 (en) | 2004-12-14 | 2005-12-14 | Apparatus for encoding and decoding image and method thereof |
CN2005800430028A CN101080931B (en) | 2004-12-14 | 2005-12-14 | Apparatus for encoding and decoding image and method thereof |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
KR10-2004-0105383 | 2004-12-14 | ||
KR20040105383 | 2004-12-14 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2006065078A1 true WO2006065078A1 (en) | 2006-06-22 |
Family
ID=36588098
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/KR2005/004310 WO2006065078A1 (en) | 2004-12-14 | 2005-12-14 | Apparatus for encoding and decoding image and method thereof |
Country Status (6)
Country | Link |
---|---|
US (1) | US9008451B2 (en) |
EP (1) | EP1825687B1 (en) |
JP (1) | JP5001853B2 (en) |
KR (1) | KR100928968B1 (en) |
CN (1) | CN101080931B (en) |
WO (1) | WO2006065078A1 (en) |
Families Citing this family (17)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
AU2004321469B2 (en) * | 2004-07-14 | 2009-08-20 | Agency For Science, Technology And Research | Context-based encoding and decoding of signals |
US7684632B2 (en) * | 2005-05-16 | 2010-03-23 | Hewlett-Packard Development Company, L.P. | Estimating image compression quantization parameter values |
US8126054B2 (en) * | 2008-01-09 | 2012-02-28 | Motorola Mobility, Inc. | Method and apparatus for highly scalable intraframe video coding |
KR101359500B1 (en) * | 2008-07-28 | 2014-02-11 | 에스케이 텔레콤주식회사 | Quantization/Inverse Quantization Apparatus and Method and Video Encoding/Decoding Apparatus Using Same |
CN101686383B (en) * | 2008-09-23 | 2013-05-01 | Utc消防和保安美国有限公司 | Method and system for transmitting medium through network |
KR101624649B1 (en) * | 2009-08-14 | 2016-05-26 | 삼성전자주식회사 | Method and apparatus for video encoding considering hierarchical coded block pattern, and method and apparatus for video decoding considering hierarchical coded block pattern |
KR101487687B1 (en) | 2010-01-14 | 2015-01-29 | 삼성전자주식회사 | Method and apparatus for encoding and decoding image using large transform unit |
CN102934428B (en) * | 2010-12-16 | 2015-08-05 | 北京航空航天大学 | The wavelet coefficient quantization method of human vision model is utilized in a kind of image compression |
KR102106534B1 (en) * | 2011-06-28 | 2020-05-04 | 삼성전자주식회사 | Method and apparatus for entropy coding and decoding |
EP2828822B1 (en) * | 2012-03-21 | 2018-07-11 | Dolby Laboratories Licensing Corporation | Systems and methods for power reduction for displays |
US9819964B2 (en) * | 2012-05-04 | 2017-11-14 | Environmental Systems Research Institute, Inc. | Limited error raster compression |
US9002126B2 (en) * | 2012-05-04 | 2015-04-07 | Environmental Systems Research Institute (ESRI) | Limited error raster compression |
US10085017B2 (en) * | 2012-11-29 | 2018-09-25 | Advanced Micro Devices, Inc. | Bandwidth saving architecture for scalable video coding spatial mode |
CN104782121A (en) * | 2012-12-18 | 2015-07-15 | 英特尔公司 | Multiple region video conference encoding |
CN105872539B (en) * | 2015-02-08 | 2020-01-14 | 同济大学 | Image encoding method and apparatus, and image decoding method and apparatus |
US10163030B2 (en) * | 2016-05-20 | 2018-12-25 | Gopro, Inc. | On-camera image processing based on image activity data |
DE102019204527B4 (en) * | 2019-03-29 | 2020-11-19 | Technische Universität München | CODING / DECODING DEVICES AND METHODS FOR CODING / DECODING VIBROTACTILE SIGNALS |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH04372074A (en) * | 1991-06-21 | 1992-12-25 | Nippon Telegr & Teleph Corp <Ntt> | Post filter |
JPH0774954A (en) * | 1993-09-01 | 1995-03-17 | Nippon Telegr & Teleph Corp <Ntt> | Method for encoding/decoding picture information |
JPH0775102A (en) * | 1993-07-19 | 1995-03-17 | Sharp Corp | Image encoder |
JP2003230023A (en) * | 2002-02-01 | 2003-08-15 | Mitsubishi Electric Corp | Aperture correction method for network camera and the network camera |
JP2003248838A (en) * | 2001-11-27 | 2003-09-05 | Samsung Electronics Co Ltd | Method and device for encoding and decoding key value data of coordinate interpolator |
Family Cites Families (15)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CA2014935C (en) | 1989-05-04 | 1996-02-06 | James D. Johnston | Perceptually-adapted image coding system |
US5426512A (en) * | 1994-01-25 | 1995-06-20 | The United States Of America As Represented By The Administrator Of The National Aeronautics And Space Administration | Image data compression having minimum perceptual error |
US5629780A (en) * | 1994-12-19 | 1997-05-13 | The United States Of America As Represented By The Administrator Of The National Aeronautics And Space Administration | Image data compression having minimum perceptual error |
JP2000506687A (en) * | 1996-03-29 | 2000-05-30 | サーノフ コーポレイション | Apparatus and method for optimizing encoding using perceptual amount and performing automatically operable image compression |
AU2001276588A1 (en) * | 2001-01-11 | 2002-07-24 | K. P. P. Kalyan Chakravarthy | Adaptive-block-length audio coder |
DE10300048B4 (en) * | 2002-01-05 | 2005-05-12 | Samsung Electronics Co., Ltd., Suwon | Image coding method for motion picture expert groups, involves image quantizing data in accordance with quantization parameter, and coding entropy of quantized image data using entropy coding unit |
US20030215013A1 (en) * | 2002-04-10 | 2003-11-20 | Budnikov Dmitry N. | Audio encoder with adaptive short window grouping |
US6961736B1 (en) * | 2002-05-31 | 2005-11-01 | Adobe Systems Incorporated | Compact color feature vector representation |
US7302105B2 (en) * | 2002-07-22 | 2007-11-27 | Canon Kabushiki Kaisha | Moving image coding apparatus, moving image decoding apparatus, and methods therefor |
GB0228556D0 (en) | 2002-12-06 | 2003-01-15 | British Telecomm | Video quality measurement |
AU2003900531A0 (en) * | 2003-01-30 | 2003-02-20 | Canon Kabushiki Kaisha | Compression into a fixed buffer |
US7318027B2 (en) * | 2003-02-06 | 2008-01-08 | Dolby Laboratories Licensing Corporation | Conversion of synthesized spectral components for encoding and low-complexity transcoding |
CN1461112A (en) * | 2003-07-04 | 2003-12-10 | 北京阜国数字技术有限公司 | Quantized voice-frequency coding method based on minimized global noise masking ratio criterion and entropy coding |
ES2305852T3 (en) * | 2003-10-10 | 2008-11-01 | Agency For Science, Technology And Research | PROCEDURE FOR CODING A DIGITAL SIGNAL IN A SCALABLE BINARY FLOW, PROCEDURE FOR DECODING A SCALABLE BINARY FLOW. |
AU2004321469B2 (en) * | 2004-07-14 | 2009-08-20 | Agency For Science, Technology And Research | Context-based encoding and decoding of signals |
-
2005
- 2005-12-14 JP JP2007545388A patent/JP5001853B2/en active Active
- 2005-12-14 US US11/793,147 patent/US9008451B2/en active Active
- 2005-12-14 KR KR1020077013456A patent/KR100928968B1/en active IP Right Grant
- 2005-12-14 EP EP05822193.8A patent/EP1825687B1/en active Active
- 2005-12-14 WO PCT/KR2005/004310 patent/WO2006065078A1/en active Application Filing
- 2005-12-14 CN CN2005800430028A patent/CN101080931B/en active Active
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH04372074A (en) * | 1991-06-21 | 1992-12-25 | Nippon Telegr & Teleph Corp <Ntt> | Post filter |
JPH0775102A (en) * | 1993-07-19 | 1995-03-17 | Sharp Corp | Image encoder |
JPH0774954A (en) * | 1993-09-01 | 1995-03-17 | Nippon Telegr & Teleph Corp <Ntt> | Method for encoding/decoding picture information |
JP2003248838A (en) * | 2001-11-27 | 2003-09-05 | Samsung Electronics Co Ltd | Method and device for encoding and decoding key value data of coordinate interpolator |
JP2003230023A (en) * | 2002-02-01 | 2003-08-15 | Mitsubishi Electric Corp | Aperture correction method for network camera and the network camera |
Non-Patent Citations (1)
Title |
---|
See also references of EP1825687A4 * |
Also Published As
Publication number | Publication date |
---|---|
EP1825687B1 (en) | 2015-10-28 |
JP2008523706A (en) | 2008-07-03 |
US20080131014A1 (en) | 2008-06-05 |
EP1825687A1 (en) | 2007-08-29 |
KR20070090191A (en) | 2007-09-05 |
CN101080931A (en) | 2007-11-28 |
KR100928968B1 (en) | 2009-11-26 |
EP1825687A4 (en) | 2011-03-16 |
CN101080931B (en) | 2010-06-16 |
JP5001853B2 (en) | 2012-08-15 |
US9008451B2 (en) | 2015-04-14 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US9008451B2 (en) | Apparatus for encoding and decoding image and method thereof | |
CN1232123C (en) | Method of and system for activity-based frequency weighting for FGS enhancement layers | |
JP5838258B2 (en) | Method and apparatus for lossy compression encoding data and corresponding method and apparatus for reconstructing data | |
EP1305953B1 (en) | Signal compression and decompression | |
KR101247011B1 (en) | Adaptive coding and decoding of wide-range coefficients | |
EP1788820A2 (en) | Image encoding method and apparatus and image decoding method and apparatus using characteristics of the human visual system | |
US20150043630A1 (en) | Image encoding device, image decoding device, image encoding method, and image decoding method | |
KR100529311B1 (en) | Apparatus and method for selecting the length of variable length coding bit stream using neural network | |
US5719961A (en) | Adaptive technique for encoder and decoder signal transformation | |
JP2002501687A (en) | Embedded image encoder with rate-distortion optimization | |
WO2005112434A2 (en) | Noise filter for video processing | |
JP2004032718A (en) | System and method for processing video frame by fading estimation/compensation | |
WO2020261314A1 (en) | Image encoding method and image decoding method | |
IL288888B1 (en) | Encoder and method of encoding a sequence of frames | |
JP2004528791A (en) | Inter-frame encoding method and apparatus | |
Li et al. | Layered DCT still image compression | |
KR100909373B1 (en) | Jpec 2000 Compressed Image Quality Evaluation Method Using Context Information | |
JP3853708B2 (en) | Digital image encoding apparatus, encoding method, and program | |
JP2004266794A (en) | Multi-resolution video coding and decoding | |
US7613347B1 (en) | Low-complexity image coding | |
PL | A Study on Image Compression Technology | |
CN115486070A (en) | Entropy encoded transform coefficient ordering | |
WO2003094523A2 (en) | Snr scalable quantizer | |
Kumar et al. | A survey of contemporary medical image compression techniques | |
Arun Sankar et al. | Performance Analysis of Wavelet Based Transforms in JPEG |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
AK | Designated states |
Kind code of ref document: A1 Designated state(s): AE AG AL AM AT AU AZ BA BB BG BR BW BY BZ CA CH CN CO CR CU CZ DE DK DM DZ EC EE EG ES FI GB GD GE GH GM HR HU ID IL IN IS JP KE KG KM KN KP KR KZ LC LK LR LS LT LU LV LY MA MD MG MK MN MW MX MZ NA NG NI NO NZ OM PG PH PL PT RO RU SC SD SE SG SK SL SM SY TJ TM TN TR TT TZ UA UG US UZ VC VN YU ZA ZM ZW |
|
AL | Designated countries for regional patents |
Kind code of ref document: A1 Designated state(s): GM KE LS MW MZ NA SD SL SZ TZ UG ZM ZW AM AZ BY KG KZ MD RU TJ TM AT BE BG CH CY CZ DE DK EE ES FI FR GB GR HU IE IS IT LT LU LV MC NL PL PT RO SE SI SK TR BF BJ CF CG CI CM GA GN GQ GW ML MR NE SN TD TG |
|
121 | Ep: the epo has been informed by wipo that ep was designated in this application | ||
DPE2 | Request for preliminary examination filed before expiration of 19th month from priority date (pct application filed from 20040101) | ||
WWE | Wipo information: entry into national phase |
Ref document number: 2005822193 Country of ref document: EP |
|
WWE | Wipo information: entry into national phase |
Ref document number: 2007545388 Country of ref document: JP |
|
WWE | Wipo information: entry into national phase |
Ref document number: 11793147 Country of ref document: US Ref document number: 200580043002.8 Country of ref document: CN Ref document number: 1020077013456 Country of ref document: KR |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
WWP | Wipo information: published in national office |
Ref document number: 2005822193 Country of ref document: EP |