`
`DECLARATION OF TRANSLATOR
`
`
`
`I, Corey Colling hereby declare as follows:
`
`1.
`
`My name is Corey Colling. I provide this declaration on behalf of
`
`Park IP at the request of Unified Patents, LLC.
`
`2.
`
`I am a professional translator, fluent in Korean and English. I work as
`
`a contractor for Park IP.
`
`3.
`
`I reviewed the original Korean version of Korean Patent KR 0135364
`
`BI and prepared the English translation attached hereto. The English translation is a true,
`
`complete, and accurate translations of KR 0135364 BI.
`
`4.
`
`In signing this declaration, I recognize that the declaration will be filed
`
`as evidence in a case before the Patent Trial and Appeal Board of the United States Patent
`
`and Trademark Office. I also recognize that I may be subject to cross-examination in the
`
`case and that cross-examination will take place within the United States. If cross-
`
`examination is required of me, I will appear for cross-examination within the United States
`
`during the time allotted for cross-examination.
`
`5.
`
`All statements made herein of my own knowledge are true and all
`
`statements made on information and belief are believed to be true. These statements
`
`were made with the knowledge that willful false statements and the like so made are
`
`punishable by fine or imprisonment, or both (18 U.S.C. § 1001).
`
`
`
`Dated: December 28, 2020
`
`
`
`
`
`
`
`Corey Colling
`
`15 W. 37th Street 8th Floor
`New York, NY 10018
`212.581.8870
`ParkIP.com
`
`Unified Patents, LLC v. Elects. & Telecomm. Res. Inst., et al.
`
`Ex. 1010, p. 1
`
`
`
`(51) . Int. Cl. 6
`H04N 7/133
`(21) Application No.
`(22) Application date
`
`(73) Assignee(s)
`
`(72) Inventor(s)
`
`1019940025774
`
`(19) Korean Intellectual Property Office (KR)
`(12) Publication of Registration (B1)
`
`
`(11) Registration No. Patent 0135364
`(24) Registration date
`January 13, 1998
`(65) Publication No.
`Patent 1996-016550
`(43) Publication date
`May 22, 1996
`
`HONG, Doo-Pyo
`
`Patent 1994-025774
`October 8, 1994
`
`KOREAN BROADCASTING SYSTEM
`18, Yeouido-dong, Yeongdeungpo-gu, Seoul
`LEE, Jong-Hwa
`1460-2, Sillim 5-dong, Gwanak-gu, Seoul
`MOON, Jong-Hwan
`#2-710, Jangmi Apt., 7, Sincheon-dong, Songpa-gu, Seoul
`WON, Hui-Seon
`11/1, 115, Seongbuk-dong 1-ga, Seongbuk-gu, Seoul
`DO, Myeong-Gyu
`366-12, Sindaebang-dong, Dongjak-gu, Seoul
`YANG, Gyeong-Ho
`#106-201, Woosung Apt., 365, Singil 3-dong, Yeongdeungpo-gu, Seoul
`Soo-Won Kang
`11-8, Sillim 9-dong, Gwanak-gu, Seoul
`
`(74) Agent(s)
`
`KIM Byung-Jin, BAIK Myung-Ja
`
`
`Examiner: KWON, Jang-Woo (Booklet Gazette No. 5341)
`(54) METHOD AND APPARATUS FOR ENCODING DCT BLOCKS USING BLOCK-ADAPTING SCAN
`Abstract
`
`The present invention relates to a method and apparatus for coding DCT blocks using block-adapting scan, and in particular, the encoder (100)
`is composed of a scan pattern storage unit (11), a DCT conversion unit (12), a feature extraction unit (13), a scan pattern determination unit
`(14), a processor unit (15), and an entropy encoding unit (16), and the decoder (200) is composed of a scan pattern storage unit (21), a DCT
`conversion unit (22), a feature extraction unit (23), a scan pattern determination unit (24), an entropy decoding unit (25), a processor unit (26),
`which can greatly improve the efficiency by selecting an arbitrary set of scan patterns, analyzing the features of each DCT block of an image,
`and applying a suitable scan pattern to the image, instead of selecting a scan pattern reflecting the features of a general image and adapting it
`uniformly as in the prior art.
`
`Representative figure
`
`FIG. 1
`
`Specifications
`
`[Title of the Invention]
`
`Text Information Display Apparatus
`
`[Brief description of the drawings]
`
`FIG. 1 is an original image taken by a camera.
`
`FIG. 2 is an image for which the current image has been motion compensated from the previous image.
`
`FIG. 3 is a difference image when the current image is predicted with a motion compensated image.
`
`FIG. 4 is an exemplary diagram of an embodiment of a set of scanning patterns adapted to the present invention.
`
`FIG. 5 is a diagram showing the CDT coefficient used for feature extraction.
`
`FIG. 6 is a flowchart of an encoder for explaining the method of the present invention.
`
`FIG. 7 is a flowchart of a decoder for explaining the method of the present invention.
`
`FIG. 8 is a system configuration diagram of an encoder in the apparatus of the present invention.
`
`FIG. 9 is a system configuration diagram of a decoder in the apparatus of the present invention.
`
`* Description of reference numerals for major parts of the drawings
`
`8-1
`
`Unified Patents, LLC v. Elects. & Telecomm. Res. Inst., et al.
`
`Ex. 1010, p. 2
`
`
`
`1019940025774
`
`11, 21: Scan pattern storage unit
`
`13, 23: Feature extraction unit
`
`15, 26: Processor unit
`
`
`
`25: Entropy decoding unit
`
`200: Decoder
`
`
`
`
`
`
`
`
`
`[Detailed description of the invention]
`
`12, 22: DCT conversion unit
`
`14, 24: Scan determination unit
`
`16: Entropy encoding unit
`
`100: Encoder
`
`The present invention relates to a coding method for compressing and restoring image data with a very large amount of information, such as
`high-definition TV (HDTV), that is, a quantization scanning technique for a DCT (Discrete Cosine Transform) coefficient in an image
`encoding and decoding process, and in particular, to a method and apparatus for encoding DCT blocks using a block-adaptive scan to increase
`the efficiency of variable-length coding by adaptively transforming a scan pattern.
`
`In general, the image data compression method draws a great deal of interest in various application fields, such as broadcasting and storage of
`communication and video signals. Since the processing of the digital signal has many advantages in data compression applications over the
`processing of analog signals, the main focus of the image transmission field is on encoding digital images.
`
`Furthermore, such trend is gaining more attention with the introduction of digital HDTV systems and image transmission through a network,
`and the standardization of image encoding methods in various application fields is being studied. Examples include CCITT Recommendation
`H.261 and MPEG.
`
`Most of these systems encode a video signal using motion compensation and DCT (MC-DCT encoder), and when motion compensation and
`DCT are performed, temporal and spatial redundancy of video data gets removed.
`
`In a general MC-DCT encoder, a motion compensation prediction difference image is conceptually encoded in the following three steps.
`
`1. 2D DCT of a motion compensation prediction difference image
`
`2. Scalar quantization of DCT coefficients
`
`3. Entropy encoding of quantized coefficients
`
`Since most of the DCT coefficients are quantized to zero and non-zero coefficients are sparsely distributed, an effective coding technique
`called “run-length encoding” is often used with this characteristic. Run-length encoding, that is, entropy encoding is performed on the
`consecutive zeros together with following non-zero values.
`
`For effective run-length encoding, a scan pattern should be created to make the entropy and average coding length small, and from this point
`of view, a zigzag scan is generally an effective scan pattern.
`
`Since the MC-DCT encoder allocates approximately 90% of the total number of bits to encode the quantized DCT coefficients, it is very
`important to display them to be effective in encoding them.
`
`For this purpose, several methods for analyzing the features of DCT coefficients have been proposed, but most of the image encoding scan
`methods developed so far adopt a method for selecting a scan pattern that reflects the general features of an image and applying it uniformly
`to each image, which lowers efficiency.
`
`Unlike the conventional method, the present invention provides a method and apparatus of encoding DCT blocks using block-adapting scan
`capable of improving efficiency by determining an arbitrary set of scan patterns, analyzing the features of each DCT block of an image, and
`applying the suitable scan pattern.
`
`In other words, the present invention is a method for encoding quantized DCT blocks to reduce the number of bits by adaptively changing the
`scan pattern of each block, and the scan pattern is selected to make the run-length as short as possible so that the scan pattern of the motion
`compensation prediction difference is determined generally by using a block of a corresponding motion compensated image having a similar
`contour feature to that of the scan pattern.
`
`The most important characteristic of the present invention is that it uses a motion compensated image, which is present in both the encoder
`and the decoder, to analyze the block features of an image and does not require other additional information, and that it is compatible with the
`international standard MPEG-2.
`
`When the method according to the present invention is applied, a reduction effect of approximately 5% to 10% can be expected based on the
`bit rate depending on the image.
`
`Hereinafter, the present invention will be described in detail with reference to the accompanying figures.
`
`FIGS. 6 and 7 are flowcharts for explaining the method of the present invention, and the process for encoding an image is composed of a step
`for initializing a block address with the first block of an image to be encoded (S1); a step for DCT-converting the current block indicated by
`the block address in the motion compensated image (S2); a step for calculating a value (t) of the extraction function (F) to identify the contour
`features of the DCT block generated in step (S2) (S3); a step for determining (I) a scan pattern suitable for the current DCT block from the set
`of scan patterns defined using the t value calculated in step (S3) and the selection function (X) (S4); a step for DCT-converting and quantizing
`the current block of the motion compensation prediction difference image (S5); a step for coding the DCT coefficients of the block generated
`in step (S5) in variable-length in the order of the scan patterns determined in step (S4) and outputting them to a bit string (S6); a step for
`increasing the address of the block (S7); and a step for detecting whether it is the end of the valid block, and ending the encoding process if it
`
`8-2
`
`Unified Patents, LLC v. Elects. & Telecomm. Res. Inst., et al.
`
`Ex. 1010, p. 3
`
`
`
`1019940025774
`
`is the end and repeating the process after step (S2) if it is not the end of the valid block (S8), and also, the process for decoding an image is
`composed of a step for initializing a block address with the first block of an image to be decoded (S11); a step for DCT-converting the current
`block indicated by the block address in the motion compensated image (S12); a step for calculating the value (t) of the extraction function (F)
`to identify the contour features of the DCT block generated in step (S12) (S13); a step for determining (I) a suitable scan pattern from a set of
`scan patterns defined using the t value calculated in step (S13) and the selection function (X) (S14); a step for coding the compressed bit string
`coming from the encoder in variable-length and outputting DCT coefficients (S15); a step for reconstructing the block with the DCT
`coefficients decoded in step (S15) by referring to the scan pattern determined in step (S14) (S16); a step for increasing the address of the block
`(S17); and a step for detecting whether it is the end of the valid block, and ending the decoding process if it is the end and repeating the process
`after step (S12) if it is not the end of the valid block (S18).
`
`FIGS. 8 and 9 show an embodiment of and encoder and a decoder in the apparatus of the present invention, and the encoder (100) is composed
`of a scan pattern storage unit (11) for storing data of an arbitrary set of scan patterns; a DCT conversion unit (12) for receiving the input of the
`current block of a motion compensated image and converting it to DCT; a feature extraction unit (13) for receiving a DCT block from the
`DCT conversion unit (12) and calculating a value of an extraction function (F) to identify the contour features of the block; a scan pattern
`determination unit (14) for determining (I) a scan pattern based on the value of the extraction function (F) output from the feature extraction
`unit (13); a processor unit (15) for referring to the scan pattern determined in the scan pattern storage unit (14) in the scan pattern storage unit
`(11) and applying it to the quantized DCT block of the motion compensation prediction difference image; and an entropy encoding unit (16)
`for variable-length encoding DCT coefficients coming in from the processor unit (15), and the decoder (200) is composed of a scan pattern
`storage unit (21) for storing data of an arbitrary set of scan patterns; a DCT conversion unit (22) for receiving the input of the current block of
`a motion compensated image and converting it to DCT; a feature extraction unit (23) for calculating the value of an extraction function (F) to
`identify the contour of the block; a scan pattern determination unit (24) for determining (I) the scan pattern based on the value of the extraction
`function (F) output from the feature extraction unit (23); an entropy decoding unit (25) for variable-length decoding the compressed bit string;
`and a processor unit (26) for referring to the scan pattern determined in the scan pattern storage unit (24) in the scan pattern storage unit (21)
`and reconstructing the DCT coefficients coming in from the entropy decoding unit (25).
`
`The effects of the present invention are as follows.
`
`First, the basic idea of the present invention is that the contour features of the original image captured by the camera, the motion compensated
`image, and the motion compensation prediction difference image are similar to each other. FIGS. 1, 2, and 3 are examples showing the original
`image, the motion compensated image, and the motion compensation prediction difference image presented above.
`
`In addition, in order to apply the block-adapting scan method, an arbitrary set of scan patterns reflecting the characteristics of DCT coefficients
`must be defined according to the direction of the contour.
`
`FIG. 4 illustrates a set of five scan patterns, and other various scan pattern sets can be configured.
`
`The sets of scan patterns are stored in the scan pattern storage units (11)(21) installed in the encoder (100) and the decoder (200), respectively,
`and when the scan pattern is determined (I) in the scan pattern determination units (14)(24), the processor (15) refers to it.
`
`On the other hand, the present invention uses a motion compensation image that is commonly effective in the encoder (100) and the decoder
`(200) to determine a suitable scan pattern from a set of scan patterns for each DCT-converted block of the motion compensation prediction
`difference image, and converts the block of the motion compensated image into DCT, and analyzes and uses the coefficient for efficient and
`simple extraction of contour features.
`
`In other words, the DCT conversion units (12)(22) receive the current block indicated by the block address in the motion compensated image
`and perform DCT conversion.
`
`In addition, the feature extraction units (13)(23) that have received the output signals of the DCT conversion units (12)(22) calculate the value
`(t) of the extraction function (F) to identify the contour features of the block.
`
`Also, the scan pattern determination units (14)(24) determine (I) the scan pattern based on the extraction function (F) value (t) and the selection
`function (X) output from the feature extraction units (13)(23).
`
`FIG. 5 shows the locations of DCT coefficients used for feature extraction in each of the feature extraction units (13)(23), and here, Hi refers
`to the position of the coefficients indicating the horizontal contour feature, and Vi refers to that indicating the vertical contour feature. When
`the primary value of the scanning pattern is indicated as I for P, which is a block of a DCT-converted motion compensated image, I is defined
`as follows.
`
`I(P)=X(F(P))
`
`Here, F is a function for extracting required features from P, and X is a selection function to be matched to the scan pattern according to the
`value calculated in F(P).
`
`The function F for extracting the features of the contour is defined as the ratio of the state of horizontal and vertical energy as follows.
`
`
`
`In this case, it is preferable that the scan pattern selection function X is determined to select No. 1 or No. 2 for the vertical contour, and No. 4
`or No. 5 contour for the horizontal contour when viewed from the scan pattern set in FIG. 4. The selection function X for the set of scan
`patterns in FIG. 4 is defined as follows.
`
`8-3
`
`Unified Patents, LLC v. Elects. & Telecomm. Res. Inst., et al.
`
`Ex. 1010, p. 4
`
`
`
`1019940025774
`
`
`
`ε1 and ε2 in the above equation are threshold values determined in consideration of the motion compensation prediction difference image and
`the extraction function (F).
`
`In addition, the processor unit (16) in the encoder (100) receives the scan pattern determined by the scan pattern determination unit (14) from
`the scan pattern storage unit (11), and processes it after applying it to the current quantized DCT block indicated by the block address in the
`motion compensation prediction difference image.
`
`When the quantized DCT block is output in the order of the scan patterns from the processor unit (15) in this manner, the entropy encoding
`unit (16) receiving this input codes the DCT coefficients in variable-length, outputs them as a bit string, and then increases the address of the
`block, and this process is repeated until the valid block is finished.
`
`On the other hand, the entropy decoding unit (25) in the decoder (200) decodes the bit string coming from the encoder in variable-length, and
`outputs DCT coefficients.
`
`The processor unit (26) receiving the DCT coefficients from the entropy decoding unit (25) reconstruct the DCT block by referring to the scan
`pattern selected in the scan pattern storage unit (21) and then increases the block address, and this process is repeated until the valid block is
`finished.
`
`As described above, according to the present invention, the efficiency can be greatly improved by selecting an arbitrary set of scan patterns,
`analyzing the features of each DCT block of an image, and applying a suitable scan pattern to the image, instead of uniformly applying a scan
`pattern reflecting the general features of an image as in the prior art.
`
`(57) Scope of claims
`
`Claim 1
`
`A method for encoding DCT blocks using block-adapting scan in a system to which a method for encoding a predetermined image and
`decoding the encoded image back to the original image is applied, wherein the image encoding process is composed of a step for initializing
`a block address with the first block of an image to be encoded (S1); a step for DCT-converting the current block indicated by the block address
`in the motion compensated image (S2); a step for calculating a value (t) of the extraction function (F) to identify the contour features of the
`DCT block generated in step (S2) (S3); a step for determining (I) a scan pattern suitable for the current DCT block from the set of scan patterns
`defined using the t value calculated in step (S3) and the selection function (X) (S4); a step for DCT-converting and quantizing the current
`block of the motion compensation prediction difference image (S5); a step for coding the DCT coefficients of the block generated in step (S5)
`in variable-length according to the scan patterns determined in step (S4) (S6); a step for increasing the address of the block (S7); and a step
`for detecting whether it is the end of the valid block, and ending the encoding process if it is the end and repeating the process after step (S2)
`if it is not the end of the valid block (S8), wherein the process for decoding an image is composed of a step for initializing a block address
`with the first block of an image to be decoded (S11); a step for DCT-converting the current block indicated by the block address in the motion
`compensated image (S12); a step for calculating the value (t) of the extraction function (F) to identify the contour features of the DCT block
`generated in step (S12) (S13); a step for determining (I) a suitable scan pattern from the set of scan patterns defined using the t value calculated
`in step (S13) and the selection function (X) (S14); a step for reconstructing the block into DCT coefficients by decoding the compressed bit
`string in variable-length (S16); a step for increasing the address of the block (S17); a step for detecting whether it is the end of the valid block,
`and ending the decoding process if it is the end if it is not the end, with step (S17); and a step for detecting whether it is the end of the valid
`block, and ending the decoding process if it is the end and repeating the process after step (S12) if it is not the end of the valid block (S18).
`
`Claim 2
`
`The method for encoding DCT blocks using the block-adapting scan of claim 1, wherein the determination of the scan pattern (I(P)) for the
`block (P) of the DCT-converted motion compensated image calculates the extraction function (F) for the block (P) and is defined by the scan
`pattern selection function (X) value (I(P)=X((F(P)) for that value.
`
`Claim 3
`
`The method for encoding DCT blocks using the block-adapting scan of claim 1, wherein the extraction function (F) is defined by
`
`
`
`(provided, Hi is the horizontal contour feature and Vi is the vertical contour feature)
`
`
`
`Claim 4
`
`The method for encoding DCT blocks using the block-adapting scan of claim 1, wherein the selection function (X) is determined by threshold
`values (ε1, ε2) determined in consideration of the motion compensation prediction difference image and extraction function (F).
`
`
`
`8-4
`
`Unified Patents, LLC v. Elects. & Telecomm. Res. Inst., et al.
`
`Ex. 1010, p. 5
`
`
`
`1019940025774
`
`Claim 5
`
`An apparatus for encoding DCT blocks using a block-adapting scan composed of an encoder (100) composed of a scan pattern storage unit
`(11) for storing data of an arbitrary set of scan patterns; a DCT conversion unit (12) for receiving the block of a motion compensated image
`and converting it to DCT; a feature extraction unit (13) for receiving an input of the output signal of the DCT conversion unit (12) and
`calculating the value of an extraction function (F) to identify the contour features of the block; a scan pattern determination unit (14) for
`determining (I) the scan pattern based on the value of the extraction function (F) output from the feature extraction unit (13); a processor unit
`(15) for bringing the scan pattern determined in the scan pattern storage unit (14) and the scan pattern determined in the scan pattern storage
`unit (11) from the scan pattern storage unit (11) and applying it to the quantized DCT block of the motion compensation prediction difference
`image; and an entropy encoding unit (16) for variable-length encoding DCT coefficients inputted from the processor unit (15), a scan pattern
`storage unit (21) for storing data of an arbitrary set of scan patterns; a DCT conversion unit (22) for receiving the input of the current block of
`a motion compensated image and converting it to DCT; a feature extraction unit (23) for calculating the value of an extraction function (F) to
`identify the contour of the block; a scan pattern determination unit (24) for determining (I) the scan pattern based on the value of the extraction
`function (F) output from the feature extraction unit (23); an entropy decoding unit (25) for variable-length decoding the compressed bit string;
`and a processor unit (26) for referring to the scan pattern determined in the scan pattern storage unit (24) in the scan pattern storage unit (21)
`and reconstructing the DCT coefficients coming in after being decoded through the entropy decoding unit (25) into blocks according to the
`selected scan pattern.
`
`Figure 1
`
`Figure 2
`
`Figure 3
`
`
`
`Figures
`
`
`
`
`
`
`
`
`
`
`
`
`
`8-5
`
`
`
`
`
`
`
`Unified Patents, LLC v. Elects. & Telecomm. Res. Inst., et al.
`
`Ex. 1010, p. 6
`
`
`
`1019940025774
`
`
`
`Figure 4
`
`No. 1
`
`No. 2
`
`No. 3
`
`No. 4
`
`No. 5
`
`
`
`Figure 5
`
`
`
`
`
`
`
`
`
`
`
`
`
`8-6
`
`Unified Patents, LLC v. Elects. & Telecomm. Res. Inst., et al.
`
`Ex. 1010, p. 7
`
`
`
`1019940025774
`
`
`
`Figure 6
`
`Start
`
`Initialize the block address of the image
`
`DCT-convert the current block of the motion compensated image
`
`Calculate value t of extraction function F for the
`block
`
`Determine the suitable scan pattern from a set of scan
`patterns by the value of t and selection function X
`
`DCT-convert and quantize the current block of the
`motion compensation prediction difference image
`
`DCT-convert and quantize the current block of the
`motion compensation prediction difference image
`
`Increase the block address
`
`Is it the end of valid block?
`
`No
`
`Yes
`
`End
`
`
`
`
`
`
`
`
`
`Figure 7
`
`Start
`
`Initialize the block address of the image
`
`DCT-convert the current block of the motion compensated image
`
`Calculate value t of extraction function F for the
`block
`
`Determine the suitable scan pattern from a set of scan patterns
`by the value of t and selection function X
`
`Variable-length decode the encoded bit sting
`
`Reconfigure the block by the scan pattern with
`decoded DCT coefficients
`
`Increase the block address
`
`Is it the end of valid block?
`
`No
`
`Yes
`
`End
`
`
`
`8-7
`
`Unified Patents, LLC v. Elects. & Telecomm. Res. Inst., et al.
`
`Ex. 1010, p. 8
`
`
`
`
`
`
`
`
`
`Figure 8
`
`Figure 9
`
`1019940025774
`
`
`
`
`
`8-8
`
`Unified Patents, LLC v. Elects. & Telecomm. Res. Inst., et al.
`
`Ex. 1010, p. 9
`
`