Apparatus for prediction coding or decoding image signal and method therefor

• US 7,010,168 B2
• Filed: 03/15/2002
• Issued: 03/07/2006
• Est. Priority Date: 12/26/2001
• Status: Active Grant

• Current Assignee: Mtekvision Co.Ltd.
• Original Assignee: Electronics and Telecommunications Research Institute, VK Corporation
• Inventors: Kim, Kyung-soo, Park, Seong-mo, Suh, Ki-bum, Kim, Seong-min, Kim, Ig-kyun, Koo, Bon-tae
• Primary Examiner(s): Couso, Jose L.
• Application Number:
  US10/098,213
• Publication Number:
  US 20030118239A1
• Time in Patent Office: 1,453 Days
• Field of Search: 382232-233, 382235-236, 382/238, 382/248, 382250-251, 348/384.1, 348/394.1, 348/395.1, 3484001-4031, 3484071-4161, 3484201-4211, 348/425.3, 348/426.1, 3484301-4311, 348/699, 348/715, 348/717, 37524012-24013, 37524014-24015, 375/240.16, 375/240.18, 375/240.2, 375/240.22, 375/240.24, 375/240.25, 375/240.26, 386/109, 386/111, 386/112
• US Class Current: 382/238
• CPC Class Codes: H04N 19/105 Selection of the reference unit for prediction within a chosen coding or prediction mode, e.g. adaptive choice of position and number of pixels used for prediction
• CPC Class Codes: H04N 19/14 Coding unit complexity, e.g. amount of activity or edge presence estimation H04N19/146 takes precedence
• CPC Class Codes: H04N 19/42 characterised by implementation details or hardware specially adapted for video compression or decompression, e.g. dedicated software implementation H04N19/635 takes precedence
• CPC Class Codes: H04N 19/423 characterised by memory arrangements H04N19/433 takes precedence
• CPC Class Codes: H04N 19/593 involving spatial prediction techniques
• CPC Class Codes: H04N 19/61 in combination with predictive coding
• CPC Class Codes: H04N 19/63 using sub-band based transform, e.g. wavelets

Abstract

A hardware structure for adaptively prediction coding an image signal using spatial correlation and a method therefor are provided. The prediction coding apparatus has a packet change detection unit which determines whether or not blocks used to determine a prediction direction for a current block for which prediction is currently performed is included in a same packet to which the current block belongs; a prediction reference value providing unit which provides a prediction reference value used in prediction according to the result of determination by the packet change detection unit; and a prediction calculation unit which outputs a predictive coded value or a predictive decoded value using the prediction reference value provided by the prediction reference value providing unit and the DCT component of the current block according to an operation mode. According to the apparatus and method, the size of a prediction apparatus is reduced, and one apparatus is used both in coding and decoding.

Claims

1. An apparatus for predicting a Discrete Cosine Transform (DCT) component of an image signal in an apparatus for prediction coding or decoding an image signal, the apparatus for predicting a DCT component comprising:

a packet change detection unit which determines whether or not blocks used to determine a prediction direction for a current block for which prediction is currently performed is included in a same packet to which the current block belongs;

a prediction reference value providing unit which provides a prediction reference value used in prediction according to the result of determination by the packet change detection unit; and

a prediction calculation unit which outputs a predictive coded value or a predictive decoded value using the prediction reference value provided by the prediction reference value providing unit and the DCT component of the current block according to an operation mode.

2. The apparatus of claim 1, wherein the prediction calculation unit performs subtraction of the DCT component of the current block and the prediction reference value if the operation mode is a coding mode, and performs addition of the DCT component of the current block and the prediction reference value if the operation mode is a decoding mode.

3. The apparatus of any one of claims 1, wherein the apparatus for predicting a DCT component further comprises an inverse quantizing unit which inverse quantizes the DCT component of the current block if the operation mode is a coding mode, and inverse quantizes the prediction decoded value which is output from the prediction calculation unit if the operation mode is a decoding mode, and provides the inverse quantized value to the prediction reference value providing unit.

4. The apparatus of any one of claim 2, wherein the apparatus for predicting a DCT component further comprises an (inverse) quantizing unit which inverse quantizes the DCT component of the current block if the operation mode is a coding mode, and inverse quantizes the prediction decoded value which is output from the prediction calculation unit if the operation mode is a decoding mode, and provides the inverse quantized value to the prediction reference value providing unit.

5. The apparatus of claim 1, wherein the packet change detection unit comprises:

a first packet change signal output unit which outputs a first packet change signal indicating whether or not the packet of a macro block to the left of the current macro block is the same as the packet of the current macro block, using a current packet change signal indicating whether or not the packet to which the current macro block belongs changed;

a storage unit which is formed with a plurality of storage areas of which number corresponds to the number of macro blocks forming X axis of a scene, and stores the current packet change signal in a storage area corresponding to a location of the current block on the X axis of the scene;

a selection signal generating unit which generates a selection signal indicating the location value on the X axis of the scene of the current macro block;

a first calculation unit which calculates a first output value using the selection signal and a value stored in the storage unit;

a second calculation unit which outputs a second packet change signal indicating whether or not the packet of a macro block immediately above the current macro block is the same as the packet to which the current macro block belongs, using the first output value and the current packet change signal; and

a third calculation unit which outputs a third packet change signal indicating whether or not the packet of a macro block immediately above and to the left of the current macro block is the same as the packet to which the current macro block belongs, using the second packet change signal and a value stored in a storage area corresponding to the location on the X axis of the scene of the current macro block among storage areas forming the storage unit.

6. The apparatus of claim 5, wherein the current packet change signal is input to the packet change detection unit at a first clock, the first through third packet change signals are output at a second clock, the current packet change signal is stored in a storage area corresponding to the location on the X axis of the scene of the current macro block among storage areas forming the storage unit at a third clock, and the first through the third clocks are sequentially generated.

7. The apparatus of claim 5, wherein a storage area forming the storage unit is a 1-bit register.

8. The apparatus of claim 1, wherein the prediction reference value providing unit comprises:

a horizontal direction memory unit which has storage areas capable of storing a horizontal DCT component of a color difference signal block of a slice forming a scene and a horizontal DCT component of any one brightness signal block of an upper brightness signal block and a lower brightness signal block of a slice;

a vertical direction memory unit which has storage areas capable of storing a vertical DCT component of any one column of a vertical DCT component forming a slice;

an upper-left direction memory unit which has storage areas capable of storing the Direct Current (DC) component of a block immediately above and to the left of the current block for which prediction is currently performed; and

a memory control unit which stores the DC component of a block immediately above the current block in the upper-left memory unit, stores the horizontal DCT component of the current block in the horizontal direction memory unit, and stores the vertical DCT component of the current block in the vertical direction memory unit.

9. The apparatus of claim 8, wherein the prediction reference value providing apparatus further comprises a prediction direction determining unit which determines a prediction direction using the DC component of the block to the left of the current block stored in the vertical direction memory unit, the DC component of the block immediately above the current block stored in the horizontal direction memory unit, and the DC component of the block immediately above and to the left of the current block.

10. The apparatus of claim 9, wherein the memory control unit provides the horizontal DCT component of the upper block of the current block from the horizontal direction memory unit to the calculation unit if the prediction direction output from the prediction direction determining unit is a first direction, and provides the vertical DCT component of the left-hand block of the current block from the vertical direction memory unit to the calculation unit if the prediction direction output from the prediction direction determining unit is a second direction.

11. The apparatus of claim 8, wherein the memory control unit stores the horizontal DCT component of the current block in the horizontal direction memory unit, replacing the horizontal DCT component of the upper block of the current block, and stores the vertical DCT component of the current block in the vertical direction memory unit, replacing the vertical DCT component of the left-hand block of the current block.

12. A method for predicting an image signal in a method for predictive coding or predictive decoding an image signal using a prediction memory which comprises a horizontal direction memory unit which has storage areas capable of storing horizontal DCT components of a slice forming a scene, a vertical direction memory unit which has storage areas capable of storing a vertical DCT component, and an upper-left memory unit storage areas capable of storing a DC component of a block immediately above and to the left of a current block for which prediction is performed, the method for prediction an image signal comprises:

(a) determining whether or not the packet to which the current macro block belongs is different from the packet of a macro block to the left of the current macro block, the macro block immediately above and to the left of the current block, or a macro block immediately above the current macro block, using a current packet change signal indicating whether or not the packet to which the current macro block belongs changed;

(b) if the packet to which the current macro block belongs is the same as the packet of the macro block to the left of the current macro block, the macro block immediately above and to the left of the current block, or the macro block immediately above the current macro block, determining the direction of prediction, using the DC component of the block to the left of the current block stored in the vertical direction memory unit, the DC component of the block immediately above the current block stored in the horizontal direction memory unit, and the DC component of the block immediately above and to the left of the current block;

(c) if the prediction direction is a first direction, outputting a reference prediction value by using a value stored in the vertical direction memory unit and if the prediction direction is a second direction, outputting a reference prediction value by using a value stored in the horizontal direction memory unit;

(d) if the packet to which the current macro block belongs is different from the packet of the macro block to the left of the current macro block, the macro block immediately above and to the left of the current block, or the macro block immediately above the current macro block, outputting a predetermined reference value;

(e) performing prediction calculation by using the reference prediction value of step (c) or the predetermined reference value of step (c) with the DCT component of the current block; and

(f) storing the DCT component of the current block in the prediction memory.

13. The method for predicting an image signal of claim 12, wherein step (a) comprises:

(aa) generating a selection signal indicating the location value on the X axis of the scene of the current macro block;

(ab) outputting a first output value by using a value stored in a storage unit which is formed with a plurality of storage areas of which number corresponds to the number of macro blocks forming X axis of a scene, and stores the current packet change signal in a storage area corresponding to a location of the current block on the X axis of the scene, and the selection signal;

(ac) outputting a first packet change signal indicating whether or not the packet of the macro block to the left of the current macro block is the same as the packet of the current macro block, using a current packet change signal indicating whether or not the packet to which the current macro block belongs changed;

(ad) outputting a second packet change signal indicating whether or not the packet of the macro block immediately above the current macro block is the same as the packet to which the current macro block belongs, using the first output value and the current packet change signal; and

(ae) outputting a third packet change signal indicating whether or not the packet of the macro block immediately above and to the left of the current macro block is the same as the packet to which the current macro block belongs, using the second packet change signal and a value stored in a storage area corresponding to the location on the X axis of the scene of the current macro block among storage areas forming the storage unit.

14. The method for predicting an image signal of claim 12, wherein in step (f), the DC component of the block immediately above the current block is stored in the upper-left direction memory unit, the horizontal DCT component of the current block is stored in the horizontal direction memory unit, and the vertical DCT component of the current block is stored in the vertical direction memory unit.

15. A computer readable medium having embodied thereon a computer program for implementing a method for predicting an image signal in a method for predictive coding or predictive decoding an image signal using a prediction memory which comprises a horizontal direction memory unit which has storage areas capable of storing horizontal DCT components of a slice forming a scene, a vertical direction memory unit which has storage areas capable of storing vertical DCT components, and an upper-left memory unit storage areas capable of storing a DC component of a upper left-hand block of a current block for which prediction is performed, the method for prediction an image signal comprises:

(a) determining whether or not the packet to which the current macro block belongs is different from the packet of a left-hand macro block of the current macro block, the upper left-hand macro block of the current macro block, or an upper macro block of the current macro block, using a current packet change signal indicating whether or not the packet to which the current macro block belongs changed;

(b) if the packet to which the current macro block belongs is the same as the packet of the left-hand macro block of the current macro block, the upper left-hand macro block of the current macro block, or the upper macro block of the current macro block, determining the direction of prediction, using the DC component of the left-hand block of the current block stored in the vertical direction memory unit, the DC component of the upper block of the current block stored in the horizontal direction memory unit, and the DC component of the upper left-hand block of the current block;

(c) if the prediction direction is a first direction, outputting a reference prediction value by using a value stored in the vertical direction memory unit and if the prediction direction is a second direction, outputting a reference prediction value by using a value stored in the horizontal direction memory unit;

(d) if the packet to which the current macro block belongs is different from the packet of the left-hand macro block of the current macro block, the upper left-hand macro block of the current macro block, or the upper macro block of the current macro block, outputting a predetermined reference value;

(e) performing prediction calculation by using the reference prediction value of step (c) or the predetermined reference value of step (c) with the DCT component of the current block; and

(f) storing the DCT component of the current block in the prediction memory.