Apparatus and method for adaptively encoding pictures in accordance with information quantity of respective pictures and inter-picture correlation

• US 5,883,672 A
• Filed: 07/18/1997
• Issued: 03/16/1999
• Est. Priority Date: 09/29/1994
• Status: Expired due to Fees

• Current Assignee: Sony Corporation (Sony Group Corp.)
• Original Assignee: Sony Corporation (Sony Group Corp.)
• Inventors: Suzuki, Kazuhiro, Ando, Yuji, Mitsuhashi, Satoshi
• Primary Examiner(s): LIN, BRYAN T
• Application Number:
  US08/896,585
• Time in Patent Office: 606 Days
• Field of Search: 348/419, 348/416, 348/415, 348/409, 348/402, 348/401, 348/400, 348/405, 348/390, 348/384
• US Class Current: 375/240.04
• CPC Class Codes: H04N 19/107 between spatial and temporal predictive coding, e.g. picture refresh
• CPC Class Codes: H04N 19/114 Adapting the group of pictures [GOP] structure, e.g. number of B-frames between two anchor frames H04N19/107 takes precedence
• CPC Class Codes: H04N 19/124 Quantisation
• 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 19/142 Detection of scene cut or scene change
• CPC Class Codes: H04N 19/146 Data rate or code amount at the encoder output
• CPC Class Codes: H04N 19/149 by estimating the code amount by means of a model, e.g. mathematical model or statistical model
• CPC Class Codes: H04N 19/15 by monitoring actual compressed data size at the memory before deciding storage at the transmission buffer
• CPC Class Codes: H04N 19/172 the region being a picture, frame or field
• CPC Class Codes: H04N 19/176 the region being a block, e.g. a macroblock
• CPC Class Codes: H04N 19/177 the unit being a group of pictures [GOP]
• CPC Class Codes: H04N 19/179 the unit being a scene or a shot
• CPC Class Codes: H04N 19/60 using transform coding
• CPC Class Codes: H04N 19/61 in combination with predictive coding
• CPC Class Codes: H04N 19/87 involving scene cut or scene change detection in combination with video compression

Abstract

A method and apparatus for encoding a picture with improved prediction accuracy in predicting the basic quantization step. The apparatus for encoding a picture includes a picture data storage unit, a picture information evaluating unit, an inter-picture correlation detection unit, an orthogonal transform unit for orthogonally transforming the picture data for generating orthogonal transform coefficients, a quantization unit for quantizing the orthogonal transform coefficients produced by the orthogonal transform unit at a pre-set quantization step, an adaptive compression method selection unit and a quantization step control unit. The information quantity is evaluated from stored picture data of plural pictures, and the inter-picture correlation is detected. A picture data compression method is adaptively selected based upon the evaluated value of the information quantity and the inter-picture correlation information. The basic quantization step is predicted from the evaluated value and a pre-set compressed data quantity obtained on compressing one-picture picture data by the selected compression method and the evaluated value.

Claims

1. Apparatus for adaptively encoding pictures in accordance with information quantity of respective pictures and inter-picture correlation, comprising:

a storage device for storing picture data of a plurality of input pictures to provide stored picture data;

a picture information evaluator for evaluating the information quantity of said plurality of input pictures of said stored picture data to generate evaluated values of information quantity;

an inter-picture correlation detector for detecting the correlation between pictures of said stored picture data to generate inter-picture correlation information;

an orthogonal transformer for orthogonally transforming said stored picture data to generate orthogonal transform coefficients;

a quantizer for quantizing the orthogonal transform coefficients using a pre-set quantization step size;

a selector for adaptively selecting a compression method for each of said plurality of input pictures of said stored picture data as a function of said evaluated values of the information quantity and the inter-picture correlation information to provide a selected compression method; and

a controller for predicting a basic quantization step size for said quantizer as a function of said evaluated values of the information quantity and an estimated quantity of compressed data from compressing one-picture picture data in accordance with said selected compression method;

wherein said controller learns the relation between a quantization step size employed for compression, a post-compression data quantity and the evaluated values of information quantity to generate learning parameters and predicting the basic quantization step size for said quantizer as a function of said learning parameters.

2. The apparatus as claimed in claim 1, wherein said controller sums the evaluated value for each macro-block obtained from dividing said picture and the evaluated value for a full picture to generate a summed evaluated value and predicts said basic quantization step size as function of said summed evaluated value.

3. The apparatus as claimed in claim 1, wherein said picture information evaluator generates said evaluated values from a sum of macro-block-based absolute values of a difference between pixel data of a macro-block of a reference picture corresponding to a motion vector obtained by motion detection and pixel data of a macro-block of an input picture.

4. The apparatus as claimed in claim 1, wherein said controller sums said evaluated values during a time interval corresponding to a period of intra-frame coding to allocate bits for quantization.

5. The apparatus as claimed in claim 4, wherein the quantity of picture-based bit allocation for quantization and the sum of the evaluated values are related in a pre-set manner for each of the selected compression methods;

and wherein said controller calculates said basic quantization step size from previously determined learning parameters and said pre-set relation.

6. Method for adaptively encoding pictures in accordance with information quantity of respective pictures and inter-picture correlation, said method comprising the steps of:

storing picture data of a plurality of input pictures to provide stored picture data;

evaluating the information quantity of said plurality of input pictures of said stored picture data to generate evaluated values of information quantity;

detecting the correlation between pictures of said stored picture data to generate inter-picture correlation information;

orthogonally transforming said stored picture data to generate orthogonal transform coefficients;

quantizing the orthogonal transform coefficients by use of a quantizer using a pre-set quantization step size;

adaptively selecting a compression method for each of said plurality of input pictures of said stored picture data as a function of said evaluated values of the information quantity and the inter-picture correlation information to provide a selected compression method; and

predicting a basic quantization step size by use of a controller for said quantizer as a function of said evaluated values of the information quantity and an estimated quantity of compressed data from compressing one-picture picture data in accordance with said selected compression method;

wherein said controller learns the relation between a quantization step size employed for compression, a post-compression data quantity and the evaluated values of information quantity to generate learning parameters and predicting the basic quantization step size for said quantizer as a function of said learning parameters.

7. The method as claimed in claim 6, wherein said controller sums the evaluated value for each macro-block obtained from dividing said picture and the evaluated value for a full picture to generate a summed evaluated value and predicts said basic quantization step size as function of said summed evaluated value.

8. The method as claimed in claim 6, wherein the step of evaluating includes generating said evaluated values from a sum of macro-block-based absolute values of a difference between pixel data of a macro-block of a reference picture corresponding to a motion vector obtained by motion detection and pixel data of a macro-block of an input picture.

9. The method as claimed in claim 6, wherein said controller sums said evaluated values during a time interval corresponding to a period of intra-frame coding to allocate bits for quantization.

10. The method as claimed in claim 9, wherein the quantity of picture-based bit allocation for quantization and the sum of the evaluated values are related in a pre-set manner for each of the selected compression methods;

and wherein said controller calculates said basic quantization step size from previously determined learning parameters and said pre-set relation.