Data processing apparatus and data processing method

• US 7,031,513 B1
• Filed: 11/05/1999
• Issued: 04/18/2006
• Est. Priority Date: 11/09/1998
• Status: Expired due to Fees

• Current Assignee: Sony Corporation (Sony Group Corp.)
• Original Assignee: Sony Corporation (Sony Group Corp.)
• Inventors: Kondo, Tetsujiro
• Primary Examiner(s): Ahmed, Samir
• Assistant Examiner(s): Kim, Charles
• Application Number:
  US09/434,662
• Time in Patent Office: 2,356 Days
• Field of Search: 382155-159, 706/12, 706/14, 706/20, 706/21, 706/25, 700/47, 700/48
• US Class Current: 382/155
• CPC Class Codes: G06T 2207/20081 Training; Learning
• CPC Class Codes: G06T 5/20 using local operators
• CPC Class Codes: G06T 5/70 Denoising; Smoothing
• CPC Class Codes: H04N 5/21 Circuitry for suppressing or minimising disturbance, e.g. moiré or halo
• CPC Class Codes: H04N 5/213 Circuitry for suppressing or minimising impulsive noise for suppressing or minimising disturbance in image signal generation H04N23/81
• CPC Class Codes: H04N 5/46 for receiving on more than one standard at will deflecting circuits of multi-standard receivers H04N3/27
• CPC Class Codes: H04N 7/0135 involving interpolation processes interpolation-based image scaling G06T3/4007; interpolation for video coding H04N19/587, H04N19/59
• CPC Class Codes: H04N 7/0145 the interpolation being class adaptive, i.e. it uses the information of class which is determined for a pixel based upon certain characteristics of the neighbouring pixels

Abstract

A data processing apparatus processes input data and outputs the processed data. The data processing apparatus includes a data processing section and a real-time learning section. The data processing section processes the input data by a predetermined processing method and outputs the processed data. The real-time learning section controls such that the processing method is learned in real time and the data processing section processes the input data by the learned processing method, so that the output data is improved as time elapses.

Claims

1. A data processing apparatus for processing input data and outputting the processed data, comprising:

data processing means for processing the input data by a predetermined processing method and outputting the processed data as output data;

evaluating means for evaluating a reliability of the output data; and

real-time learning means for controlling such that the processing method is learned in real time and said data processing means processes the input data by the learned processing method, so that the output data is improved as time elapses,

wherein said real-time learning means learns the processing method according to the reliability of the output data;

whereby the processing method is learned without repeatedly determining the output data value for a given point in time.

2. A data processing apparatus according to claim 1, wherein said real-time learning means learns the processing method according to the input data and the output data.

3. A data processing apparatus according to claim 1, wherein said real-time learning means learns the processing method according to time-sequentially input learning data.

4. A data processing apparatus according to claim 3, wherein said real-time learning means learns the processing method according to data having a value within a predetermined range from the current value among the time-sequentially input learning data.

5. A data processing apparatus according to claim 3, wherein said real-time learning means learns the processing method while said real-time learning means ignores past learning data at a predetermined rate.

6. A data processing apparatus according to claim 1, further comprising:

classification means for classifying normal quality data according to a plurality of classes;

formulating means for formulating equations for each class; and

solving means for solving said equations and outputting prediction coefficients for each class.

7. A data processing method for processing input data and outputting the processed data, comprising:

a data processing step for processing the input data by a predetermined processing method and outputting the processed data as output data;

an evaluating step for evaluating a reliability of the output data; and

a real-time learning step for controlling such that the processing method is learned in real time and the input data is processed by the learned processing method in said data processing step, so that the output data is improved as time elapses,

wherein in said real-time learning step the processing method is learned according to the reliability of the output data;

whereby the processing method is learned without repeatedly determining the output data value for a given point in time.

8. A data processing apparatus for processing input data and outputting the processed data, comprising:

data processing means for processing the input data by a linear coupling of the input data and predetermined prediction coefficients and outputting the processed data; and

real-time learning means for learning the prediction coefficients in real time so that the output data has higher quality as time elapses and for transferring the prediction coefficients to said data processing means,

wherein the real-time learning means includes formulating means for formulating equations to obtain the prediction coefficients based on at least higher quality data than the input data, and solving means for solving said equations and outputting the prediction coefficients;

whereby the prediction coefficients are learned without repeatedly determining the output data value for a given point in time.

9. A data processing apparatus according to claim 8, wherein said real-time learning means learns the prediction coefficient according to time-sequentially input learning data.

10. A data processing apparatus according to claim 9, wherein said real-time learning means learns the prediction coefficient while said real-time learning means ignores past learning data at a predetermined rate.

11. A data processing apparatus according to claim 8, wherein said formulating means formulates the equations based on higher quality data than the input data and normal quality data generated by the higher quality data.

12. A data processing method for processing input data and outputting the processed data, comprising:

a data processing step for processing the input data by a linear coupling of the input data and predetermined prediction coefficients and outputting the processed data; and

a real-time learning step for learning the prediction coefficients in real time so that the output data has higher quality as time elapses and for transferring the prediction coefficients to said data processing step,

wherein the real-time learning step includes formulating equations to obtain the prediction coefficients based on at least higher quality data than the input data, solving said equations, and outputting the prediction coefficients;

whereby the prediction coefficients are learned without repeatedly determining the output data value for a given point in time.

13. A storage medium for storing a computer controllable program for processing input data and outputting the processed data, the program comprising:

a data processing step for processing the input data by a predetermined processing method and outputting the processed data as output data;

an evaluating step for evaluating a reliability of the output data; and

a real-time learning step for controlling such that the processing method is learned in real time and the input data is processed by the learned processing method in said data processing step, so that the output data is improved as time elapses,

wherein in said real-time learning step the processing method is learned according to the reliability of the output data;

whereby the processing method is learned without repeatedly determining the output data value for a given point in time.

14. A storage medium for storing a computer controllable program for processing input data and outputting the processed data, the program comprising:

a data processing step for processing the input data by a linear coupling of the input data and predetermined prediction coefficients and outputting the processed data; and

a real-time learning step for learning the prediction coefficients in real time so that the output data has higher quality as time elapses and for transferring the prediction coefficients to said data processing step,

wherein the real-time learning step includes formulating equations to obtain the prediction coefficients based on at least higher quality data than the input data, solving said equations, and outputting the prediction coefficients;

whereby the prediction coefficients are learned without repeatedly determining the output data value for a given point in time.