Data processing apparatus and data processing method

• US 7,088,768 B2
• Filed: 03/18/2004
• Issued: 08/08/2006
• 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, Kobayashi, Naoki, Ohta, Koji
• Primary Examiner(s): Kumar, Pankaj
• Application Number:
  US10/804,266
• Publication Number:
  US 20040174930A1
• Time in Patent Office: 873 Days
• Field of Search: 375/224, 375/225, 375/226, 375/295, 375/316, 375/219, 700/1, 700/90, 370/290, 370/291, 370/286, 370/282, 370/376, 379/406.08, 379/406.06, 379/406.01
• US Class Current: 375/224
• CPC Class Codes: H03H 21/0012 Digital adaptive filters
• CPC Class Codes: H04N 5/21 Circuitry for suppressing or minimising disturbance, e.g. moiré or halo

Abstract

A data processing apparatus processes input data and outputs the processed data. The data processing apparatus includes a data processing section for processing the input data by a predetermined processing method and outputting the processed data, an input-data evaluation section for evaluating the input data, an output-data evaluation section for evaluating the output data, and a real-time learning section for controlling such that the processing method is learned in real time according to the evaluation results obtained by the input-data evaluation section and the output-data evaluation section and the data processing section processes the input data according to the learned processing method, and thereby, improves the output data every moment.

Claims

1. A data processing apparatus for processing input data to create processed data for outputting, comprising:

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

an input data evaluator configured to evaluate the input data and calculate a reliability of the input data;

an output data evaluator configured to evaluate the output data and calculate a reliability of the output data, and

a real time learning portion consisting of a single component configured to learn the processing of the data processor in real time using the reliability of the input data calculated by said input data evaluator and the reliability of the output data calculated by said output data evaluator to create learned processing and control the data processor to process the input data according to the learned processing.

2. A data processing apparatus according to claim 1, further comprising an input data storage unit for storing a predetermined number of time-sequentially input data.

3. A data processing apparatus according to claim 2, wherein said input data evaluator calculates a dispersion of the predetermined number of the time-sequentially input data stored by said input data storage unit and evaluates current input data according to the dispersion.

4. A data processing apparatus according to claim 2, wherein said input data evaluator calculates an average of the predetermined number of the time-sequentially input data stored by said input data storage unit, and evaluates current input data according to an error of each of the input data against the average.

5. A data processing apparatus according to claim 2, wherein said input data evaluator calculates a dispersion and an average of the predetermined number of the time-sequentially input data stored by said input data storage unit;

obtains an error of each input data against the average; and

evaluates current input data according to the dispersion and a respective error.

6. A data processing apparatus according to claim 1, further comprising an output data storage unit for storing the output data, wherein said data processor adds previous output data stored by said output data storage unit and current input data to obtain the output data corresponding to the current input data.

7. A data processing apparatus according to claim 6, wherein said real time learning portion learns a predetermined weight coefficient according to the reliability of the input data and the reliability of the output data, and

said data processor obtains the output data corresponding to the current input data according to the weight coefficient.

8. A data processing method for processing input data and creating processed data for outputting comprising the steps of:

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

evaluating the input data and calculating a reliability of the input data;

evaluating the output data and calculating a reliability of the output data, and

learning the step of processing the input data in real time in a single component using the calculated reliability of the input data and the calculated reliability of the output data to create a learned processing and the input data is processed by the step of processing the input data by the learned processing.

9. A data processing method according to claim 8, further comprising an input data storing step of storing a predetermined number of time-sequentially input data.

10. A data processing method according to claim 9, wherein, in the step of evaluating the input data, a dispersion of the predetermined number of the time-sequentially input data stored in the input data storing step is calculated and current input data is evaluated according to the dispersion.

11. A data processing method according to claim 9, wherein, in the step of evaluating the input data, an average of the predetermined number of the time-sequentially input data stored in the input data storing step is calculated, and current input data is evaluated according to an error of each input data against the average.

12. A data processing method according to claim 9, wherein, in the step of evaluating the input data, a dispersion and an average of the predetermined number of the input data stored in the input data storing step are calculated;

an error of each input data against the average is obtained; and

current input data is evaluated according to the dispersion and a respective error.

13. A data processing method according to claim 8, further comprising an output data storing step of storing the output data, wherein, in the step of processing the input data, previous output data stored in the output data storing step and current input data are added to obtain the output data corresponding to the current input data.

14. A data processing method according to claim 13, wherein a predetermined weight coefficient is learned according to the reliability of the input data and the reliability of the output data in the step of learning, and

the output data corresponding to the current input data is obtained according to the weight coefficient in the step of processing the input data.

15. A storage medium storing a computer-controllable program for processing input data and creating processed data for outputting the program comprising the steps of:

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

evaluating the input data and calculating a reliability of the input data;

evaluating the output data and calculating a reliability of the output data, and

learning the step of processing the input data in real time in a single component using the calculated reliability of the input data and the calculated reliability of the output data to create a learned processing, and the input data is processed by the step of processing the input data by the learned processing.