Data processing apparatus

A data processing method for processing encoded data obtained by encoding data, characterized by comprising:

• a decoding step of decoding the encoded data and of outputting decoded data;
• a re-encoding step of encoding the decoded data and of outputting re-encoded data;
• a prediction-tap generation step of generating a prediction tap used for a predetermined prediction calculation with tap coefficients obtained by learning, from the decoded data and information obtained from the re-encoded data;
• a tap-coefficient acquisition step of acquiring the tap coefficients;
• and
• a prediction step of obtaining a prediction value corresponding to master data used as a master in the learning, by the predetermined prediction calculation with the use of the prediction tap and the tap coefficients.

A program for making a computer process encoded data obtained by encoding data, characterized by comprising:

• a decoding step of decoding the encoded data and of outputting decoded data;
• a re-encoding step of encoding the decoded data and of outputting re-encdded data;
• a prediction-tap generation step of generating a prediction tap used for a predetermined prediction calculation with tap coefficients obtained by learning, from the decoded data and information obtained from the re-encoded data;
• a tap-coefficient acquisition step of acquiring the tap coefficients;
• and
• a prediction step of obtaining a prediction value corresponding to master data used as a master in the learning, by the predetermined prediction calculation with the use of the prediction tap and the tap coefficients.

A recording medium storing a program which makes a computer process encoded data obtained by encoding data, characterized by recording the program comprising:

• a decoding step of decoding the encoded data and of outputting decoded data;
• a re-encoding step of encoding the decoded data and of outputting re-encoded data;
• a prediction-tap generation step of generating a prediction tap used for a predetermined prediction calculation with tap coefficients obtained by learning, from the decoded data and information obtained from the re-encoded data;
• a tap-coefficient acquisition step of acquiring the tap coefficients;
• and
• a prediction step of obtaining a prediction value corresponding to master data used as a master in the learning, by the predetermined prediction calculation with the use of the prediction tap and the tap coefficients.

A data processing apparatus for learning predetermined tap coefficients used for processing encoded data obtained by encoding data, characterized by comprising:

• apprentice-data generation means for generating from master data serving as a master, decoded data obtained by encoding the master data and then decoding it, and information obtained from re-encoded data obtained by encoding the decoded data, as apprentice data serving as an apprentice;
• prediction-tap generation means for generating a prediction tap used for predicting the master data, from the apprentice data;
• and
• learning means for performing learning such that the prediction error of a prediction value for the master data, obtained by performing a predetermined prediction calculation by using the prediction tap and tap coefficients, is statistically minimum, and for obtaining the tap coefficients.

A data processing method for learning predetermined tap coefficients used for processing encoded data obtained by encoding data, characterized by comprising:

• an apprentice-data generation step of generating from master data serving as a master, decoded data obtained by encoding the master data and then decoding it, and information obtained from re-encoded data obtained by encoding the decoded data, as apprentice data serving as an apprentice;
• a prediction-tap generation step of generating a prediction tap used for predicting the master data, from the apprentice data;
• and
• a learning step of performing learning such that the prediction error of a prediction value for the master data, obtained by a predetermined prediction calculation by using the prediction tap and tap coefficients, is statistically minimum, and of obtaining the tap coefficients.

A program for making a computer execute data processing for learning predetermined tap coefficients used for processing encoded data obtained by encoding data, characterized by comprising:

• an apprentice-data generation step of generating from master data serving as a master, decoded data obtained by encoding the master data and then decoding it, and information obtained from re-encoded data obtained by encoding the decoded data, as apprentice data serving as an apprentice;
• a prediction-tap generation step of generating a prediction tap used for predicting the master data, from the apprentice data;
• and
• a learning step of performing learning such that the prediction error of a prediction value for the master data, obtained by a predetermined prediction calculation by using the prediction tap and tap coefficients, is statistically minimum, and of obtaining the tap coefficients.

A recording medium storing a program which makes a computer execute data processing for learning predetermined tap coefficients used for processing encoded data obtained by encoding data, characterized by recording the program comprising:

• an apprentice-data generation step of generating from master data serving as a master, decoded data obtained by encoding the master data and then decoding it, and information obtained from re-encoded data obtained by encoding the decoded data, as apprentice data serving as an apprentice;
• a prediction-tap generation step of generating a prediction tap used for predicting the master data, from the apprentice data;
• and
• a learning step of performing learning such that the prediction error of a prediction value for the master data, obtained by a predetermined prediction calculation by using the prediction tap and tap coefficients, is statistically minimum, and of obtaining the tap coefficients.

A data processing apparatus for performing class classification which classifies information, characterized by comprising:

• decoding means for decoding encoded data obtained by encoding data and for outputting decoded data;
• re-encoding means for encoding the decoded data and for outputting re-encoded data;
• class-tap generation means for generating a class tap used for the class classification, from the decoded data and information obtained from the re-encoded data;
• and
• class classification means for performing class classification according to the class tap.

A data processing method for performing class classification which classifies information, characterized by comprising:

• a decoding step of decoding encoded data obtained by encoding data and for outputting decoded data;
• a re-encoding step of encoding the decoded data and for outputting re-encoded data;
• a class-tap generation step of generating a class tap used for the class classification, from the decoded data and information obtained from the re-encoded data;
• and
• a class classification step of performing class classification according to the class tap.

A program for making a computer execute data processing for performing class classification which classifies information, characterized by comprising:

• a decoding step of decoding encoded data obtained by encoding data and for outputting decoded data;
• a re-encoding step of encoding the decoded data and for outputting re-encoded data;
• a class-tap generation step of generating a class tap used for the class classification, from the decoded data and information obtained from the re-encoded data;
• and
• a class classification step of performing class classification according to the class tap.

A recording medium storing a program which makes a computer execute data processing for performing class classification which classifies information, characterized by recording the program comprising:

• a decoding step of decoding encoded data obtained by encoding data and for outputting decoded data;
• a re-encoding step of encoding the decoded data and for outputting re-encoded data;
• a class-tap generation step of generating a class tap used for the class classification, from the decoded data and information obtained from the re-encoded data;
• and
• a class classification step of performing class classification according to the class tap.