Image encoding apparatus and image decoding apparatus

Claim 2

An image encoding apparatus which performs lossless encoding, comprising: image inputting means for inputting image information; packing means for packing the image information inputted through said image inputting means in unit data of a fixed length; a plurality of pixel value predicting means, each of which predicts a value of the unit data outputted from said packing means by a predetermined manner; a plurality of prediction error calculating means, each of which calculates a prediction error between the value of the unit data outputted from said packing means and the value predicted by a corresponding one of said plurality of pixel value predicting means; prediction error selecting means for selecting an optimum prediction error from the prediction errors outputted from said plurality of prediction error calculating means; selecting means for selecting identification information of one of said plurality of pixel value predicting means which succeeds in prediction, and in the case where no pixel value predicting means succeeds in prediction, selecting the prediction error selected by said prediction error selecting means; encoding means for encoding the identification information or the prediction error selected by said selecting means by a predetermined manner; and code outputting means for outputting the code generated by said encoding means.

Claim 3

The image encoding apparatus according to claim 1, wherein said prediction error selecting means selects and outputs the unit data as the prediction error rather than the prediction errors outputted by said plurality of prediction error calculating means when a predetermined condition is satisfied.

Claim 4

The image encoding apparatus according to claim 2, wherein an operation used for prediction error calculation in each of said plurality of prediction error calculating means is one of subtraction and exclusive OR, and each of the subtraction and the exclusive OR is used by at least one of said plurality of prediction error calculating means.

Claim 5

The image encoding apparatus according to claim 2, wherein a selection signal is inputted to said prediction error selecting means from the outside of said image encoding apparatus and the unit data outputted from said packing means is outputted as the prediction error by said prediction error selecting means in accordance with the selection signal.

Claim 6

The image encoding apparatus according to claim 2, wherein said prediction error selecting means comprises: entropy calculating means for calculating entropy for each of the unit data outputted from said packing means and prediction errors calculated by said plurality of prediction error calculating means by a predetermined procedure; selection signal generating means for selecting a smallest entropy from those calculated by said entropy calculating means and generating a selection signal corresponding to the smallest entropy; and means for selecting one of the unit data outputted from said packing means and the prediction errors calculated by said plurality of prediction error calculating means in accordance with the selection signal.

Claim 7

The image encoding apparatus according to claim 1, wherein the image information input by the image inputting means includes digital data having different bit lengths, and the unit data output by the packing means has a constant, fixed bit length.

Claim 8

The image encoding apparatus according to claim 6, wherein the digital data includes pixel data having a pixel data bit length, and tag data having a tag data bit length less than the pixel data bit length, and the unit data also has said pixel data bit length.

Claim 9

An image decoding apparatus which decodes image information encoded by an image encoding apparatus which packs the image information in unit data of a fixed length, predicts a plurality of values of the unit data using a plurality of pixel value predicting means, calculates a plurality of prediction errors between the value of the unit data and the predicted values, selects an optimum prediction error from the plurality of prediction errors, selects identification information of one of said plurality of pixel value predicting means which succeeds in prediction, and when no pixel value predicting means succeeds in prediction, selects the optimum prediction error, and encodes the selected identification information or the optimum prediction error, comprising: code inputting means for inputting code encoded by said image encoding apparatus; decoding means for decoding the code inputted through said code inputting means by a predetermined manner; selecting means for outputting a direction to one of a plurality of predicting means or a prediction error based on information obtained by said decoding means; each of said plurality of predicting means predicting a value of the unit data inputted through said code inputting means in accordance with a direction from said selecting means; a plurality of prediction error adding means, each of which calculates a value of the unit data based on the prediction error sent from said selecting means and a predicted value predicted by a predetermined manner; predicted value selecting means for selecting an optimum predicted value from those calculated by said plurality of prediction error adding means; unpacking means for unpacking the unit data outputted from one of said predicting means and said predicted value selecting means by a predetermined manner to restore the image information; and image outputting means for outputting the image information sent from said unpacking means.

Claim 10

An image decoding apparatus which decodes image information encoded by an image encoding apparatus which packs the image information in unit data of a fixed length, predicts a plurality of values of the unit data using a plurality of pixel value predicting means, calculates a plurality of prediction errors between the value of the unit data and the predicted values, selects an optimum prediction error from the plurality of prediction errors, selects identification information of one of said plurality of pixel value predicting means which succeeds in prediction, and when no pixel value predicting means succeeds in prediction, selects the optimum prediction error, and encodes the selected identification information or the optimum prediction error, comprising: code inputting means for inputting code encoded by said image encoding apparatus; decoding means for decoding the code inputted through said code inputting means by a predetermined manner; selecting means for outputting a direction to one of a plurality of predicting means or a prediction error based on information obtained by said decoding means; each of said plurality of predicting means predicting a value of the unit data inputted through said code inputting means in accordance with a direction from said selecting means; a plurality of prediction error adding means, each of which calculates a value of the unit data based on the prediction error sent from said selecting means and a predicted value predicted by a predetermined manner; predicted value selecting means for selecting an optimum predicted value from those calculated by said plurality of prediction error adding means; unpacking means for unpacking the unit data outputted from one of said predicting means and said predicted value selecting means by a predetermined manner to restore the image information; and image outputting means for outputting the image information sent from said unpacking means.

Claim 11

The image decoding apparatus according to claim 8, wherein said predicted value selecting means selects and outputs the prediction error outputted from said selecting means as the predicted value rather than the predicted values outputted from said plurality of predicting means when a predetermined condition is satisfied.

Claim 12

The image decoding apparatus according to claim 9, wherein an operation used for calculation of the value of the unit data in each of said plurality of prediction error adding means is one of addition and exclusive OR, and each of the addition and the exclusive OR is used by at least one of said plurality of prediction error adding means.

Claim 13

The image decoding apparatus according to claim 9, wherein a selection signal is inputted to said predicted value selecting means from the outside of said image decoding apparatus and one of the prediction error sent from said selecting means and the values of the unit data calculated by said plurality of prediction error adding means is selected in accordance with the selection signal.

Claim 14

The image decoding apparatus according to claim 9, wherein said predicted value selecting means comprises: a plurality of prediction error calculating means, each of which calculates a prediction error between the predicted value predicted by a predetermined manner and the value predicted by a corresponding one of said plurality of predicting means or the optimum predicted value selected by said predicted value selecting means; entropy calculating means for calculating entropy for each of the values outputted by said plurality of predicting means and said predicted value selecting means and the prediction errors calculated by said plurality of prediction error calculating means by a predetermined procedure; selection signal generating means for selecting a smallest entropy from those calculated by said entropy calculating means and generating a selection signal corresponding to the smallest entropy; and means for selecting one of the prediction error sent from said selecting means and the predicted values calculated by said plurality of prediction error adding means in accordance with the selection signal.

Claim 15

The image decoding apparatus according to claim 8, wherein the image information encoded by the image encoding apparatus and restored by the unpacking means includes digital data having different bit lengths, and the unit data packed by the image encoding apparatus and unpacked by the unpacking means has a constant, fixed bit length.

Claim 16

The image decoding apparatus according to claim 13, wherein the digital data includes pixel data having a pixel data bit length, and tag data having a tag data bit length less than the pixel data bit length, and the unit data also has said pixel data bit length.

Claim 17

An image encoding method which performs lossless encoding, comprising the steps of: using an image inputting means to input image information; using a packing means to pack the image information inputted through said image inputting means in unit data of a fixed length; using a plurality of pixel value predicting means, each of which predicts a value of the unit data outputted from said packing means by a predetermined manner, to predict a plurality of values of the unit data; using a plurality of prediction error calculating means, each of which calculates a prediction error between the value of the unit data outputted from said packing means and the value predicted by a corresponding one of said plurality of pixel value predicting means; using a prediction error selecting means to select an optimum prediction error from the prediction errors outputted from said plurality of prediction error calculating means; using a selecting means to select identification information of one of said plurality of pixel value predicting means which succeeds in prediction, and when no pixel value predicting means succeeds in prediction, selecting the prediction error selected by said prediction error selecting means; using an encoding means to encode the identification information or the prediction error selected by said selecting means by a predetermined manner; and using a code outputting means to output the code generated by said encoding means.

Claim 18

An image encoding method which performs lossless encoding, comprising the steps of: using an image inputting means to input image information; using a packing means to pack the image information inputted through said image inputting means in unit data of a fixed length; using a plurality of pixel value predicting means, each of which predicts a value of the unit data outputted from said packing means by a predetermined manner, to predict a plurality of values of the unit data; using a plurality of prediction error calculating means, each of which calculates a prediction error between the value of the unit data outputted from said packing means and the value predicted by a corresponding one of said plurality of pixel value predicting means; using a prediction error selecting means to select an optimum prediction error from the prediction errors outputted from said plurality of prediction error calculating means; using a selecting means to select identification information of one of said plurality of pixel value predicting means which succeeds in prediction, and when no pixel value predicting means succeeds in prediction, selecting the prediction error selected by said prediction error selecting means; using an encoding means to encode the identification information or the prediction error selected by said selecting means by a predetermined manner; and using a code outputting means to output the code generated by said encoding means.

Claim 19

An image decoding method which decodes image information encoded by an image encoding apparatus which packs the image information in unit data of a fixed length, predicts a plurality of values of the unit data using a plurality of pixel value predicting means, calculates a plurality of prediction errors between the value of the unit data and the predicted values, selects an optimum prediction error from the plurality of prediction errors, selects identification information of one of said plurality of pixel value predicting means which succeeds in prediction, and when no pixel value predicting means succeeds in prediction, selects the optimum prediction error, and encodes the selected identification information or the optimum prediction error, comprising the steps of: using a code inputting means to input code encoded by said image encoding apparatus; using a decoding means to decode the code inputted through said code inputting means by a predetermined manner; using a selecting means to output a direction to one of a plurality of predicting means or a prediction error based on information obtained by said decoding means; each of said plurality of predicting means predicting a value of the unit data inputted through said code inputting means in accordance with a direction from said selecting means; using a plurality of prediction error adding means, each of which calculates a value of the unit data based on the prediction error sent from said selecting means and a predicted value predicted by a predetermined manner; using a predicted value selecting means to select an optimum predicted value from those calculated by said plurality of prediction error adding means; using an unpacking means to unpack the unit data outputted from one of said predicting means and said predicted value selecting means by a predetermined manner to restore the image information; and using an image outputting means to output the image information sent from said unpacking means.

Claim 20

An image decoding method which decodes image information encoded by an image encoding apparatus which packs the image information in unit data of a fixed length, predicts a plurality of values of the unit data using a plurality of pixel value predicting means, calculates a plurality of prediction errors between the value of the unit data and the predicted values, selects an optimum prediction error from the plurality of prediction errors, selects identification information of one of said plurality of pixel value predicting means which succeeds in prediction, and when no pixel value predicting means succeeds in prediction, selects the optimum prediction error, and encodes the selected identification information or the optimum prediction error, comprising the steps of: using a code inputting means to input code encoded by said image encoding apparatus; using a decoding means to decode the code inputted through said code inputting means by a predetermined manner; using a selecting means to output a direction to one of a plurality of predicting means or a prediction error based on information obtained by said decoding means; each of said plurality of predicting means predicting a value of the unit data inputted through said code inputting means in accordance with a direction from said selecting means; using a plurality of prediction error adding means, each of which calculates a value of the unit data based on the prediction error sent from said selecting means and a predicted value predicted by a predetermined manner; using a predicted value selecting means to select an optimum predicted value from those calculated by said plurality of prediction error adding means; using an unpacking means to unpack the unit data outputted from one of said predicting means and said predicted value selecting means by a predetermined manner to restore the image information; and using an image outputting means to output the image information sent from said unpacking means.