Image processing method and apparatus

Claim 2

An image processing apparatus for processing image data in units of blocks each consisting of a plurality of pixels, comprising: reversible compression means for reversibly compressing a block which satisfies D ≦ L (D is the compressed data amount obtained when the block is reversibly compressed, and L is a predetermined value); and irreversible compression means for irreversibly compressing a block which satisfies D > L.

Claim 3

The apparatus according to claim 2, wherein said reversible compression means comprises: sorting means for one-dimensionally sorting pixel data of the block; latch means for latching the pixel data output from said sorting means; first comparison means for comparing first pixel data latched by said latch means with second pixel data output from said sorting means after the first pixel data, and outputting a first comparison result; second comparison means for comparing pixel data held in advance in at least one holding means with the second pixel data, and outputting at least one second comparison result; updating means for, when the first comparison result indicates a non-coincidence, and the second comparison result indicates a coincidence, updating pixel data in the holding means which holds the pixel data coinciding with the second pixel data to the pixel data latched by said latch means, and for, when both the first and second comparison results indicate a non-coincidence, updating pixel data in the holding means which holds the oldest pixel data to the pixel data latched by said latch means; and output means for, when the first comparison result indicates a coincidence, outputting the first comparison result, for, when the first comparison result indicates a non-coincidence, and the second comparison result indicates a coincidence, outputting the first comparison result, and the second comparison result in an order named, and for, when both the first and second comparison results indicate a non-coincidence, outputting the first comparison result, the second comparison result, and the second pixel data in an order named.

Claim 4

The apparatus according to claim 3, wherein the first comparison result is a 1-bit signal, and the second comparison result is a signal consisting of at least 1 bit.

Claim 5

The apparatus according to claim 3, wherein said reversible compression means further comprises: calculation means for calculating the compressed data amount D on the basis of the first and second comparison results; discrimination means for discriminating a relationship between the compressed data amount D obtained by said calculation means and the limit value L; and recovery means for, when a discrimination result of said discrimination means indicates D > L, restoring the pixel data latched by said latch means and pixel data held in the holding means to values before the beginning of processing of a present block.

Claim 6

The apparatus according to claim 5, wherein said calculation means includes: a first counter for obtaining a first count value by counting the number of times of non-coincidences indicated by the first comparison result; and a second counter for obtaining a second count value by counting the number of times of non-coincidences simultaneously indicated by the first and second comparison results, and said calculation means calculates the compressed data amount D on the basis of the first and second count values.

Claim 7

The apparatus according to claim 6, wherein the compressed data amount D is calculated by: D = m x n + R x P + T x Q where m x n is the number of pixels of the block, P is the first count value, Q is the second count value, R is the number of bits of the second comparison result, and T is the number of bits of the pixel data.

Claim 8

The apparatus according to claim 5, wherein said discrimination means includes: a difference calculation unit for calculating a difference S (= L - D) between the limit value L and the compressed data amount D of a processed block; and an accumulation unit for, when the discrimination result in the present block to be processed indicates D ≦ L, updating the limit value L with a sum (C + S) of a predetermined value C and the difference S obtained by said difference calculation unit.

Claim 9

The apparatus according to claim 2, wherein said irreversible compression means comprises: color conversion means for converting a color space of image data; orthogonal transform means for orthogonally transforming image data converted by said color conversion means; quantization means for quantizing transform coefficients output from said orthogonal transform means; and encoding means for encoding quantized data output from said quantization means.

Claim 10

The apparatus according to claim 9, wherein said encoding means distributes and outputs codes obtained by encoding the quantized data in accordance with spatial frequencies of the quantized data.

Claim 11

An image processing method for expanding compressed data obtained by compressing image data in units of blocks each consisting of a plurality of pixels, comprising the steps of: expanding the compressed data by selecting one of reversible expansion processing and irreversible expansion processing on the basis of information indicating a compression method of the compressed data; and reproducing original image data by synthesizing image data expanded by the reversible expansion processing and image data expanded by the irreversible expansion processing.

Claim 12

The method according to claim 11, wherein the reversible expansion processing includes: the first input step of inputting a first bit corresponding to a pixel of the compressed data; the second input step of inputting compression information consisting of at least 1 bit of a second bit and the following bits corresponding to the pixel when the first bit is '1'; the exchange step of exchanging pixel data latched by latch means with pixel data held in at least one holding means corresponding to a predetermined value when the compression information has the predetermined value; the updating step of updating the pixel data held by said holding means to the pixel data latched by said latch means, and thereafter, updating pixel data to be latched by said latch means to pixel data following the compression information when the compression information is different from the predetermined value; and the output step of outputting pixel data latched by said latch means.

Claim 13

The method according to claim 11, wherein the irreversible compression processing includes: the decoding step of decoding the compressed data; the inverse quantization step of inversely quantizing quantized data decoded in the decoding step; the inverse orthogonal transform step of inversely orthogonally transforming orthogonal transform coefficients inversely quantized in the inverse quantization step; and the color conversion step of converting a color space of image data output in the inverse orthogonal transform step.

Claim 14

An image processing apparatus for expanding compressed data obtained by compressing image data in units of blocks each consisting of a plurality of pixels, comprising: reversible expansion means for expanding reversibly compressed data; irreversible expansion means for expanding irreversibly compressed data; selection means for selecting one of said reversible expansion means and said irreversible expansion means on the basis of information indicating a compression method of the compressed data; and synthesizing means for reproducing original image data by synthesizing image data expanded by said reversible expansion means and image data expanded by said irreversible expansion means.

Claim 15

An image processing method for reversibly compressing image data, comprising: the latch step of latching input pixel data; the first comparison step of comparing first pixel data latched in the latch step with second pixel data input after the first pixel data, and outputting a first comparison result; the second comparison step of comparing at least two different first and second holding data, which are held in advance, with the second pixel data, and outputting a second comparison result; the updating step of updating the first holding data to the first pixel data when the second comparison result indicates a coincidence between the first holding data and the second pixel data, and of updating the second holding data to the first holding data, and thereafter, updating the first holding data to the first pixel data when the second comparison result indicates a coincidence between the second holding data and the second pixel data or when both the first and second comparison results indicate a non-coincidence; and the output step of outputting a first code when the first comparison result indicates a coincidence, of outputting a second code when the second comparison result indicates a coincidence between the first holding data and the second pixel data, of outputting a third code when the second comparison result indicates a coincidence between the second holding data and the second pixel data, and of outputting a fourth code and the second pixel data in an order named when both the comparison results indicate a non-coincidence.

Claim 16

The method according to claim 15, wherein the first code is 1-bit data '1', the second code is 2-bit data '01', the third code is 3-bit data '001', and the fourth code is 3-bit data '000'.

Claim 17

An image processing apparatus for reversibly compressing image data, comprising: first holding means for holding input pixel data; second holding means for holding in advance at least two different first and second holding data; first comparison means for comparing first pixel data held by said first holding means with second pixel data input after the first pixel data, and outputting a first comparison result; second comparison means for comparing the first and second holding data held by said second holding means with the second pixel data, and outputting a second comparison result; updating means for, when the second comparison result indicates a coincidence between the first holding data and the second pixel data, updating the first holding data to the first pixel data, and for, when the second comparison result indicates a coincidence between the second holding data and the second pixel data or when both the first and second comparison results indicate a non-coincidence, updating the second holding data to the first holding data, and thereafter, updating the first holding data to the first pixel data; and output means for outputting a first code when the first comparison result indicates a coincidence, for outputting a second code when the second comparison result indicates a coincidence between the first holding data and the second pixel data, for outputting a third code when the second comparison result indicates a coincidence between the second holding data and the second pixel data, and for outputting a fourth code and the second pixel data in an order named when both the first and second comparison results indicate a non-coincidence.

Claim 18

The apparatus according to claim 17, wherein the first code is 1-bit data '1', the second code is 2-bit data '01', the third code is 3-bit data '001', and the fourth code is 3-bit data '000'.

Claim 19

An image processing method for reproducing image data by expanding reversibly compressed data, comprising: the output step of outputting one of at least three different pixel data stored in holding means or pixel data included in the reversibly compressed data in accordance with a priority order and 3-bit data of first to third bits of the input reversibly compressed data; and the updating step of updating the pixel data held by said holding means and the priority order in accordance with the priority order and the 3-bit data.

Claim 20

An image processing apparatus for reproducing image data by expanding reversibly compressed data, comprising: holding means for holding in advance at least three different pixel data; storage means for storing a priority order of the pixel data held by said holding means; output means for outputting one of the pixel data stored in said holding means or pixel data included in the reversibly compressed data in accordance with the priority order and 3-bit data of first to third bits of the input reversibly compressed data; and updating means for updating the pixel data held by said holding means and the priority order in accordance with the priority order and the 3-bit data.

Claim 21

The apparatus according to claim 20, wherein said output means outputs: pixel data having the highest priority when the first bit of the reversibly compressed data is '1'; pixel data having the second highest priority when the first and second bits of the reversibly compressed data are '01'; pixel data having the third highest priority when the first to third bits of the reversibly compressed data are '001'; and pixel data starting from a fourth bit of the reversibly compressed data when the first to third bits of the reversibly compressed data are '000'.

Claim 22

The apparatus according to claim 20, wherein said updating means exchanges the priority order of the pixel data currently having the highest priority and the pixel data currently having the second highest priority when the first and second bits of the reversibly compressed data are '01', updates the priority order of the pixel data currently having the highest priority to the second highest priority, the priority order of the pixel data currently having the second highest priority to the third highest priority, and the priority order of the pixel data currently having the third highest priority to the highest priority when the first to third bits of the reversibly compressed data are '001', and updates the pixel data currently having the third highest priority to pixel data starting from a fourth bit of the reversibly compressed data, the priority order of the pixel data currently having the highest priority to the second highest priority, the priority order of the pixel data currently having the second highest priority to the third highest priority, and the priority order of the pixel data currently having the third highest priority to the highest priority when the first to third bits of the reversibly compressed data are '000'.

Claim 23

An image processing method for processing image data including pixel type information in units of blocks each consisting of m x n pixels, comprising: the analysis step of outputting first block information according to first pixel type information and second block information according to second pixel type information in units of blocks; the replacing step of replacing a pixel having the first pixel type information with first pixel data, and replacing a pixel having the second pixel type information and the same value as the first pixel data with second pixel data; the reversible compression step of reversibly compressing the pixel data replaced in the replacing step in accordance with the first block information and the second block information, and outputting reversibly compressed data and an area discrimination result; the interpolation step of interpolating pixels according to the pixel type information and the area discrimination result to pixel data according to other pixels in a block; the irreversible compression step of irreversibly compressing the block interpolated in the interpolation step in accordance with the first block information and the area discrimination result; the block information output step of outputting third block information in accordance with the second block information and the area discrimination result; and the holding step of holding the reversibly compressed data output in the reversible compression step, and outputting the reversibly compressed data in accordance with the third block information.

Claim 24

An image processing apparatus for processing image data including pixel type information in units of blocks each consisting of m x n pixels, comprising: analysis means for outputting first block information according to first pixel type information and second block information according to second pixel type information in units of blocks; replacing means for replacing a pixel having the first pixel type information with first pixel data, and replacing a pixel having the second pixel type information and the same value as the first pixel data with second pixel data; reversible compression means for reversibly compressing the pixel data replaced by said replacing means in accordance with the first block information and the second block information, and outputting reversibly compressed data and an area discrimination result; interpolation means for interpolating pixels according to the pixel type information and the area discrimination result to pixel data according to other pixels in a block; irreversible compression means for irreversibly compressing the block interpolated by said interpolation means in accordance with the first block information and the area discrimination result; block information output means for outputting third block information in accordance with the second block information and the area discrimination result; and holding means for holding the reversibly compressed data output from said reversible compression means, and outputting the reversibly compressed data in accordance with the third block information.

Claim 25

The apparatus according to claim 24, wherein the first pixel type information indicates a natural image pixel, and the second pixel type information indicates a pixel other than a natural image.

Claim 26

The apparatus according to claim 25, wherein the first block information indicates the presence/absence of a pixel having the first pixel type information in the block, and the second block information indicates the presence/absence of a pixel having the second pixel type information in the block.

Claim 27

The apparatus according to claim 25, wherein when the first block information and the second block information indicate a block consisting of only pixels other than a natural image, said reversible compression means calculates a compressed data amount D to be obtained upon reversible compression of the block, and outputs an area discrimination result indicating a first area and reversibly compressed data if D ≦ L (limit value), or outputs only an area discrimination result indicating a second area if D = m x n or D > L; when the first block information and the second block information indicate a block including both natural image pixels and pixels other than the natural image, said reversible compression means calculates the compressed data amount D to be obtained upon reversible compression of the block, and outputs an area discrimination result indicating the first area and reversibly compressed data if D ≦ L, or outputs only an area discrimination result indicating the second area if D > L; and when the first block information and the second block information indicate a block consisting of only natural image pixels, said reversible compression means outputs only an area discrimination result indicating the second area.

Claim 28

The apparatus according to claim 27, wherein said reversible compression means comprises: a first holding unit for holding input pixel data; a second holding unit for holding at least two different first and second holding data; a first comparison unit for comparing first pixel data held by said first holding unit with second pixel data input after the first pixel data, and outputting a first comparison result; a second comparison unit for comparing the first and second holding data held by said second holding unit with the second pixel data, and outputting a second comparison result; an updating unit for, when the second comparison result indicates a coincidence between the first holding data and the second pixel data, updating the first holding data to the first pixel data, and for, when the second comparison result indicates a coincidence between the second holding data and the second pixel data or when both the first and second comparison results indicate a non-coincidence, updating the second holding data to the first holding data, and thereafter, updating the first holding data to the first pixel data; and an output unit for outputting a first code when the first comparison result indicates a coincidence, for outputting a second code when the second comparison result indicates a coincidence between the first holding data and the second pixel data, for outputting a third code when the second comparison result indicates a coincidence between the second holding data and the second pixel data, and for outputting a fourth code and the second pixel data in an order named when both the first and second comparison results indicate a non-coincidence.

Claim 29

The apparatus according to claim 28, wherein the first code is 1-bit data '1', the second code is 2-bit data '01', the third code is 3-bit data '001', and the fourth code is 3-bit data '000'.

Claim 30

The apparatus according to claim 27, wherein when the area discrimination result indicates the first area, said interpolation means interpolates pixels other than the natural image indicated by the pixel type information.

Claim 31

The apparatus according to claim 27, wherein said irreversible compression means irreversibly compresses a block corresponding to the first block information indicating the presence of the natural image pixel, or a block corresponding to the area discrimination information indicating the second area.

Claim 32

The apparatus according to claim 31, wherein said irreversible compression means comprises: a color conversion unit for converting a color space of image data; an orthogonal transform unit for orthogonally transforming image data converted by said color conversion unit; a quantization unit for quantizing transform coefficients output from said orthogonal transform unit; and an encoding unit for encoding quantized data output from said quantization unit.

Claim 33

The apparatus according to claim 32, wherein said encoding unit distributes and outputs codes obtained by encoding the quantized data in accordance with spatial frequencies of the quantized data.

Claim 34

The apparatus according to claim 27, wherein the third block information is 1-bit information which is '1' when the second block information indicates that a block includes a pixel other than the natural image, and the area discrimination result indicates the first area.

Claim 35

The apparatus according to claim 34, wherein said holding means outputs held reversibly compressed data when the third block information is '1'.

Claim 36

An image processing method for expanding compressed data obtained by compressing image data in units of blocks each consisting of m x n pixels, comprising the steps of: irreversibly expanding the compressed data when method information indicating a compression method in units of blocks indicates irreversible compression; irreversibly expanding a pixel corresponding to pixel type information, obtained by analyzing the compressed data, indicating a predetermined pixel type, and reversibly expanding a pixel corresponding to the pixel type information indicating a pixel type other than the predetermined pixel type when the method information indicates reversible compression; and reproducing original image data by synthesizing reversibly expanded image data and irreversibly expanded image data.

Claim 37

An image processing apparatus for expanding compressed data obtained by compressing image data in units of blocks each consisting of m x n pixels, comprising: irreversible expansion means for irreversibly expanding the compressed data; reversible expansion means for reversibly expanding the compressed data, and outputting pixel data and pixel type information; and selection means for selecting one of the output from said reversible expansion means and the output from said irreversible expansion means in accordance with method information indicating a compression method in units of blocks, and the pixel type information, wherein original image data is reproduced by synthesizing the pixel data expanded by said reversible expansion means and image data expanded by said irreversible expansion means.

Claim 38

The apparatus according to claim 37, wherein said irreversible expansion means comprises: decoding means for decoding the compressed data; inverse quantization means for inversely quantizing quantized data decoded by said decoding means; inverse orthogonal transform means for inversely orthogonally transforming orthogonal transform coefficients inversely quantized by said inverse quantization means; and color conversion means for converting a color space of image data output from said inverse orthogonal transform means.

Claim 39

The apparatus according to claim 37, wherein said reversible expansion means comprises: a latch unit for latching predetermined pixel data; a holding unit for holding in advance at least three different holding data; a storage unit for storing updating information of the holding data held by said holding unit; an updating unit for, when 1-bit data of a first bit of the compressed data is '1', and 2-bit data of second and third bits is other than '11', exchanging the pixel data latched by said latch unit with holding data corresponding to the 2-bit data, and for, when the 1-bit data is '1', and the 2-bit data is '11', updating holding data indicated by the updating information to the pixel data latched by said latch unit, and thereafter updating the pixel data latched by said latch unit to pixel data of a fourth bit and the following bits of the compressed data; and an output unit for outputting area information accordance with 3-bit data of the first to third bits of the compressed data, and outputting the pixel data latched by said latch unit.

Claim 40

The apparatus according to claim 39, wherein said output unit outputs area information having the same value as a value of an immediately preceding pixel when the 1-bit data is '0', outputs area information indicating "area 1" when the 1-bit data is '1' and the 2-bit data is other than '11', and outputs pixel type information indicating "area 2" if pixel data starting from the fourth bit of the compressed data has a predetermined value or pixel type information indicating "area 1" if the pixel data has a value other than the predetermined value when the 3-bit data is '111'.

Claim 41

The apparatus according to claim 40, wherein when the method information indicates reversible compression, and the pixel type information indicates "area 1", said selection means selects the output from said reversible expansion means; otherwise, said selection means selects the output from said irreversible expansion means.