Image processing method and apparatus

Claim 2

An image processing method comprising: an image separation process for separating an input image into plural blocks; a temporary digitization threshold value calculating process for calculating a temporary digitization threshold value for each of the blocks separated by the image separation process; a subject presence evaluation process for determining whether a predetermined subject, comprising text characters, is present in each of the blocks separated by the image separation process; a block information storing process for storing as the block information the digitization threshold values of each block for which the final digitization threshold value has been obtained, and the temporary digitization threshold value and subject presence information obtained for each block by the temporary digitization threshold value calculating process and the subject presence evaluation process, respectively; a final digitization threshold value determining process for determining the final digitization threshold value for each of the blocks using the block information of the one block and peripheral ones of the blocks stored by the block information storing process; and a digitization process for digitizing the input image using the final digitization threshold value determined for each block by the final digitization threshold value determining process.

Claim 3

An image processing method according to claim 1 wherein the temporary digitization threshold value calculating process is characterized by: a histogram generating process for generating a density histogram of the input image; an among-class dispersion calculating process for calculating the among-class dispersion value of each density value in the density histogram obtained by the histogram generating process; and a maximum value detecting process for detecting the maximum among-class dispersion values obtained by the among-class dispersion calculating process, and outputting as the temporary digitization threshold value the density value of the among-class dispersion value peak.

Claim 4

An image processing method according to claim 1 wherein the temporary digitization threshold value calculating process is characterized by: a histogram generating process for generating a density histogram of the input image; a peak detection process for detecting only the peaks required to obtain the temporary digitization threshold value from the density histogram obtained by the histogram generating process; and a temporary digitization threshold value calculating process for obtaining the temporary digitization threshold value based on the peaks detected by said peak detection process.

Claim 5

An image processing method according to claim 3 wherein the peak detection process determines, when there are plural peak candidates in the density histogram obtained by the histogram generating process, which of the plural peak candidates is the peak required to obtain the temporary digitization threshold value based on the frequencies of occurrence of the density value in each of the plural peak candidates, and the comparative relationship between the frequency of occurrence of the density value in peak candidates being compared.

Claim 6

An image processing method according to claim 4 wherein the temporary digitization threshold value calculating process is characterized by: defining as the temporary digitization threshold value the density value corresponding to a predetermined position at either the rise side or the drop side of the detected peak when there is only one peak detected by the peak detection process, and defining as the temporary digitization threshold value the density value corresponding to a predetermined position between adjacent peaks when plural peaks are detected by the peak detection process.

Claim 7

An image processing method according to claim 1 wherein the subject presence evaluation process is characterized by: a feature quantity extraction process for extracting a feature quantity based on the density distribution of the input image in each separated block; and a subject presence discrimination signal output process for determining the subject presence by comparing the feature quantity obtained by the feature quantity extraction process with a predetermined evaluation value supplied as an input parameter of the comparison operation.

Claim 8

An image processing method according to claim 6 wherein the feature quantity of the density distribution is the density dispersion value of the input image in each block.

Claim 9

An image processing method according to claim 6 wherein the feature quantity of the density distribution is the maximum among-class dispersion value obtained based on the density histogram obtained from the input image in each block.

Claim 10

An image processing method according to claim 6 wherein the feature quantity of the density distribution is the maximum density difference of the input image in each block.

Claim 11

An image processing method according to claim 1 wherein the subject presence evaluation process is characterized by: a feature quantity extraction process for extracting as the feature quantity at least two of following three feature quantities of the input image in each block: the density dispersion value, the maximum among-class dispersion value, and the maximum density difference; and a subject presence discrimination signal output process for determining the subject presence by comparing the feature quantities obtained by the feature quantity extraction process with predetermined evaluation values supplied as input parameters of the comparison operation.

Claim 12

An image processing method according to claim 1 wherein the final digitization threshold value determining process defines a reference range comprising plural blocks around the target block for which the digitization threshold value is to be determined, and determines the final digitization threshold value of the target block based on the temporary digitization threshold value of the target block and an already-determined final digitization threshold value or temporary digitization threshold value of at least one block in said reference range.

Claim 13

An image processing method according to claim 11 where when subject text or objects are determined to not be present in any blocks in the reference range, excluding the target block, the digitization threshold value of the target block is obtained by determining whether subject text or objects are present in the target block, and is determined to be the temporary digitization threshold value of the target block when subject text or objects are determined to be present in the target block, and is determined by referencing an already-determined digitization threshold value in the reference range when subject text or objects are determined to not be present in the target block.

Claim 14

An image processing method according to claim 11 where when subject text or objects are determined to be present in one or more blocks of the reference range, excluding the target block, the digitization threshold value of the target block is obtained by determining whether subject text or objects are present in the target block, and is determined from a temporary digitization threshold value in the reference range and the temporary digitization threshold value of the target block when subject text or objects are determined to be present in the target block, and is determined from the temporary digitization threshold values in the reference range when subject text or objects are determined to not be present in the target block.

Claim 15

An image processing method according to claim 11 or claim 13 wherein the final digitization threshold value of the target block is obtained by calculation from the temporary digitization threshold value of the target block and the temporary digitization threshold value of at least one block in the reference range.

Claim 16

An image processing method according to claim 11 or claim 13 wherein the final digitization threshold value of the target block is determined by means of a logic operation obtaining the minimum value of the temporary digitization threshold values of the target block and each block in the reference range, and defining this minimum temporary digitization threshold value as the final digitization threshold value of the target block.

Claim 17

An image processing method according to claim 15 where when the minimum temporary digitization threshold value in the reference range is defined as the final digitization threshold value to be obtained, the final digitization threshold value of the target block is defined, depending upon the conditions, to be the sum of a predetermined value added to the minimum temporary digitization threshold value in the reference range.

Claim 18

An image processing method comprising: an image separation process for separating an input image into plural blocks; a temporary digitization threshold value calculating process for calculating a temporary digitization threshold value for each of the blocks separated by the image separation process; a subject presence evaluation process for determining whether a predetermined subject, comprising text characters, is present in each of the blocks separated by the image separation process; a block information storing process for storing as the block information the digitization threshold values of each block for which the final digitization threshold value has been obtained, and the temporary digitization threshold value and subject presence information obtained for each block by the temporary digitization threshold value calculating process and the subject presence evaluation process, respectively; a final digitization threshold value determining process for determining the final digitization threshold value for each block using the block information stored by the block information storing process; and a digitization process for digitizing the input image using the final digitization threshold value determined for each block by the final digitization threshold value determining process, wherein the final digitization threshold value determining process defines a reference range comprising plural blocks around the target block for which the digitization threshold value is to be determined, and determines the final digitization threshold value of the target block based on the temporary digitization threshold value of the target block and an already-determined final digitization threshold value or temporary digitization threshold value of at least one block in said reference range, and wherein the reference range is defined relative to the target block: in the line direction as a variable number of blocks dependent upon the detected subject presence in each block, and in the column direction as at least one block line above and one block line below the block line of the target block.

Claim 19

An image processing method comprising: an image separation process for separating an input image into plural blocks; a temporary digitization threshold value calculating process for calculating a temporary digitization threshold value for each of the blocks separated by the image separation process; a subject presence evaluation process for determining whether a predetermined subject, comprising text characters, is present in each of the blocks separated by the image separation process; a block information storing process for storing as the block information the digitization threshold values of each block for which the final digitization threshold value has been obtained, and the temporary digitization threshold value and subject presence information obtained for each block by the temporary digitization threshold value calculating process and the subject presence evaluation process, respectively; a final digitization threshold value determining process for determining the final digitization threshold value for each block using the block information stored by the block information storing process; and a digitization process for digitizing the input image using the final digitization threshold value determined for each block by the final digitization threshold value determining process, wherein the final digitization threshold value determining process defines a reference range comprising plural blocks around the target block for which the digitization threshold value is to be determined, and determines the final digitization threshold value of the target block based on the temporary digitization threshold value of the target block and an already-determined final digitization threshold value or temporary digitization threshold value of at least one block in said reference range, and wherein the reference range is defined relative to the target block: in the line direction as a variable number of blocks dependent upon the detected subject presence in each block, and in the column direction as at least one block line above and one block line below the block line of the target block.

Claim 20

An image processing apparatus comprising: an image separator for separating an input image into plural blocks; a temporary digitization threshold value calculator for calculating a temporary digitization threshold value for each of the blocks separated by the image separator; a subject presence evaluator for determining whether a predetermined subject, comprising text characters, is present in each of the blocks separated by the image separator; a block information store for storing as the block information the digitization threshold values of each block for which the final digitization threshold value has been obtained, and the temporary digitization threshold value and subject presence information obtained for each block by the temporary digitization threshold value calculator and the subject presence evaluator, respectively; a final digitization threshold value determiner for determining the final digitization threshold value for each of the blocks using the block information of the one block and peripheral ones of the blocks stored by the block information store; and a digitizer for digitizing the input image using the final digitization threshold value determined for each block by the final digitization threshold value determiner.

Claim 21

An image processing apparatus comprising: an image separator for separating an input image into plural blocks; a temporary digitization threshold value calculator for calculating a temporary digitization threshold value for each of the blocks separated by the image separator; a subject presence evaluator for determining whether a predetermined subject, comprising text characters, is present in each of the blocks separated by the image separator; a block information store for storing as the block information the digitization threshold values of each block for which the final digitization threshold value has been obtained, and the temporary digitization threshold value and subject presence information obtained for each block by the temporary digitization threshold value calculator and the subject presence evaluator, respectively; a final digitization threshold value determiner for determining the final digitization threshold value for each of the blocks using the block information of the one block and peripheral ones of the blocks stored by the block information store; and a digitizer for digitizing the input image using the final digitization threshold value determined for each block by the final digitization threshold value determiner.

Claim 22

An image processing apparatus according to claim 18 wherein the temporary digitization threshold value calculator is characterized by: a histogram generating means for generating a density histogram of the input image; an among-class dispersion calculator for calculating the among-class dispersion value of each density value in the density histogram obtained by the histogram generating means; and a maximum value detecting means for detecting the maximum among-class dispersion values obtained by the among-class dispersion calculator, and outputting as the temporary digitization threshold value the density value of the among-class dispersion value peak.

Claim 23

An image processing apparatus according to claim 18 wherein the temporary digitization threshold value calculator is characterized by: a histogram generating means for generating a density histogram of the input image; a peak detection means for detecting only the peaks required to obtain the temporary digitization threshold value from the density histogram obtained by the histogram generating means; and a temporary digitization threshold value calculator for obtaining the temporary digitization threshold value based on the peaks detected by said peak detection means.

Claim 24

An image processing apparatus according to claim 20 wherein the peak detection means determines, when there are plural peak candidates in the density histogram obtained by the histogram generating means, whether one of plural peak candidates is the peak required to obtain the temporary digitization threshold value based on the frequencies of occurrence of the density value in each of the plural peak candidates, and the comparative relationship between the frequencies of occurrence of the density value in peak candidates being compared.

Claim 25

An image processing apparatus according to claim 21 wherein the temporary digitization threshold value calculator is characterized by: defining as the temporary digitization threshold value the density value corresponding to a predetermined position at either the rise side or the drop side of the detected peak when there is only one peak detected by the peak detection process, and defining as the temporary digitization threshold value the density value corresponding to a predetermined position between adjacent peaks when plural peaks are detected by the peak detection process.

Claim 26

An image processing apparatus according to claim 18 wherein the subject presence evaluator is characterized by: a feature quantity extraction means for extracting a feature quantity based on the density distribution of the input image in each separated block; and a subject presence discrimination signal generator for determining the subject presence by comparing the feature quantity obtained by the feature quantity extraction means with a predetermined evaluation value supplied as an input parameter of the comparison operation.

Claim 27

An image processing apparatus according to claim 23 wherein the feature quantity of the density distribution is the density dispersion value of the input image in each block.

Claim 28

An image processing apparatus according to claim 23 wherein the feature quantity of the density distribution is the maximum among-class dispersion value obtained based on the density histogram obtained from the input image in each block.

Claim 29

An image processing apparatus according to claim 23 wherein the feature quantity of the density distribution is the maximum density difference of the input image in each block.

Claim 30

An image processing apparatus according to claim 18 wherein the subject presence evaluator is characterized by: a feature quantity extraction means for extracting as the feature quantity at least two of following three feature quantities of the input image in each block: the density dispersion value, the maximum among-class dispersion value, and the maximum density difference; and a subject presence discrimination signal generator for determining the subject presence by comparing the feature quantities obtained by the feature quantity extraction means with predetermined evaluation values supplied as input parameters of the comparison operation.

Claim 31

An image processing apparatus according to claim 18 wherein the final digitization threshold value determiner defines a reference range comprising plural blocks around the target block for which the digitization threshold value is to be determined, and determines the final digitization threshold value of the target block based on the temporary digitization threshold value of the target block and an already-determined final digitization threshold value or temporary digitization threshold value of at least one block in said reference range.

Claim 32

An image processing apparatus according to claim 28 where when subject text or objects are determined to not be present in any blocks in the reference range, excluding the target block, the digitization threshold value of the target block is obtained by determining whether subject text or objects are present in the target block, and is determined to be the temporary digitization threshold value of the target block when subject text or objects are determined to be present in the target block, and is determined by referencing an already-determined digitization threshold value in the reference range when subject text or objects are determined to not be present in the target block.

Claim 33

An image processing apparatus according to claim 28 where when subject text or objects are determined to be present in one or more blocks of the reference range, excluding the target block, the digitization threshold value of the target block is obtained by determining whether subject text or objects are present in the target block, and is determined from a temporary digitization threshold value in the reference range and the temporary digitization threshold value of the target block when subject text or objects are determined to be present in the target block, and is determined from the temporary digitization threshold values in the reference range when subject text or objects are determined to not be present in the target block.

Claim 34

An image processing apparatus according to claim 28 or claim 30 wherein the final digitization threshold value of the target block is obtained by calculation from the temporary digitization threshold value of the target block and the temporary digitization threshold value of at least one block in the reference range.

Claim 35

An image processing apparatus according to claim 28 or claim 30 wherein the final digitization threshold value of the target block is determined by means of a logic operation obtaining the minimum value of the temporary digitization threshold values of the target block and each block in the reference range, and defining this minimum temporary digitization threshold value as the final digitization threshold value of the target block.

Claim 36

An image processing apparatus according to claim 32 where when the minimum temporary digitization threshold value in the reference range is defined as the final digitization threshold value to be obtained, the final digitization threshold value of the target block is defined, depending upon the conditions, to be the sum of a predetermined value added to the minimum temporary digitization threshold value in the reference range.

Claim 37

An image processing apparatus comprising: an image separator for separating an input image into plural blocks: a temporary digitization threshold value calculator for calculating a temporary digitization threshold value for each of the blocks separated by the image separator: a subject presence evaluator for determining whether a predetermined subject, comprising text characters, is present in each of the blocks separated by the image separator: a block information store for storing as the block information the digitization threshold values of each block for which the final digitization threshold value has been obtained, and the temporary digitization threshold value and subject presence information obtained for each block by the temporary digitization threshold value calculator and the subject presence evaluator, respectively; a final digitization threshold value determiner for determining the final digitization threshold value for each block using the block information stored by the block information store; and a digitizer for digitizing the input image using the final digitization threshold value determined for each block by the final digitization threshold value determiner wherein the temporary digitization threshold value calculator is characterized by: a histogram generating means for generating a density histogram of the input image; a peak detection means for detecting only the peaks required to obtain the temporary digitization threshold value from the density histogram obtained by the histogram generating means; and a temporary digitization threshold value calculator for obtaining the temporary digitization threshold value based on the peaks detected by said peak detection means, and wherein the reference range is defined relative to the target block: in the line direction as a variable number of blocks dependent upon the presence of the subject in each block, and in the column direction as at least one block line above and one block line below the block line of the target block.

Claim 38

An image processing apparatus comprising: an image separator for separating an input image into plural blocks: a temporary digitization threshold value calculator for calculating a temporary digitization threshold value for each of the blocks separated by the image separator: a subject presence evaluator for determining whether a predetermined subject, comprising text characters, is present in each of the blocks separated by the image separator: a block information store for storing as the block information the digitization threshold values of each block for which the final digitization threshold value has been obtained, and the temporary digitization threshold value and subject presence information obtained for each block by the temporary digitization threshold value calculator and the subject presence evaluator, respectively; a final digitization threshold value determiner for determining the final digitization threshold value for each block using the block information stored by the block information store; and a digitizer for digitizing the input image using the final digitization threshold value determined for each block by the final digitization threshold value determiner wherein the temporary digitization threshold value calculator is characterized by: a histogram generating means for generating a density histogram of the input image; a peak detection means for detecting only the peaks required to obtain the temporary digitization threshold value from the density histogram obtained by the histogram generating means; and a temporary digitization threshold value calculator for obtaining the temporary digitization threshold value based on the peaks detected by said peak detection means, and wherein the reference range is defined relative to the target block: in the line direction as a variable number of blocks dependent upon the presence of the subject in each block, and in the column direction as at least one block line above and one block line below the block line of the target block.