Image processing apparatus having image area recognition function and image processing method

Claim 2

An image processing apparatus comprising: means for fetching an image; means for forming a density histogram based on the fetched image; means for deciding a sort of the fetched image corresponding to a histogram distribution of a background area when the density histogram is divided into the background area and a character area having an image density higher than the background area, a first image density frequency having a maximum frequency within a range of the background area, and each frequency of the density histogram in left and right sides of a crest having the first image density frequency; and means for correcting a gradation of the fetched image, in accordance with a decision result of the deciding means.

Claim 3

An image processing apparatus according to claim 1, wherein the deciding means includes deciding means for deciding the fetched image as a character image when at least one crest of a histogram having a frequency of a predetermined percentage or less of the first image density frequency of the background area exists in predetermined range of each of left and right sides of a crest having the first image density frequency, and for deciding the fetched image as a photograph image when not.

Claim 4

An image processing apparatus according to claim 1, wherein the deciding means includes: means for obtaining the first image density frequency having a maximum frequency within a range of the background area, and the second image density frequency having a maximum frequency within a range of the character area; and means for deciding the fetched image as a character area when a total of a subtotal of frequencies of a predetermined number of image densities in left and right sides of a first crest having the first image density frequency and a subtotal of frequencies of a predetermined number of image densities in left and right sides of a crest having the second image density frequency is greater than a predetermined percentage of a frequency total of frequencies of the entire density histogram; and means for deciding the fetched image as a photograph area when the total is equal to or smaller than the predetermined percentage of the frequency total of the entire density histogram.

Claim 5

An image processing apparatus according to claim 1, wherein the deciding means includes: means for calculating an accumulative frequency by summing frequencies of a histogram crest having a density lower by a predetermined second density value than a first image density value of the first image density frequency to a histogram crest having a minimum density value when the first image density value of the first image density frequency is greater than a predetermined first density value; and means for deciding the fetched image as a photograph image when the accumulative frequency is greater than a predetermined frequency, and for deciding the fetched image as a character image when the accumulative frequency is equal to or smaller than the predetermine frequency.

Claim 6

An image processing apparatus according to claim 1, wherein the deciding means includes: first deciding means for deciding the fetched image as a character image when at least one crest of a histogram having a frequency of a predetermined percentage or less of a first crest having the first image density frequency of the background area exists in each of left and right sides of the first image density frequency, and for deciding the fetched image as a photograph image when not; means for obtaining the first image density frequency having a maximum frequency within a range of the background area, and a second image density frequency having a maximum frequency within a range of the character area; second deciding means for deciding the fetched image as a character area when a total of a subtotal of frequencies of a predetermined number of image densities in left and right sides of the first crest and a subtotal of frequencies of a predetermined number of image densities in left and right sides of a second crest having the second image density frequency is greater than a predetermined second percentage of a frequency total of frequencies of the entire density histogram, and for deciding the fetched image as a photograph area when the total is equal to or smaller than the predetermined second percentage of the frequency total of the entire density histogram; means for calculating an accumulative frequency by summing frequencies of a histogram crest having a density lower by a predetermined second density value than a first density value having the first image density frequency to a histogram crest having a minimum density value when the first image density value is greater than a predetermined first density value; and third deciding means for deciding the fetched image as a photograph image when the accumulative frequency is greater than a predetermined frequency, and for deciding the fetched image as a character image when the accumulative frequency is equal to or smaller than the predetermine frequency; fourth deciding means for deciding the fetched image as a photograph image when any one of the first, second, and third deciding means decides the fetched image as a photograph image; and means for correcting a gradation of the fetched image accordance with a decision result of the fourth deciding means.

Claim 7

An image processing apparatus according to claim 1, wherein the correcting means includes means for correcting a gradation of the fetched image, in correspondence with a second image sort different from a first image sort, after recognizing that a predetermined number of decisions of the second image sort are made continuously after the deciding means makes a decision of the first image sort.

Claim 8

An image processing method comprising: a step of fetching an image; a step of forming a density histogram based on the fetched image; a step of deciding a sort of the fetched image corresponding to a histogram distribution of a background area when the density histogram is divided into the background area and a character area having an image density higher than the background area, a first image density frequency having a maximum frequency within a range of the background area, and each frequency of the density histogram in left and right sides of a crest having the first image density frequency; and a step of correcting a gradation of the fetched image, in accordance with a decision result of the deciding step.

Claim 9

An image processing method comprising: a step of fetching an image; a step of forming a density histogram based on the fetched image; a step of deciding a sort of the fetched image corresponding to a histogram distribution of a background area when the density histogram is divided into the background area and a character area having an image density higher than the background area, a first image density frequency having a maximum frequency within a range of the background area, and each frequency of the density histogram in left and right sides of a crest having the first image density frequency; and a step of correcting a gradation of the fetched image, in accordance with a decision result of the deciding step.

Claim 10

An image processing method according to claim 7, wherein the deciding step includes a deciding step of deciding the fetched image as a character image when at least one crest of a histogram having a frequency of a predetermined percentage or less of the first image density frequency of the background area exists in predetermined range of each of left and right sides of a crest having the first image density frequency, and of deciding the fetched image as a photograph image when not.

Claim 11

An image processing method according to claim 7, wherein the deciding step includes: a step of obtaining the first image density frequency having a maximum frequency within a range of the background area, and the second image density frequency having a maximum frequency within a range of the character area; and a step of deciding the fetched image as a character area when a total of a subtotal of frequencies of a predetermined number of image densities in left and right sides of a first crest having the first image density frequency and a subtotal of frequencies of a predetermined number of image densities in left and right sides of a second crest having the second image density frequency is greater than a predetermined percentage of a frequency total of frequencies of the entire density histogram; and a step of deciding the fetched image as a photograph area when the total is equal to or smaller than the predetermined percentage of the frequency total of the entire density histogram.

Claim 12

An image processing method according to claim 7, wherein the deciding step includes: a step of calculating an accumulative frequency by summing frequencies of a histogram crest having a density lower by a predetermined second density value than a first image density value having the first image density frequency to a histogram crest having a minimum density value when the first image density value is greater than a predetermined first density value; and a step of deciding the fetched image as a photograph image when the accumulative frequency is greater than a predetermined frequency, and for deciding the fetched image as a character image when the accumulative frequency is equal to or smaller than the predetermine frequency.

Claim 13

An image processing method according to claim 7, wherein the deciding step includes: a first deciding step of deciding the fetched image as a character image when at least one crest of a histogram having a frequency of a predetermined percentage or less of the first image density frequency of the background area exists in each of left and right sides of a first crest having the first image density frequency, and of deciding the fetched image as a photograph image when not; a step of obtaining the first image density frequency having a maximum frequency within a range of the background area, and a second image density frequency having a maximum frequency within a range of the character area; a second deciding step of deciding the fetched image as a character area when a total of a subtotal of frequencies of a predetermined number of image densities in left and right sides of the first crest and a subtotal of frequencies of a predetermined number of image densities in left and right sides of a second crest having the second image density frequency is greater than a predetermined second percentage of a frequency total of frequencies of the entire density histogram, and of deciding the fetched image as a photograph area when the total is equal to or smaller than the predetermined second percentage of the frequency total of the entire density histogram; a step of calculating an accumulative frequency by summing frequencies of a histogram crest having a density lower by a predetermined second density value than a density value of a first image density value having the first image density frequency to a histogram crest having a minimum density value when the first image density frequency is greater than a predetermined first density value; and a third deciding step of deciding the fetched image as a photograph image when the accumulative frequency is greater than a predetermined frequency, and for deciding the fetched image as a character image when the accumulative frequency is equal to or smaller than the predetermine frequency; a fourth deciding step for deciding the fetched image as a photograph image when any of the first, second, and third deciding steps decides the fetched image as a photograph image; and a step of correcting a gradation of the fetched image accordance with a decision result of the fourth deciding means.

Claim 14

An image processing method according to claim 7, wherein the correcting step includes a step for correcting a gradation of the fetched image, in correspondence with a second image sort different from a first image sort, after recognizing that a predetermined number of decisions of the second image sort are made continuously after the deciding step makes a decision of the first image sort.

Claim 15

An image processing apparatus comprising: means for fetching an image; means for forming a density histogram based on the fetched image; means for detecting a background area and a character area corresponding to the density histogram formed by the forming means; means for deciding a sort of the fetched image corresponding to a first image density frequency having a maximum frequency within a range of the background area and a histogram distribution of the background area; and means for correcting a gradation of the fetched image, in accordance with a decision result of the deciding means.

Claim 16

An image processing apparatus comprising: means for fetching an image; means for forming a density histogram based on the fetched image; means for detecting a background area and a character area corresponding to the density histogram formed by the forming means; means for deciding a sort of the fetched image corresponding to a first image density frequency having a maximum frequency within a range of the background area and a histogram distribution of the background area; and means for correcting a gradation of the fetched image, in accordance with a decision result of the deciding means.

Claim 17

An image processing apparatus according to claim 13, wherein the correcting means includes means for correcting a gradation of the fetched image within a predetermined range of change value, in accordance with a decision result of the deciding means.

Claim 18

An image processing apparatus according to claim 13, wherein the deciding means includes means for deciding a sort of the fetched image on the basis of a density having a maximum frequency within the background area and density histograms of right and left areas of background.

Claim 19

An image processing apparatus according to claim 13, wherein the deciding means includes means for deciding a sort of the fetched image on the basis of a density histogram corresponding to densities that are lower than a density having a maximum frequency within the background area.

Claim 20

An image processing method comprising: fetching an image; forming a density histogram based on the fetched image; detecting a background area and a character area corresponding to the density histogram formed by the forming; deciding a sort of the fetched image corresponding to a first image density frequency having a maximum frequency within a range of the background area and a histogram distribution of the background area; and correcting a gradation of the fetched image, in accordance with a decision result of the deciding.

Claim 21

An image processing method comprising: fetching an image; forming a density histogram based on the fetched image; detecting a background area and a character area corresponding to the density histogram formed by the forming; deciding a sort of the fetched image corresponding to a first image density frequency having a maximum frequency within a range of the background area and a histogram distribution of the background area; and correcting a gradation of the fetched image, in accordance with a decision result of the deciding.

Claim 22

An image processing method according to claim 17, wherein the correcting step includes a step of correcting a gradation of the fetched image within a predetermined range of change value, in accordance with a decision result of the deciding step.

Claim 23

An image processing method according to claim 17, wherein the deciding includes deciding a sort of the fetched image on the basis of a density having a maximum frequency within the background area and density histograms of right and left areas of background.

Claim 24

An image processing method according to claim 17, wherein the deciding includes deciding a sort of the fetched image on the basis of a density histogram corresponding to densities that are lower than a density having a maximum frequency within the background area.