Information processing device, image modification method, and computer program product
View Patent ↗According to an embodiment, an information processing device includes: a first division unit divides an image into a text containing region and a background region other than the text containing region; a second division unit divides a text containing region into a character region constituted by lines forming characters and a character background region other than the character region; a calculator calculates a first representative value of an attribute of the character region, a second representative value of the attribute of the character background region, and a third representative value of the attribute of the background region; a modification unit makes modification so that a first difference based on the first and third representative values, a second difference based on the first and second representative values, and a third difference based on the second and third representative values become larger; and an output unit outputs a modified image.
1. An information processing device comprising:
an acquisition unit to acquire an image;
a first division unit to divide the image into a text containing region containing one or more characters and a background region other than the text containing region;
a second division unit to divide the text containing region into a character region constituted by lines forming the characters and a character background region other than the character region;
a calculator to calculate a first representative value of a predetermined attribute of the character region, a second representative value of the predetermined attribute of the character background region, and a third representative value of the predetermined attribute of the background region;
a modification unit to modify at least one of the first representative value, the second representative value, and the third representative value in such a way that a first difference based on the first representative value and the third representative value, a second difference based on the first representative value and the second representative value, and a third difference based on the second representative value and the third representative value become larger; and
an output unit to output a modified image obtained by
changing the value of the predetermined attribute of the character region to a modified first representative value when the first representative value is modified,
changing the value of the predetermined attribute of the character background region to a modified second representative value when the second representative value is modified, or
changing the value of the predetermined attribute of the background region to a modified third representative value when the third representative value is modified.
2. The device according to claim 1 , wherein the predetermined attribute is an attribute capable of expressing easiness of recognition by a human.
3. The device according to claim 2 , wherein
the predetermined attribute is hue,
the first representative value is a representative color of a hue in the character region,
the second representative value is a representative color of a hue in the character background region,
the third representative value is a representative color of a hue in the background region,
the first difference is a visual recognition distance when the first representative value is a color of an object and the third representative value is a color of a background of the object,
the second difference is a visual recognition distance when the first representative value is a color of an object and the second representative value is a color of a background of the object, and
the third difference is a visual recognition distance when the second representative value is a color of an object and the third representative value is a color of a background of the object.
4. The device according to claim 3 , wherein the modification unit further modifies the first representative value and the second representative value in such a way that a sum of the second difference and the third difference is maximized.
5. The device according to claim 2 , wherein
the predetermined attribute is brightness, and
the first representative value is a representative value of a brightness in the character region,
the second representative value is a representative value of a brightness in the character background region,
the third representative value is a representative value of a brightness in the background region, and
the first difference, the second difference, and the third difference are differences in brightness.
6. The device according to claim 2 , wherein
the predetermined attribute is saturation, and
the first representative value is a representative value of a saturation in the character region,
the second representative value is a representative value of a saturation in the character background region,
the third representative value is a representative value of a saturation in the background region, and
the first difference, the second difference, and the third difference are differences in saturation.
7. The device according to claim 2 , wherein
the predetermined attribute is a color, and
the first representative value is a representative value of a color in the character region,
the second representative value is a representative value of a color in the character background region,
the third representative value is a representative value of a color in the background region, and
the first difference, the second difference, and the third difference are differences in color.
8. The device according to claim 1 , wherein
the first division unit divides the image into multiple text containing regions and the background region, and
the second division unit divides each of the text containing regions into the character region and the character background region,
the information processing device further comprises a determination unit to determine significance of each of the character regions and determine a character region having a significance equal to or lower than a threshold and a character background region of the character region to be a background region.
9. The device according to claim 1 , wherein the output unit outputs the modified image to a display unit for display.
10. The device according to claim 1 , wherein the output unit outputs the modified image to an external device.
11. An image modification method comprising:
acquiring an image;
dividing the image into a text containing region containing one or more characters and a background region other than the text containing region;
dividing the text containing region into a character region constituted by lines forming the characters and a character background region other than the character region;
calculating a first representative value of a predetermined attribute of the character region, a second representative value of the predetermined attribute of the character background region, and a third representative value of the predetermined attribute of the background region;
modifying at least one of the first representative value, the second representative value, and the third representative value in such a way that a first difference based on the first representative value and the third representative value, a second difference based on the first representative value and the second representative value, and a third difference based on the second representative value and the third representative value become larger; and
outputting a modified image obtained by
changing the value of the predetermined attribute of the character region to a modified first representative value when the first representative value is modified,
changing the value of the predetermined attribute of the character background region to a modified second representative value when the second representative value is modified, or
changing the value of the predetermined attribute of the background region to a modified third representative value when the third representative value is modified.
12. A computer program product comprising a non-transitory computer-readable medium containing programmed instructions, the instructions causing a computer to execute:
acquiring an image;
dividing the image into a text containing region containing one or more characters and a background region other than the text containing region;
dividing the text containing region into a character region constituted by lines forming the characters and a character background region other than the character region;
calculating a first representative value of a predetermined attribute of the character region, a second representative value of the predetermined attribute of the character background region, and a third representative value of the predetermined attribute of the background region;
modifying at least one of the first representative value, the second representative value, and the third representative value in such a way that a first difference based on the first representative value and the third representative value, a second difference based on the first representative value and the second representative value, and a third difference based on the second representative value and the third representative value become larger; and
outputting a modified image obtained by
changing the value of the predetermined attribute of the character region to a modified first representative value when the first representative value is modified,
changing the value of the predetermined attribute of the character background region to a modified second representative value when the second representative value is modified, or
changing the value of the predetermined attribute of the background region to a modified third representative value when the third representative value is modified.