Image processing method, image processing apparatus, and image forming apparatus

An image processing method for identifying, from a binary image, a plurality of pixel regions where pixels determined to be having the same pixel values are adjacent to each other, extracting, based on a pixel coordinate value of the identified pixel region, a feature point of the pixel region, calculating features of an image based on the extracted feature point, and determining similarity among images based on the calculated features, the method comprising steps of:

• counting the number of the feature points within a predetermined range when extracting the feature point;
• extracting, from the counted feature points, a feature point included in a region to be processed;
• determining whether the number of the extracted feature points is lower than a third threshold value;
• calculating the features based on a neighboring feature point located within a surrounding region of the extracted feature point when the number of the extracted feature points has been determined to be equal to or higher than the third threshold value;
• and
• changing a range of the surrounding region when the number of the extracted feature points has been determined to be lower than the third threshold value.

An image processing apparatus for identifying, from a binary image, a plurality of pixel regions where pixels determined to be having the same pixel values are adjacent to each other, extracting, based on a pixel coordinate value of the identified pixel region, a feature point of the pixel region, calculating features of an image based on the extracted feature point, and determining similarity among images based on the calculated features, the apparatus comprising:

• a deciding section for deciding a neighboring feature point located within a surrounding region of the extracted feature point;
• a feature point counting section for counting the number of the decided neighboring feature points;
• and
• a determination section for determining whether the counted number of the feature points is equal to or lower than a second threshold value,
• wherein the features is calculated based on the neighboring feature point when the counted number of the feature points has been determined to be above the second threshold value, and a range of the surrounding region is changed when the counted number of the feature points has been determined to be equal to or lower than the second threshold value.

An image processing apparatus for identifying, from a binary image, a plurality of pixel regions where pixels determined to be having the same pixel values are adjacent to each other, extracting, based on a pixel coordinate value of the identified pixel region, a feature point of the pixel region, calculating features of an image based on the extracted feature point, and determining similarity among images based on the calculated features, the apparatus comprising:

• a feature point counting section for counting the number of the feature points within a predetermined range when extracting the feature point;
• an extraction section for extracting, from the counted feature points, a feature point included in a region to be processed;
• and
• a determination section for determining whether the number of the extracted feature points is lower than a third threshold value,
• wherein the features is calculated based on a neighboring feature point located within a surrounding region of the extracted feature points when the number of the extracted feature points has been determined to be equal to or higher than the third threshold value, and a range of the surrounding region is changed when the number of the extracted points has been determined to be lower than the third threshold value.

A computer-readable memory product that stores a computer program for causing a computer to identify, from a binary image, a plurality of pixel regions where pixels determined to be having the same pixel values are adjacent to each other, to extract, based on a pixel coordinate value of the identified pixel region, a feature point of the pixel region, to calculate features of an image based on the extracted feature point, and to determine similarity among images based on the calculated features,

• wherein the computer program comprises steps of;
• causing the computer to decide a neighboring feature point located within a surrounding region of the extracted feature point;
• causing the computer to count the number of the decided neighboring feature points;
• causing the computer to determine whether the counted number of the feature points is equal to or lower than a second threshold value;
• causing the computer to calculate the features based on the neighboring feature point when the counted number of the feature points has been determined to be above the second threshold value;
• and
• causing the computer to change a range of the surrounding region when the counted number of the feature points has been determined to be equal to or lower than the second threshold value.

A computer-readable memory product that stores a computer program for causing a computer to identify, from a binary image, a plurality of pixel regions where pixels determined to be having the same pixel values are adjacent to each other, to extract, based on a pixel coordinate value of the identified pixel region, a feature point of the pixel region, to calculate features of an image based on the extracted feature point, and to determine similarity among images based on the calculated features,

• wherein the computer program comprises steps of;
• causing the computer to count the number of the feature points within a predetermined range when extracting the feature point;
• causing the computer to extract, from the counted feature points, a feature point included in a region to be processed;
• causing the computer to determine whether the number of the extracted feature points is lower than a third threshold value;
• causing the computer to calculate the features based on a neighboring feature point located within a surrounding region of the extracted feature points when the number of the extracted feature points has been determined to be equal to or higher than the third threshold value;
• and
• causing the computer to change a range of the surrounding region when the number of the extracted feature points has been determined to be lower than the third threshold value.