Image processing apparatus for converting image in characteristic region of original image into image of brushstroke patterns

An image processing method for an image processing apparatus including a CPU and a memory storing an original image, the image processing method comprising executing, with the CPU:

• detecting a characteristic region of the original image stored in the memory;
• calculating an importance region by calculating importance information from the original image stored in the memory;
• and
• calculating an importance barycenter of the calculated importance region, based on importance of each pixel of the calculated importance region;
• adding a margin region to the characteristic region based on the calculated importance barycenter;
• and
• generating and outputting a final output image in which the margin region has been added to the characteristic region of the original image;
• wherein calculating the importance region comprises calculating a region with an importance evaluation value being 90 or higher as an importance region, in a case in which an upper limit of values of the calculated importance is 100.

An image processing method for an image processing apparatus including a CPU and a memory storing an original image, the image processing method comprising executing, with the CPU:

• detecting a characteristic region of the original image stored in the memory;
• calculating an importance region by calculating importance information from the original image stored in the memory;
• calculating an importance barycenter based on an importance of each pixel of the calculated importance region;
• adding a margin region to the characteristic region based on the calculated importance barycenter;
• and
• generating and outputting a final output image in which the margin region has been added to the characteristic region of the original image;
• wherein calculating the importance region comprises calculating the importance for each pixel composing the original image, and calculating the importance barycenter of the original image, based on the calculated importance of each pixel;
• wherein the importance barycenter is calculated by applying equations (1) and (2) to pixels having importance of at least a threshold value, where the importance barycenter in spatial distribution of the importance of each pixel is (x, y);
• wherein the threshold value is 90, in a case in which an upper limit of values of the importance is 100, and
• wherein equations (1) and (2) are as follows:
• mx=1/(n+sumw)*Σ(x*w(x,y)) (1)
• my=1/(n+sumw)*Σ(y*w(x,y)) (2)
• where n represents a number of pixels having importance of at least the threshold value, sumw represents a sum of importance, w (x, y) represents a value of importance in (x, y), mx represents an x coordinate of the importance barycenter, and my represents a y coordinate of the importance barycenter.

An image processing apparatus comprising:

• a CPU;
• and
• a memory storing an original image,
• wherein the CPU is configured to:
• detect a characteristic region of the original image stored in the memory;
• calculate an importance region by calculating importance information from the original image stored in the memory;
• calculate an importance barycenter of the calculated importance region based on an importance of each pixel of the calculated importance region;
• add a margin region to the characteristic region based on the calculated importance barycenter;
• and
• generate and output a final output image in which the margin region has been added to the characteristic region of the original image;
• wherein, to calculate the importance information, the CPU is configured to:
• detect the importance of each pixel composing the original image based on an indicator of the importance, by using a color, brightness, or an edge direction as the indicator of importance, from the original image;
• and
• generate an importance map indicating spatial distribution of the importance detected for each pixel, to thereby calculate the importance information of the original image.