Method and system for processing images of forms which have irregular construction and/or determining whether characters are interior to a form

A method of analyzing whether character candidates are inside a form, comprising the steps of:

• inputting an image containing a form and character candidates;
• and
• determining if the character candidates are inside or outside the form, including the steps of,
• determining if the character candidates are between a first perimeter of the form in a first viewpoint;
• determining if the character candidates are between a second perimeter of the form, the second perimeter being opposite the first perimeter, in a second viewpoint opposite the first viewpoint;
• and
• determining that the character candidates are inside the form when it is determined that character candidates are not between the first perimeter in the first viewpoint, and are not between the second perimeter in the second viewpoint.

A system for processing an image, comprising:

• means for obtaining an image which includes document image data;
• means for determining a form candidate using the document image data;
• means for extracting provisional ruled lines from the form candidate;
• means for determining a structure of the form candidate by analyzing the provisional lines in a horizontal and vertical direction, including,
• determining whether or not the form candidate corresponds to a figure,
• determining whether or not the form candidate corresponds to a graph,
• determining whether or not the form candidate corresponds to a frame, and
• determining whether or not the form candidate corresponds to a form;
• and
• means for determining if character candidates which are proximate to the form candidate which is determined to correspond to a form are inside or outside the corresponding form,
• wherein the means for determining if the character candidates which are proximate to the form are inside or outside the corresponding form comprises:
• means for determining if the character candidates are between a first perimeter of the corresponding form in a first viewpoint;
• means for determining if the character candidates are between a second perimeter of the corresponding form, the second perimeter being opposite the first perimeter, in a second viewpoint opposite the first viewpoint;
• and
• means for determining that the character candidates are inside the corresponding form when it is determined that character candidates are not between the first perimeter in the first viewpoint, and are not between the second perimeter in the second viewpoint.

A system of analyzing whether character candidates are inside a form, comprising:

• means for inputting an image containing a form and character candidates;
• and
• means for determining if the character candidates are inside or outside the form, including,
• determining if the character candidates are between a first perimeter of the form in a first viewpoint;
• means for determining if the character candidates are between a second perimeter of the form, the second perimeter being opposite the first perimeter, in a second viewpoint opposite the first viewpoint;
• and
• means for determining that the character candidates are inside the form when it is determined that character candidates are not between the first perimeter in the first viewpoint, and are not between the second perimeter in the second viewpoint.