Image processing device and method for controlling image processing device

An image processing device comprising:

• a layout analyzing portion which analyzes a layout of image data and which extracts from the image data a character region that includes characters;
• an OCR processing portion which recognizes the characters included in the character region and which generates text data;
• a document file generating portion which
• calculates a proportion by dividing an area of the character region by a total area of the character region and a region other than the character region,
• generates a document file by a first generation process when the proportion is lower than a threshold value, and
• generates the document file by a second generation process when the proportion is equal to or higher than the threshold value, wherein
• the first generation process is a process of generating the document file having a text box including the text data attached thereto, and
• the second generation process is a process of generating the document file by writing the text data directly in the document file without using the text box;
• and
• a size recognizing portion which recognizes a size of the characters included in the character region, wherein
• as a number of different sizes increases, the document file generating portion sets the threshold value higher.

A method for controlling an image processing device, the method comprising:

• analyzing a layout of image data and extracting from the image data a character region that includes characters;
• recognizing the characters included in the character region and generating text data;
• calculating a proportion by dividing an area of the character region by a total area of the character region and a region other than the character region;
• generating a document file by a first generation process when the proportion is lower than a threshold value;
• generating the document file by a second generation process when the proportion is equal to or higher than the threshold value, wherein
• the first generation process is a process of generating the document file having a text box including the text data attached thereto, and
• the second generation process is a process of generating the document file by writing the text data directly in the document file without using the text box;
• recognizing a size of the characters included in the character region;
• and
• setting the threshold value higher as a number of different sizes increases.