| 摘要 |
PROBLEM TO BE SOLVED: To obtain an image reader capable of eliminating the need for input of document size and reading out a document, even when the document is slightly inaccurately set within a readable range by detecting four corners of the document from boundary information and segmenting the document, based on the four corners of the document. SOLUTION: A coordinate storing means 30 stores the positional information of the ends of a document. A rotation display means 40 displays the number of detections in each of right, left, upper, and lower ends of the document. A document position defining means 50 generates the four-corner positional information of the document, based on the positional information of the document ends and generates an line inclination. A line inclination correction means 60 corrects the line inclination of the document, stored in a read data storing means 10 based on the line inclination and corrects the four-corner positional information of the document. A document segmentation means 70 extracts the line inclination corrected document stored in the means 10 and stores the extracted document in an image storing means 80. The means 80 stores the image data of the document and outputs the stored data.
|
