Method and apparatus for performing optical character recognition (OCR) and text stitching
View Patent ↗A method of generating an electronic text file from a paper-based document that includes a plurality of characters includes capturing a plurality of partially overlapping digital images of the document. Optical character recognition is performed on each one of the plurality of captured digital images, thereby generating a corresponding plurality of electronic text files. Each one of the electronic text files includes a portion of the plurality of characters in the document. The plurality of electronic text files are compared with one another to identify characters that are in common between the electronic text files. The plurality of electronic text files are combined into a combined text file based on the comparison. The combined text file includes the plurality of characters in the document.
1. A method of processing a document that includes a plurality of characters, the method comprising:
capturing a plurality of partially overlapping digital images of the document with an image capture device;
maintaining image sequence information identifying a sequential order in which the partially overlapping images were captured;
receiving direction information indicative of a direction of relative movement between the image capture device and the document during the capture of the plurality of partially overlapping digital images;
performing optical character recognition on the plurality of captured partially overlapping digital images to generate a corresponding plurality of electronic text files;
selecting two of the plurality of electronic text files based on the sequence information; and
combining the two selected electronic text files into a combined text file based on the direction information and based on a character sequence appearing in common in the two selected electronic text files.
2. The method of claim 1 , wherein the plurality of partially overlapping digital images are captured with a digital camera, the method further comprising:
providing the direction information with a user input device of the digital camera.
3. The method of claim 1 , wherein the plurality of partially overlapping digital images are captured with a digital camera, the method further comprising:
automatically detecting the direction information.
4. The method of claim 1 , wherein the plurality of partially overlapping digital images are captured automatically at a predefined time interval.
5. A digital camera comprising:
a memory adapted to store image sequence information representing an order in which a plurality of digital images are captured; and
a controller configured to: receive direction information indicative of a direction of movement of the digital camera during capture of the plurality of digital images, perform optical character recognition on the plurality of digital images to generate electronic text files, and stitch at least two of the electronic text files together based on the image sequence information, the direction information, and by identifying overlapping text appearing in the at least two electronic text files.
6. The digital camera of claim 5 , further comprising:
a user input device for inputting the direction information.
7. The digital camera of claim 5 , further comprising:
a motion detector for automatically detecting the direction information.
8. The digital camera of claim 5 , wherein the controller is configured to capture the plurality of digital images automatically at a predefined time interval.
9. An image capturing apparatus comprising:
a memory for storing image sequence information representing an order in which images are captured by the image capturing apparatus; and
a processor configured to perform optical character recognition on the images, generate electronic text files corresponding to the images, and to stitch the electronic text files together based at least in part on the image sequence information, on direction information indicative of a direction of relative movement of the image capturing apparatus, and on character sequence information.
10. The image capturing apparatus of claim 9 , wherein the image capturing apparatus is implemented in one of a cellular telephone, a personal digital assistant device, and a laptop computer.
11. The image capturing apparatus of claim 9 , further comprising an input device to allow a user to enter the direction information.
12. The image capturing apparatus of claim 9 , further comprising
a motion detector to automatically detect the direction information.
13. The image capturing apparatus of claim 9 , wherein the processor is configured to cause the images to be captured automatically at a predefined time interval.