Automatic image capture
An improved automatic image capture system for an intelligent mobile device having a camera guides a user to position the camera so only a single image needs to be automatically captured. Syntactic features, using a view finder on a display of the intelligent mobile device, are used to guide a user to maximize the occupancy of the view finder with the document so that the document is maximized within the view finder based upon detected corners of the document. When occupancy is maximized, the camera automatically captures the image of the document for post-processing using semantic knowledge of the document. A confidence level is computed based on the semantic knowledge to qualify an image with greater accuracy, and without user intervention, prior to transmission to a remote site.
1. A method of automatic image capture of a target document using an intelligent mobile device having a camera comprising the steps of:
showing a view finder on a display of the intelligent mobile device and a secondary view finder of corresponding shape to the view finder;
lining up the two view finders until the secondary view finder superimposed on the view finder;
guiding a user to maximize occupancy of the view finder with an image of the target document as a first condition;
automatically capturing the target document image when the occupancy is maximized to produce a captured image; and
semantically processing the captured image to qualify the captured image for transmission to a remote site.
2. The method as recited in claim 1 further comprising the step of:
testing the target document for sufficient brightness and contrast with respect to a background for the target document prior to the capturing step.
3. The method as recited in claim 2 further comprising the step of:
determining the corners of the target document image within the view finder prior to the capturing step.
4. The method as recited in claim 3 further comprising the step of:
maximizing the occupancy of the target document image within the view finder after the corner determining step and prior to the capturing step.
5. The method as recited in claim 4 further comprising the step of:
assuring focus of the camera with respect to the target document prior to the capturing step.
6. The method as recited in claim 5 further comprising the step of checking stability of the camera with respect to the target document prior to the capturing step.
7. A method of automatic image capture of a target document using an intelligent mobile device having a camera comprising the steps of:
showing a view finder on a display of the intelligent mobile device and a secondary view finder of corresponding shape to the view finder;
lining up the two view finders until the secondary view finder superimposed on the view finder;
guiding a user to maximize occupancy of the view finder with an image of the target document as a first condition;
automatically capturing the target document image when the occupancy is maximized to produce a captured image; and
semantically processing the captured image to qualify the captured image for transmission to a remote site, wherein the semantically processing step comprising the steps of:
providing semantic knowledge of the target document within the mobile device;
by using symbol character recognition, measuring a confidence level of each symbol in the target document image based upon the semantic knowledge of the target document; and
qualifying the target document image for transmission to a remote site when an average confidence level for all the symbols achieves a specified score.
8. A method of automatic image capture of a target document using an intelligent mobile device having a camera comprising the steps of:
showing a trapezoidal view finder on a display of the intelligent mobile device and
a secondary view finder of corresponding shape to the trapezoidal view finder;
lining up the trapezoidal view finder and the secondary view finder until the secondary view finder superimposed on the trapezoidal view finder;
determining the corners of the target document image within the trapezoidal view finder using edge detection and line projection;
maximizing the target document image within the trapezoidal view finder using the determined corners;
automatically capturing the target document image for post-processing; and
semantically processing the captured image to qualify the captured image for transmission to a remote site, wherein the semantically processing step comprises the steps of:
providing semantic knowledge of the target document within the mobile device;
by using optical symbol recognition, measuring a confidence level of each symbol in the target document image based upon the semantic knowledge of the target document; and
qualifying the target document image for transmission to a remote site when an average confidence level for all the symbols achieves a specified score.
9. The method as recited in claim 8 further comprising the step of:
assuring focus of the camera with respect to the target document prior to the capturing step.
10. The method as recited in claim 9 further comprising the step of:
semantically processing the captured image to qualify the captured image for transmission to a remote site.
11. The method as recited in claim 10 wherein the semantically processing step comprises the steps of:
providing semantic knowledge of the target document within the mobile device;
by using optical character recognition, measuring a confidence level of each character in the target document image based upon the semantic knowledge of the target document; and
qualifying the target document image for transmission to a remote site when an average confidence level for all the characters achieves a specified score.
12. The method as recited in claim 8 further comprising the steps of:
modifying capture parameters when the average confidence level is less than the specified score; and
repeating the measuring and modifying steps until the specified score is achieved.
13. The method as recited in claim 8 further comprising the step of:
comparing specific symbols identified by the optical symbol recognition to expected results based upon the semantic knowledge of the test document in order to improve the average confidence level prior to transmission to the remote site.
14. The method as recited in claim 13 further comprising the steps of:
modifying capture parameters when the average confidence level is less than the specified score; and
repeating the measuring, comparing and modifying steps until the specified score is achieved.
15. The method as recited in claim 7 wherein the view finder has a trapezoidal shape.
16. The method as recited in claim 7 further comprising the step of:
testing the target document for sufficient brightness and contrast with respect to a background for the target document prior to the capturing step.
17. The method as recited in claim 16 , further comprising the step of:
determining the corners of the target document image within the view finder prior to the capturing step.
18. The method as recited in claim 16 , further comprising the step of:
maximizing the occupancy of the target document image within the view finder after the corner determining step and prior to the capturing step.
19. The method as recited in claim 18 further comprising the step of:
assuring focus of the camera with respect to the target document prior to the capturing step.
20. The method as recited in claim 19 further comprising
the step of checking stability of the camera with respect to the target document prior to the capturing step.