System and method for efficient determination of recognition initial conditions

• US 6,961,465 B2
• Filed: 12/28/2000
• Issued: 11/01/2005
• Est. Priority Date: 12/28/2000
• Status: Expired due to Fees

• Current Assignee: Canon Kabushiki Kaisha (Canon Inc.)
• Original Assignee: Canon Kabushiki Kaisha (Canon Inc.)
• Inventors: Shah, Kishan B
• Primary Examiner(s): COUSO, YON JUNG
• Application Number:
  US09/750,602
• Publication Number:
  US 20020085756A1
• Time in Patent Office: 1,769 Days
• Field of Search: 382/181, 382/190, 382/224, 382/228, 382/229, 382/226, 382/227, 382/272, 382/290, 382/301, 382/309, 382/310, 382/317, 382/270, 382/299, 382/191
• US Class Current: 382/181
• CPC Class Codes: G06F 18/00 Pattern recognition
• CPC Class Codes: G06V 30/148 Segmentation of character regions

Abstract

A system (200) and method creates (312) a reduced-resolution version of an image and uses the reduced resolution version to identify (324) recognition initial conditions. The recognition initial conditions are then used (326) on the original version of the image to recognize objects such as characters in the image.

Claims

1. A method of recognizing at least one object in a digitized representation of an image, comprising the steps of:

receiving the digitized representation of the image, the representation having a first resolution;

creating a reduced-resolution version of the image responsive to the digitized representation of the image, the reduced-resolution version of the image having a second resolution lower than the first resolution;

providing a plurality of sets of initial conditions, the initial conditions including at least a condition for character recognition-processing of the image;

for each of the sets of initial conditions, identifying each confidence level of character recognition by first character recognition-processing of the reduced resolution version of the image having the second resolution based on each of the sets of initial conditions, each of the sets of initial conditions including a threshold value to binarize an image;

selecting an optimal set of initial conditions from the plurality of sets of initial conditions based on each confidence level identified in said identifying step, the optimal set of initial conditions including a threshold value to binarize the image; and

second character recognition processing of the objects represented in the digitized representation of the image having the first resolution by using the optimal set of initial conditions selected in said selecting step.

2. The method according to claim 1, wherein said selecting step selects the optimal set from the plurality of sets of initial conditions based on a highest confidence level identified in said identifying step.

3. The method according to claim 1, wherein said selecting step selects the optimal set from the plurality of sets of initial conditions based on a confidence level exceeding a threshold.

4. The method according to claim 1, wherein said creating step creates the reduced resolution version of the image by calculating an average of a plurality of pixels of the digitized representation of the image having the first resolution.

5. The method according to claim 1, additionally comprising the step of recognizing at least one additional object represcnted in the digitized representation of the image, responsive to the value of at least one initial condition identified responsive to a confidence level exceeding a threshold.

6. The method according to claim 1, additionally comprising the steps of:

attempting to recognize at least one additional object represented in the digitized representation of the image, responsive to the value of at least one initial condition identified, the attempting step comprising the step of producing a confidence level of the attempt; and

responsive to the confidence level of the attempt below a threshold;

repeating the identifying step; and

recognizing the at least one object represented in the digitized representation of the image responsive to the value of each of the at least one initial condition identified during the repeating step.

14. The method according to claim 1, wherein the set of initial condition includes at least one of a threshold grayscale value, a determination of skew correction, and a determination of type of object.

7. A computer program product comprising a computer readable medium having computer executable program code embodied therein for recognizing at least one object in a digitized representation of an image, the computer program product comprising:

computer executable program code configured to cause a computer to receive the digitized representation of the image, the representation having a first resolution;

computer executable program code configured to cause a computer to create a reduced-resolution version of the image responsive to the digitized representation of the image, the reduced-resolution version of the image having a second resolution lower than the first resolution;

computer executable program code configured to cause a computer to provide a plurality of sets of initial conditions, the initial conditions including at least a condition for character recognition-processing of the image;

for each of the sets of initial conditions, computer executable program code configured to identify each confidence level of character recognition by first character recognition-processing of the reduced resolution version of the image having the second resolution based on each of the sets of initial conditions, each of the sets of initial conditions including a threshold value to binarize an image;

computer executable program code configured to select an optimal set of initial conditions from the plurality of sets of initial conditions based on each confidence level identified, the optimal set of initial conditions including a threshold value to binarize the image; and

computer executable program code configured to cause a computer to perform second character recognition-processing of the objects represented in the digitized representation of the image having the first resolution by using the optimal set of initial conditions selected.

8. The computer program product according to claim 7, wherein the computer executable program code is configured to cause a computer to select the optimal set from the plurality of sets of initial conditions based on a highest confidence level identified.

9. The computer program product according to claim 7, wherein the computer executable program code is configured to cause a computer to select the optimal set from the plurality of sets of initial conditions based on a confidence level exceeding a threshold.

10. The computer program product according to claim 7, wherein the computer executable program code is configured to cause a computer to create a reduced resolution version of the image creates by calculating an avenge of a plurality of pixels of the digitized representation of the image having the first resolution.

11. The computer program product according to claim 7, additionally comprising computer executable program code is configured to cause a computer to recognize at least one additional object represented in the digitized representation of the image, responsive to the value of at least one initial condition identified responsive to a confidence level exceeding a threshold.

12. The computer program product according to claim 7, additionally comprising:

computer executable program code configured to cause a computer to attempt to recognize at least one additional object represented in the digitized representation of the image, responsive to the value of at least one initial condition identified, and configured to cause a computer to produce a confidence level of the attempt; and

computer executable program code configured to cause a computer to, responsive to the confidence level of the attempt below a threshold;

repeat the identifying step; and

recognize the at least one object represented in the digitized

representation of the image responsive to the value of each of the at

least one initial condition identified during the repeat operation.

15. The computer program according to claim 7, wherein the set of initial condition includes at least one of a threshold grayscale value, a determination of skew correction, and a determination of type of object.

13. A system for recognizing objects, the system comprising:

a downsampler having an input for receiving a representation of an image having a first resolution, the downsampler for producing and providing at an output thereof a reduced-resolution version of the image responsive to the representation of the image received at the downsampler input the reduced resolution version of the image having a second resolution lower than the first resolution; and

a recognition engine having a first input coupled to the downsampler output for receiving the reduced-resolution version of the image and a second input for receiving the representation of the image, the recognition engine for recognizing at least one object in the digitized representation of the image by a method comprising the steps of;

providing a plurality of sets of initial conditions, the initial conditions including at least a condition for character recognition-processing of the image;

for each of the sets of initial conditions, identifying each confidence level of character recognition by first character recognition-processing of the reduced resolution version of the image having the second resolution based on each of the sets of initial conditions, each of the sets of initial conditions including a threshold value to binarize an image;

selecting an optimal set of initial conditions from the plurality of sets of initial conditions based on each confidence level identified in said identifying step, the optimal set of initial conditions including a threshold value to binarize the image; and

second character recognition processing of the objects represented in the digitized representation of the image having the first resolution by using the optimal set of initial conditions selected in said selecting step.

16. The system according to claim 13, wherein the set of initial condition includes at least one of a threshold grayscale value, a determination of skew correction, and a determination of type of object.