Object-of-interest image capture

• US 7,095,892 B2
• Filed: 01/26/2005
• Issued: 08/22/2006
• Est. Priority Date: 01/26/2005
• Status: Active Grant

• Current Assignee: Motorola Solutions, Inc.
• Original Assignee: Motorola, Inc. (Motorola Solutions, Inc.)
• Inventors: Tang, Bei, Beuhler, Allyson J., Lee, King F.
• Primary Examiner(s): Alavi, Amir
• Application Number:
  US11/044,738
• Publication Number:
  US 20060165288A1
• Time in Patent Office: 573 Days
• Field of Search: 382/162, 382/172, 382/181, 382/182, 382/195, 382/219, 382/220, 382/276, 382/277, 358/509, 358/520, 348/169, 348/229.1, 348/234, 348/256, 348/364, 396/63, 396/65, 396/89, 396/106, 396/144, 396/155, 396/233, 396/324
• US Class Current: 382/182
• CPC Class Codes: G06V 20/62 Text, e.g. of license plates, overlay texts or captions on TV images

Abstract

A method and apparatus for object-of-interest image capture accomplishes the functions of capturing (120) a current image, determining (125) an object region of the object-of-interest in the current image, calculating (130) a contrast measurement of the object region, and identifying (135) the current image as having an optimized object region when an iteration measure meets a criterion. The current image is captured using a local exposure value determined from luminance readings of pixels within an object region of an object-of-interest of a previous image.

Claims

1. A method for object-of-interest image capture, comprising:

capturing a current image using a local exposure value determined from luminance readings of pixels within an object region of an object-of-interest of a previous image;

determining an object region of the object-of-interest in the current image;

calculating a contrast measurement of the object region; and

identifying the current image as having an optimized object region when an iteration measure meets a criterion wherein the local exposure value is determined from a sum of absolute differences (SAD) of luminance readings of adjacent pixels within the object region.

2. The method according to claim 1, further comprising:

determining an object region of the object-of-interest within a target image captured using an initial exposure value; and

setting the local exposure value based on luminance readings of pixels within the object region.

3. The method according to claim 2, wherein the initial exposure value is determined from a global luminance of the target image.

4. The method according to claim 1, further comprising:

performing object recognition of a character based object using the optimized object region.

5. The method according to claim 1, wherein the local exposure value is determined from a local average luminance of the object region.

6. The method according to claim 1, wherein the local exposure value is determined in a manner designed to improve the contrast measurement of the object region in an image captured using the local exposure value so determined.

7. The method according to claim 1, wherein the local exposure value is determined by:

8. The method according to claim 1, wherein the contrast measurement is a sum of absolute differences of luminance values of adjacent pixels, for essentially all pixels within the object region.

9. The method according to claim 1, wherein the criterion is that the contrast measurement is greater than an established value.

10. The method according to claim 1, wherein the criterion is that the current contrast measurement is greater than a contrast measurement made using a previous object region of a previous image, by an established value.

11. The method according to claim 1, wherein the criterion is that the local exposure value is different from a previous local exposure value by less than an established value.

12. The method according to claim 1, wherein the criterion includes that a number of iterations has reached a maximum value.

13. The method according to claim 1, wherein the current image is a current frame of a series of periodic video frames.

14. The method according to claim 1, wherein the object-of-interest substantially comprises characters.

15. The method according to claim 1, further comprising, when the contrast measurement fails to meet the criterion:

setting a local exposure value; and

repeating the capturing of a current image, the determining of the object region, and the calculating of a contrast measurement.

16. A means for object-of-interest image capture, comprising:

a means for capturing a current image;

a means for setting an exposure value that determines a local exposure value from a local luminance reading of pixels within an object region of an object-of-interest of a previous image;

a means for local luminance determination that generates the local luminance reading;

a means for locating the object region of the object-of-interest in the current image; and

a means for calculating a contrast measurement of the object region and identifying the current image as having an optimized object region when an iteration measure meets a criterion wherein the local exposure value is determined from a sum of absolute differences (SAD) of luminance readings of adjacent pixels within the object region.

17. An electronic device for object-of-interest image capture, comprising:

an electronic device for capturing a current image; and

a processing function that includes

an exposure value setting function that determines a local exposure value from a local luminance reading of pixels within an object region of an object-of-interest of a previous image;

a local luminance determination function that generates the local luminance reading;

a locating function that locates the object region of the object-of-interest in the current image; and

a calculation function that calculates a contrast measurement of the object region and identifying the current image as having an optimized object region when an iteration measure meets a criterion wherein the local exposure value is determined from a sum of absolute differences (SAD) of luminance readings of adjacent pixels within the object region.