Method for fast locating decipherable pattern

• US 8,469,274 B2
• Filed: 10/23/2009
• Issued: 06/25/2013
• Est. Priority Date: 12/31/2008
• Status: Active Grant

• Current Assignee: Armorlink SH Corp. (IEI Integration Corp.), IEI Integration Corp.
• Original Assignee: Armorlink SH Corp. (IEI Integration Corp.), IEI Integration Corp.
• Inventors: Jheng, Ya-Yun, Chen, Shih-Hsun, Cheng, Hai-Peng, Hsiang, Te-Heng, Tseng, Chun-Shun, Jhuang, Ming-Chi, Wang, Jung-Hua
• Primary Examiner(s): Hess, Daniel
• Assistant Examiner(s): Savusdiphol, Paultep
• Application Number:
  US12/604,402
• Publication Number:
  US 20100163632A1
• Time in Patent Office: 1,341 Days
• Field of Search: 235/454, 235/462.01, 235/462.09, 235/462.1, 235/462.12, 235/462.19, 235/462.27, 235/487, 235/494
• US Class Current: 235/462.08
• CPC Class Codes: G06V 10/25 Determination of region of interest [ROI] or a volume of interest [VOI]

Abstract

A method for fast locating a decipherable pattern in an input image, which is characterized in utilizing an overly downscaled binary image to not only reduce computation time but also facilitate extraction of skeletons for fast and accurately locating pattern, is disclosed. First, a pre-process is applied to an input image to acquire a binary image downscaled n times, from which at least one skeleton corresponding to a decipherable pattern is extracted. Coordinate values of at least one pixel of each skeleton are respectively enlarged n^1/2 times and used as the central points on the original image plane for establishing a plurality of detecting blocks with the identical size. Subsequently, a grading mechanism is employed to determine the corresponding detecting blocks of the decipherable pattern.

Claims

1. A method for fast locating a decipherable pattern in an input image, which downscales the input image to extract a plurality of skeletons, comprising the steps of:

(A) applying a pre-process to the input image to acquire a binary image downscaled n times, wherein n is a real number;

(B) identifying a plurality of black lines so as to extract at least one skeleton;

(C) enlarging each of the coordinate values of at least one pixel of the at least one skeleton by n^1/2 times, and on the input image plane, using the enlarged coordinate values as the central points for establishing a plurality of detecting blocks, wherein each detecting block contains a plurality of pixel feature values corresponding to the input image plane; and

(D) employing a grading mechanism to determine the corresponding detecting blocks of the decipherable pattern.

2. The method according to claim 1, wherein the step (A) further comprises pre-process steps of:

(A-1) downscaling the input image in order to decrease the number of pixels;

(A-2) filtering the input image by using a low-pass filter;

(A-3) compensating the illumination of the input image;

(A-4) binarizing the pixel feature values of the input image; and

(A-5) filling the broken areas of the input image and eliminating the redundant black pixels of the input image.

3. The method according to claim 2, wherein the step (A-1) uses a bilinear interpolation to downscale the input image.

4. The method according to claim 2, wherein the low-pass filter used in the step (A-2) is a mean filter.

5. The method according to claim 2, wherein the step (A-3) uses a log transformation to compensate illumination of the input image.

6. The method according to claim 2, wherein the step (A-4) uses a constant threshold or an adaptive threshold to binarize the pixel feature values of the input image.

7. The method according to claim 2, wherein the step (A-5) uses a median filter to fill the broken areas of the input image and eliminating the redundant black pixels of the input image.

8. The method according to claim 1, wherein a scanning process is used in the step (B) to detect the black lines by scanning the binary image downscaled n times, pixel by pixel in two different directions, comprising the steps of:

(B-1) detecting a first location at which a black pixel appears after a white pixel;

(B-2) detecting a second location at which a black pixel appears before a white pixel;

(B-3) connecting the first location and the second location; and

(B-4) repeating the step (B-1) to the step (B-3) until all pixels of the binary image have been scanned to acquire the black lines.

9. The method according to claim 8, wherein the directions for the scanning process are perpendicular to one another.

10. The method according to claim 9, wherein the scanning process is done in an interlaced scan manner.

11. The method according to claim 9, wherein the skeletons are extracted by finding the middle points of the black lines scanned in the first direction and then connecting the adjacent middle points;

and finding the middle points of the black lines scanned in the second direction and then connecting the adjacent middle points.

12. The method according to claim 1, wherein the detecting blocks are rectangular and have an identical size.

13. The method according to claim 12, wherein the ratio between width M of the input image and width W of the detecting blocks is M/W, wherein 12≦

M/W≦

64; and

the ratio between height N of the input image and height H of the detecting blocks is N/H, where N/H=M/W.

14. The method according to claim 1, wherein the input image is a grayscale image, and the pixel feature values are the grayscale values of the grayscale image respectively.

15. The method according to claim 1, wherein the input image is a color image transformed into a grayscale image, and the pixel feature values are the grayscale values of the grayscale image respectively.

16. The method according to claim 1, wherein the decipherable pattern contains a plurality of lines at intervals.

17. The method according to claim 16, wherein the decipherable pattern is 1-D barcode.

18. The method according to claim 16, wherein the decipherable pattern is 2-D barcode.

19. A computer-readable storage medium stores computer executable program implementing a method according to claim 1.