IMAGE PROCESSING METHOD AND APPARATUS FOR DETECTING LINES OF IMAGES AND START AND END POINTS OF LINES

• US 20090245687A1
• Filed: 03/27/2009
• Published: 10/01/2009
• Est. Priority Date: 03/31/2008
• Status: Active Grant

• Current Assignee: Sungkyunkwan University Foundation For Corporate Collaboration (SungKyunKwan University)
• Original Assignee: Sungkyunkwan University Foundation For Corporate Collaboration (SungKyunKwan University)
• Inventors: Jeon, Jae Wook, Jin, Seung Hun, Kim, Ki Hoon, Lee, Sang Jun, Kim, Dong Kyun, Nguyen, Tuong Thuy
• Granted Patent: US 8,249,387 B2
• Time in Patent Office: Days
• US Class Current: 382/281
• CPC Class Codes: G06V 10/48 by mapping characteristic values of the pattern into a parameter space, e.g. Hough transformation

Abstract

Disclosed herein is an image processing method and apparatus for detecting the lines of images and the start and end points of the lines. The image processing apparatus includes an edge creation unit, a Hough transform unit, and an effective parameter detection unit. The edge creation unit creates an edge image using external image data input from the outside. The Hough transform unit performs a Hough transform on information about the pixel coordinates of the edge image created by the edge creation unit. The effective parameter detection unit detects the lines of the edge image by checking effective line parameters using the results of the Hough transform. The image processing apparatus may further include an edge list for storing coordinates of effective pixels constituting the edge image and a line parameter list for storing the effective line parameters.

Claims

1. An image processing apparatus for detecting lines of images, comprising:

an edge creation unit for creating an edge image using external image data input from an outside;

a Hough transform unit for performing a Hough transform on information about pixel coordinates of the edge image created by the edge creation unit; and

an effective parameter detection unit for detecting lines of the edge image by checking effective line parameters using results of the Hough transform.

2. The image processing apparatus as set forth in claim 1, further comprising:

an edge list for storing coordinates of effective pixels constituting the edge image created by the edge creation unit; and

a line parameter list for storing the effective line parameters checked by the effective parameter detection unit.

3. The image processing apparatus as set forth in claim 1, wherein the Hough transform unit performs the Hough transform on a plurality of edge points for θ

values in a predetermined range by calculating the following equation;

x cos θ

+y sin θ

=ρ

where x and y are x and y coordinates values of each edge point, respectively.

4. The image processing apparatus as set forth in claim 3, wherein the Hough transform unit performs a plurality of calculations of the equation at one time in a parallel manner.

5. The image processing apparatus as set forth in claim 3, wherein the Hough transform unit increases a value of a parameter cell corresponding to ρ

and θ

acquired using the Equation, by 1.

6. The image processing apparatus as set forth in claim 5, wherein the Hough transform unit comprises a plurality of trigonometrical arithmetic units and basic arithmetic units in order to perform a plurality of calculations of the equation in a parallel manner.

7. The image processing apparatus as set forth in claim 5, wherein the effective parameter detection unit determines a parameter corresponding to each cell of the parameter memory to be an effective parameter if a value of the cell of the parameter memory is equal to or greater than a predetermined value.

8. The image processing apparatus as set forth in claim 7, wherein the effective parameter detection unit searches for a cell having the maximum of values of a plurality of parameter memory cells in a predetermined region of a parameter space and determines the found cell to be the effective parameter if a value of the found cell is greater than a reference value.

9. The image processing apparatus as set forth in claim 8, wherein the effective parameter detection unit comprises:

a first the maximum value calculator for receiving values of the plurality of parameter memory cells for each clock and outputting a maximum value stored in a parameter memory cell having the maximum value;

a plurality of registers for temporarily storing maximum cell values output by the first the maximum value calculator;

a second maximum value calculator for outputting the maximum of the maximum cell values stored in the plurality of registers; and

a reference value comparator for comparing the value output from the second maximum value calculator with the reference value.

10. The image processing apparatus as set forth in claim 9, wherein the plurality of registers has a shift register configuration.

11. An image processing apparatus for detecting start and end points of lines, comprising:

an edge creation unit for creating an edge image using external image data input from an outside and storing it in memory;

a Hough transform unit for acquiring line parameters by performing a Hough transform on information about pixel coordinates of the edge image created by the edge creation unit; and

a matching unit for detecting start and end points of an infinite line image represented by the detected line parameters using the edge image stored in the memory unit.

12. The image processing apparatus as set forth in claim 11, wherein the matching unit acquires only pixels matching the stored edge image from pixels of an infinite line image represented by the line parameters by matching the infinite line image with the edge image, and checks start and end points of segments formed by the matching pixels.

13. The image processing apparatus as set forth in claim 12, wherein the matching unit performs the process of acquiring only pixels matching the edge image from pixels of the infinite line image by matching the infinite line image with the edge image and checking start and end points of the segments formed by the matching pixels for each of the line parameters.

14. The image processing apparatus as set forth in claim 11, further comprising an effective parameter detection unit for detecting effective line parameters from the line parameters on which the Hough transform has been performed.

15. The image processing apparatus as set forth in claim 14, further comprising:

an edge image storage unit for storing coordinates of effective pixels constituting the edge image created by the edge creation unit; and

an effective line parameter list for storing the effective line parameters checked by the effective parameter detection unit.

16. The image processing apparatus as set forth in claim 14, wherein the Hough transform unit performs the Hough transform on a plurality of edge points for θ

values in a predetermined range by calculating the following equation;

x cos θ

+y sin θ

=ρ

where x and y are x and y coordinates values of each edge point, respectively.

17. The image processing apparatus as set forth in claim 16, wherein the Hough transform unit performs a plurality of calculations of the equation at one time in a parallel manner.

18. The image processing apparatus as set forth in claim 17, wherein the Hough transform unit increases a value of a parameter cell corresponding to ρ

and θ

acquired using the Equation, by 1.

19. The image processing apparatus as set forth in claim 18, wherein the effective parameter detection unit determines a parameter corresponding to each cell of the parameter memory to be an effective parameter if a value of the cell of the parameter memory is equal to or greater than a predetermined value.

20. The image processing apparatus as set forth in claim 19, wherein the effective parameter detection unit searches for a cell having the maximum of values of a plurality of parameter memory cells in a predetermined region of a parameter space and determines the found cell to be the effective parameter if a value of the found cell is greater than a reference value.