Method and apparatus for processing line pattern using convolution kernel

• US 7,657,099 B2
• Filed: 02/10/2006
• Issued: 02/02/2010
• Est. Priority Date: 02/14/2005
• Status: Active Grant

• Current Assignee: Samsung Electronics Co. Ltd.
• Original Assignee: Samsung Electronics Co. Ltd.
• Inventors: Bang, Seok-won, Park, Dong-ryeol
• Primary Examiner(s): Bali; Vikkram
• Assistant Examiner(s): ENTEZARI, MICHELLE M
• Application Number:
  US11/350,847
• Publication Number:
  US 20060182365A1
• Time in Patent Office: 1,453 Days
• Field of Search: 382/205, 382/255, 382/263, 382/193, 382/209, 382/261, 382/265
• US Class Current: 382/205
• CPC Class Codes: G06V 10/145 Illumination specially adapted for pattern recognition, e.g. using gratings
• CPC Class Codes: G06V 10/36 Applying a local operator, i.e. means to operate on image points situated in the vicinity of a given point; Non-linear local filtering operations, e.g. median filtering
• CPC Class Codes: G06V 20/10 Terrestrial scenes scenes under surveillance with static cameras G06V20/52; scenes perceived from the exterior of a vehicle G06V20/56; scenes perceived from the interior of a vehicle G06V20/59

Abstract

A method and apparatus for processing a line pattern using a convolution kernel. An apparatus for processing a line pattern, the apparatus including a light source module radiating predetermined light and a camera module capturing an image of an object, onto which the light is radiated, the apparatus processing a line pattern included in the captured image, includes: a kernel learning unit determining a kernel shape and an optimal threshold value with respect to each vertical interval of the image using a simple line pattern formed at a predetermined location; a convolution application unit applying convolution to a captured test image using the determined kernel shape and optimal threshold value; and a region selection unit scanning each pixel column in the image, to which the convolution is applied, in a vertical direction, and setting pixel values of pixel groups other than a selected pixel group to 0 when a plurality of pixel groups having pixel values other than 0 exists.

Claims

1. An apparatus for processing a line pattern, the apparatus including a light source module radiating predetermined light and a camera module capturing an image of an object, onto which the light is radiated, the apparatus processing a line pattern included in the captured image, comprising:

a kernel learning unit determining a kernel shape and an optimal threshold value with respect to each vertical interval of the image using a simple line pattern formed at a predetermined location;

a convolution application unit applying convolution to a captured test image using the determined kernel shape and optimal threshold value; and

a region selection unit scanning each pixel column in the image, to which the convolution is applied, in a vertical direction, and setting pixel values of pixel groups other than a selected pixel group to 0 when a plurality of pixel groups having pixel values other than 0 exists.

2. The line pattern processing apparatus according to claim 1, wherein the light is line laser light.

3. The line pattern processing apparatus according to claim 1, wherein the camera module comprises a band pass filter receiving only a signal having a frequency close to the frequency of the laser light.

4. The line pattern processing apparatus according to claim 1, wherein the kernel shape includes −

m, (n−

m)/2 0s, m 1s, (n−

m)/2 0s, and −

m in order, m being the minimum kernel size and n being the maximum kernel size.

5. The line pattern processing apparatus according to claim 1, wherein the kernel shape includes two −

m/2, (n−

m)/2 0s, m 1s, (n−

m)/2 0s, and two −

m/2 in order, m being the minimum kernel size and n being the maximum kernel size.

6. The line pattern processing apparatus according to claim 4, wherein the optimal threshold value is determined using the minimum kernel size m.

7. The line pattern processing apparatus according to claim 5, wherein the optimal threshold value is determined using the minimum kernel size m.

8. The line pattern processing apparatus according to claim 1, further comprising a distance information calculation unit for calculating a distance to an obstacle by applying a triangulation method to an image having passed through the region selection unit.

9. The line pattern processing apparatus according to claim 1, wherein the convolution application unit allows only pixels exceeding the optimal threshold value to pass therethrough by applying a convolution kernel corresponding to the determined kernel shape to the test image while moving along each pixel column included in the test image.

10. A method of processing a line pattern, the method capturing an image of an object, onto which predetermined light is radiated, and processing a line pattern included in the captured image, comprising:

using at least one processing device determining a kernel shape and an optimal threshold value with respect to each vertical interval of the image using a simple line pattern formed at a predetermined location;

applying convolution to a captured test image using the determined kernel shape and optimal threshold value; and

scanning each pixel column included in the image, to which the convolution is to be applied, in a vertical direction, and setting pixel values of pixel groups, other than a selected pixel group, to 0 when a plurality of pixel groups having pixel values other than 0 exists.

11. The line pattern processing method according to claim 10, wherein the kernel shape includes −

m, (n−

m)/2 0s, m 1s, (n−

m)/2 0s, and −

m in order, m being the minimum kernel size and n being the maximum kernel size.

12. The line pattern processing method according to claim 10, wherein the kernel shape includes two −

m/2, (n−

m)/2 0s, m 1s, (n−

m)/2 0s, and two −

m/2 in order, m being the minimum kernel size and n being the maximum kernel size.

13. The line pattern processing method according to claim 11, wherein the optimal threshold value is determined using the minimum kernel size m.

14. The line pattern processing method according to claim 12, wherein the optimal threshold value is determined using the minimum kernel size m.

15. The line pattern processing method according to claim 10, further comprising calculating a distance to an obstacle by applying a triangulation method to an image output after the scanning has been performed.

16. The line pattern processing method according to claim 10, wherein the applying convolution comprises allowing only pixels exceeding the optimal threshold value to pass therethrough by applying a convolution kernel corresponding to the determined kernel shape to the test image while moving along each pixel column included in the test image.

17. The line pattern processing method according to claim 10, wherein the determining a kernel shape comprises:

binarizing the simple line pattern image;

determining a minimum kernel size and a maximum kernel size depending on a vertical thickness of the line pattern of the binarized image, and selecting a kernel shape based on the determined minimum and maximum kernel sizes;

setting a threshold value interval using the minimum kernel size;

applying a convolution kernel to the image formed as the simple line pattern, using the selected kernel shape and a temporary threshold value belonging to the threshold value interval; and

changing the temporary threshold value to reduce an error between the binarized image and the image to which the convolution kernel is applied.

18. The line pattern processing method according to claim 17, wherein the determining a minimum kernel size and a maximum kernel size, applying a convolution kernel to the image, and changing a temporary threshold value are repeated until the error is less than a predetermined threshold value.

19. The line pattern processing method according to claim 10, wherein the scanning comprises:

scanning in a vertical direction a pixel group having pixel values other than 0 with respect to a certain pixel column of the test image;

determining whether a pixel thickness of a current pixel group is greater than that of a previously stored pixel group;

changing all pixel values of the previously stored pixel group to 0 when the pixel thickness of the current pixel group is greater than that of the previously stored pixel group; and

changing all pixel values of the current pixel group to 0 when the pixel thickness of the current pixel group is not greater than that of the previously stored pixel group.

20. A computer-readable storage medium encoded with processing instructions for causing a processor to execute a method of processing a line pattern, the method capturing an image of an object, onto which predetermined light is radiated, and processing a line pattern included in the captured image, the method comprising:

determining a kernel shape and an optimal threshold value with respect to each vertical interval of the image using a simple line pattern formed at a predetermined location;

applying convolution to a captured test image using the determined kernel shape and optimal threshold value; and

scanning each pixel column included in the image, to which the convolution is to be applied, in a vertical direction, and setting pixel values of pixel groups, other than a selected pixel group, to 0 when a plurality of pixel groups having pixel values other than 0 exists.