Apparatus and method for processing a depth image

Patent metadata

Field	Value
Publication number	US-9811921-B2
Application number	US-201515026870-A
Country	US
Kind code	B2
Filing date	Nov 5, 2015
Priority date	May 11, 2015
Publication date	Nov 7, 2017
Grant date	Nov 7, 2017

Abstract

Embodiments of the present invention disclose method and apparatus for processing a depth image. The method comprises: obtaining the depth image and a captured image corresponding to the depth image; segmenting the depth image to obtain a plurality of segmentation units; calculating a corresponding depth reference value for each segmentation unit; determining a standard range of depth value for the each segmentation unit based on the corresponding depth reference value; and adjusting a depth value of each pixel of the depth image to the standard range of depth value corresponding to the segmentation unit in which the each pixel is located. The method and the apparatus of embodiments of the present invention can improve the quality of the depth image and cause the depth image to be easily identified.

First claim

What is claimed is:

1. A method for processing a depth image, the method comprising:
obtaining the depth image and a captured image corresponding to the depth image;
segmenting the depth image to obtain a plurality of segmentation units;
calculating, for each segmentation unit, a corresponding reference value for a depth of said segmentation unit;
determining a standard range of depth value for each segmentation unit based on the corresponding reference value; and
adjusting a depth value of each pixel of the depth image according to the standard range of depth value corresponding to the segmentation unit in which the pixel is located.

2. The method according to claim 1 , wherein segmenting the depth image to obtain a plurality of segmentation units comprises:
performing contour segmentation on the captured image to obtain contour segmentation information; and
segmenting the depth image to obtain the plurality of segmentation units based on the contour segmentation information.

3. The method according to claim 2 , wherein performing contour segmentation on the captured image to obtain contour segmentation information comprises:
performing gray processing on the captured image to obtain a gray image;
performing edge extraction on the gray image to obtain a contour image;
performing contour dilation processing on the contour image to obtain a contour dilation image;
reversing the contour dilation image to obtain a reversed image; and
calculating the contour segmentation information of the reversed image using a watershed algorithm.

4. The method according to claim 3 , wherein the calculating a corresponding depth reference value for each segmentation unit comprises:
removing a black point pixel and a bright point pixel in the segmentation unit;
counting, for each one of different depth values of the segmentation unit in which the black point pixel and the bright point pixel have been removed, the number of pixels having said value; and
determining the depth value having the largest number of pixels as the reference value for the segmentation unit.

5. The method according to claim 2 , wherein calculating a corresponding depth reference value for each segmentation unit comprises:
removing a black point pixel and a bright point pixel in the segmentation unit;
counting, for each one of different depth values of the segmentation unit in which the black point pixel and the bright point pixel have been removed, the number of pixels having said value; and
determining the depth value having the largest number of pixels as the reference value for the segmentation unit.

6. The method according to claim 1 , wherein calculating a corresponding depth reference value for each segmentation unit comprises:
removing a black point pixel and a bright point pixel in the segmentation unit;
counting, for each one of different depth values of the segmentation unit in which the black point pixel and the bright point pixel have been removed, the number of pixels having said depth value; and
determining the depth value having the largest number of pixels as the reference value for the segmentation unit.

7. The method according to claim 1 , wherein the standard range of depth value for each segmentation unit is 0.5˜1.3 times the reference value for said segmentation unit.

8. The method according to claim 1 , wherein adjusting a depth value of each pixel of the depth image to the standard range of depth value corresponding to the segmentation unit in which the each pixel is located, comprises:
traversing a pixel of the depth image along a set direction, wherein in the traversing procedure:
if the depth value of a current pixel is beyond the standard range of depth value corresponding to the segmentation unit in which the current pixel is located, then the depth values of surrounding pixels centering on the current pixel in the depth image are sequentially read and spread outward a set number of loops;
if a pixel currently being read is within the segmentation unit and its depth value is within the standard range of depth value corresponding to the segmentation unit, then it may quit the loop and the depth value of the current pixel is adjusted as the depth value of the pixel currently being read; and
if the set number of loops is finished and the depth value of the current pixel is not adjusted, then the depth value of the current pixel is adjusted to be the reference value corresponding to the segmentation unit.

9. The method according to claim 8 , wherein the surrounding pixels are located on a cross type direction of the current pixel.

10. The method according to claim 9 , wherein:
the set number of loops is five; and
the set direction is from left to right or from right to left in a row, or from up to down or from down to up in a column.

11. An apparatus for processing a depth image, the apparatus comprising:
an obtaining module configured to obtain the depth image and a captured image corresponding to the depth image;
a segmenting module configured to segment the depth image to obtain a plurality of segmentation units;
a calculating module configured to calculate, for each segmentation unit, a corresponding reference value for a depth of said segmentation unit;
a determining module configured to determine a standard range of depth value for the each segmentation unit based on the corresponding reference value of said segmentation unit; and
an adjusting module configured to adjust a depth value of each pixel of the depth image according to the standard range of depth value corresponding to the segmentation unit in which the pixel is located.

12. The apparatus according to claim 11 , wherein the segmenting module is further configured to:
perform contour segmentation on the captured image to obtain contour segmentation information; and
segment the depth image to obtain the plurality of segmentation units based on the contour segmentation information.

13. The apparatus of according to claim 12 , wherein the segmenting module is further configured to:
perform gray processing on the captured image to obtain a gray image;
perform edge extraction on the gray image to obtain a contour image;
perform contour dilation processing on the contour image to obtain a contour dilation image;
reverse the contour dilation image to obtain a reversed image; and
calculate the contour segmentation information of the reversed image using a watershed algorithm.

14. The apparatus according to claim 13 , wherein the calculating module is further configured to:
remove a black point pixel and a bright point pixel in the segmentation unit;
count, for each one of different depth values of the segmentation unit in which the black point pixel and the bright point pixel have been removed, the number of pixels having said value; and
determine the depth value having the largest number of pixels as the reference value for the segmentation unit.

15. The apparatus according to claim 12 , wherein the calculating module is further configured to:
remove a black point pixel and a bright point pixel in the segmentation unit;
count, for each one of different depth values of the segmentation unit in which the black point pixel and the bright point pixel have been removed, the number of pixels having said value; and
determine the depth value having the largest number of pixels as the reference value for the segmentation unit.

16. The apparatus according to claim 11 , wherein the calculating module is further configured to:
remove a black point pixel and a bright point pixel in the segmentation uni

Assignees

• Boe Technology Group Co Ltd

Inventors

• Zhao Jibo
• Zhao Xingxing
• Jin Xiaoli

Classifications

• G06T7/12

  Edge-based segmentation · CPC title

• H04N2013/0081

  Depth or disparity estimation from stereoscopic image signals · CPC title

• G06T7/149

  involving deformable models, e.g. active contour models · CPC title

• G06T2207/20021

  Dividing image into blocks, subimages or windows · CPC title

• G06T7/11

  Region-based segmentation · CPC title