US 9,811,887 B2
Apparatus and method for image stabilization using image blur correction
Seungho Shin, Incheon (KR)
Assigned to SK TELECOM CO., LTD., Seoul (KR)
Filed by SK TELECOM CO., LTD., Seoul (KR)
Filed on Mar. 14, 2016, as Appl. No. 15/69,212.
Application 15/069,212 is a continuation of application No. PCT/KR2014/008564, filed on Sep. 15, 2014.
Claims priority of application No. 10-2013-0111366 (KR), filed on Sep. 16, 2013.
Prior Publication US 2016/0196639 A1, Jul. 7, 2016
Int. Cl. G06T 5/00 (2006.01); H04N 5/232 (2006.01); G06K 9/46 (2006.01); G06K 9/62 (2006.01); G06T 5/40 (2006.01); G06T 7/20 (2017.01); G06T 7/238 (2017.01); G06T 7/246 (2017.01)
CPC G06T 5/003 (2013.01) [G06K 9/4609 (2013.01); G06K 9/4652 (2013.01); G06K 9/4661 (2013.01); G06K 9/6202 (2013.01); G06T 5/40 (2013.01); G06T 7/20 (2013.01); G06T 7/238 (2017.01); G06T 7/246 (2017.01); H04N 5/23254 (2013.01); H04N 5/23267 (2013.01); G06T 2207/10024 (2013.01); G06T 2207/20021 (2013.01); G06T 2207/20201 (2013.01)] 16 Claims
OG exemplary drawing
 
1. An apparatus for stabilizing an image based on a correction of a blur of the image, the apparatus comprising:
an image receiving unit, implemented by one or more processors, configured to receive the image;
a block dividing unit, implemented by one or more processors, configured to divide a frame of the image into blocks, and merge the blocks having similar object distributions based on a density of edges and histogram distribution characteristics in each of the blocks;
an image analyzing unit, implemented by one or more processors, configured to analyze characteristics of each block to calculate a weight of each block by calculating an object distribution based on the density of edges and histogram distribution characteristics of each of the blocks;
a significant block extracting unit, implemented by one or more processors, configured to
extract, from the blocks, a block having a weight equal to or larger than a threshold value,
determine the extracted block to be a significant block;
a feature point extracting unit, implemented by one or more processors, configured to
extract pixels from the significant block, and
extract a feature point of an object in the significant block by comparing at least one of color and brightness between the extracted pixels and neighboring pixels of the extracted pixels, wherein the feature point indicates a pixel corresponding to a part of the object in the image;
a local motion vector calculating unit, implemented by one or more processors, configured to calculate at least one local motion vector of at least one block corresponding to the feature point by referring a previous frame of the frame and estimating a motion of the at least one block;
a global motion vector calculating unit, implemented by one or more processors, configured to calculate a global motion vector based on the calculated at least one local motion vector of the at least one block; and
an image correcting unit, implemented by one or more processors, configured to correct output data of the frame based on the global motion vector.