Digital image stabilization device
AssigneeSamsung Electronics Co., Ltd.
InventorsSungSoo Park, Matthias Braun
Life of this patent
Estimating the term…
Abstract
A method of Digital Image Stabilization (DIS) including a feature point sorting algorithm for selecting optimal feature points, and a computationally efficient tile-vector based Hierarchical Block-Matching search algorithm for deriving motion vectors of the selected feature points, and a feature point motion vector grouping/comparison process for grouping pairs of the selected feature points based on their vector magnitude ratio and their angular difference.
Claims26
A method of processing video data using a video processing circuit, comprising:
- receiving first image data that represents a first frame;
- identifying a plurality of feature points in the first frame;
- receiving second image data that represents a second frame;
- deriving a plurality of motion vectors, each of the plurality of motion vectors corresponding to each respective feature point;
- selecting a first vector and a second vector among the plurality of motion vectors;
- and
- comparing the first vector with the second vector based on a vector magnitude ratio and a vector angular difference of the first vector and the second vector.
A video processing circuit, comprising:
- a feature point circuit configured to identify a plurality of feature points in a first frame and deriving a plurality of motion vectors between the first frame and a second frame, each of the plurality of motion vectors corresponding to each respective feature point;
- a paring controller configured to select a first vector and a second vector among the plurality of motion vectors;
- a magnitude ratio comparator configured to compare the first vector with the second vector based on a vector magnitude ratio of the first vector and the second vector;
- and
- a vector angle comparator configured to compare the first vector with the second vector based on a vector angular difference of the first vector and the second vector.
A method of processing video data using a video processing circuit, comprising:
- estimating motion vectors for a plurality of feature points in a first frame of the video data;
- grouping the motion vectors into a plurality of motion vector subsets based on vector magnitude ratios and vector angular differences of the motion vectors;
- and
- selecting one of the plurality of motion vector subsets that contains the motion vectors representing a movement of a stationary object within a scene of the first frame.
A camera, comprising:
- an image capture circuit configured to capture images and convert a plurality of images to a first frame and a second frame of image data;
- and
- a video processing circuit chip comprising:
- a feature point circuit configured to identify a plurality of feature points in the first frame and deriving a plurality of motion vectors between the first frame and the second frame, each of the plurality of motion vectors corresponding to each respective feature point;
- a paring controller configured to select a first vector and a second vector among the plurality of motion vectors;
- a magnitude ratio comparator configured to compare the first vector with the second vector based on a vector magnitude ratio of the first vector and the second vector;
- and
- a vector angle comparator configured to compare the first vector with the second vector based on a vector angular difference of the first vector and the second vector.
Dependencies are inferred from the claim text.