Optical Flow Computation Method for Large Displacement Motion Based on Depth Matching
A calculation method and large displacement technology, applied in calculation, image analysis, image enhancement and other directions, can solve the problems of inaccurate pixel matching relationship and too few matching pixels, so as to improve the accuracy of optical flow calculation and have a wide application prospect. , Overcome the effect of inaccurate matching relationship
Active Publication Date: 2021-05-11
NANCHANG HANGKONG UNIVERSITY
View PDF5 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
However, in the scenes of local deformation, large displacement motion, motion blur, weak texture, etc., this kind of method is easy to lead to inaccurate pixel matching relationship and too few matching pixel points.
Method used
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment Construction
[0038] The present invention will be further described below in conjunction with drawings and embodiments. see Figure 1a to Figure 6 , a large-displacement motion optical flow calculation method based on depth matching, using the temple_3 sequence image for experimental description:
[0039] 1) input Figure 1a and Figure 1b It is two consecutive frames of images in the temple_3 image sequence; where: Figure 1a is the first frame image, Figure 1b is the second frame image;
[0040] 2) if figure 2 As shown, in order to effectively locate the large displacement motion area in two consecutive frames of images, use the DeepMatch matching algorithm to obtain the matching pixel point set M1 between two consecutive frames of images in the temple_3 image sequence;
[0041] 3) if image 3 As shown, using the grid approximation method, divide two consecutive frames of images into n×n non-overlapping image grids, and then assign the matched pixels to the corresponding image gri...
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More PUM
Login to View More Abstract
The invention discloses a large-displacement motion optical flow calculation method based on depth matching. First, the DeepMatch matching algorithm is used to obtain a large number of matching pixels between two consecutive frames of images, and then the matching pixels are allocated to the corresponding image grids to obtain continuous The image grid matching relationship between two frames of images; then use the structure forest fast edge detection algorithm to obtain image edge information, and use the affine transformation equation to interpolate the reference frame from sparse to dense according to the edge protection distance to obtain the initial dense Corresponding field: Bring the dense corresponding field as the initial value into the energy functional to find the optimal solution, and output the optical flow calculation result. The invention significantly improves the optical flow calculation accuracy and effectively protects the moving edge, overcomes the problems of inaccurate pixel point matching relationship and too few matching pixel points, and has higher calculation efficiency for non-rigid large displacement scenes and complex edge image sequences precision and better applicability.
Description
technical field [0001] The invention relates to an image matching optimization technology, in particular to a depth-matching-based optical flow calculation method for large-displacement motion. Background technique [0002] Optical flow calculation is the study of the temporal change and correlation of pixel intensity information in an image sequence to determine the "movement" of each pixel position, that is, to study the relationship between the temporal change of image grayscale and the structure and motion of objects in the scene. The optical flow field not only contains the motion information of the observed object, but also carries information about the three-dimensional structure of the object, so the optical flow field plays an important role in different fields, such as robot vision systems that perform various tasks, resource exploration, meteorological Cloud image analysis and forecasting, automatic analysis systems in military surveys, unmanned vehicles, etc. ...
Claims
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More Application Information
Patent Timeline
Login to View More Patent Type & Authority Patents(China)
IPC IPC(8): G06T7/246G06T7/13
CPCG06T2207/10016G06T7/13G06T7/246
Inventor 陈震张道文张聪炫江少锋尚璇
Owner NANCHANG HANGKONG UNIVERSITY