A Dense Stereo Matching Method for Fisheye Images Based on Polar Curve Geometry
A fisheye image and stereo matching technology, applied in the field of image processing, can solve the problems of low matching accuracy, complexity, loss of detail information, etc., and achieve the effect of improving matching accuracy and fast image processing.
Active Publication Date: 2019-02-01
YANSHAN UNIV
View PDF3 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
But the problem is that the fisheye image is seriously deformed, and its correction itself is a complicated problem. The iterative processing in the correction process is likely to cause the loss of detail information, which affects the effect of subsequent processing, and the existing fisheye image matching methods still have low matching accuracy. , slow matching speed and other shortcomings
In summary, the existing fisheye image processing technology cannot meet people's needs
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
[0022] The present invention will be further described below in conjunction with accompanying drawing:
[0023] Such as figure 1 Shown, the step of algorithm described in the present invention is as follows:
[0024] Step 1, deduce the polar curve analytical formula of the fisheye stereo system according to the fisheye camera imaging model;
[0025] First, the fisheye camera model is optimized by adding distortion, and then the optimized camera model is analyzed to obtain the epipolar equation.
[0026] Step 2, standardize the search path of corresponding points based on the polar curve analytical formula and select the correct window neighborhood to calculate the initial matching cost;
[0027] Use the polar curve to standardize the search range of corresponding points, divide the whole image into several parts according to the polar curve; calculate the initial matching cost based on the area, and take the point to be matched as the center, and select a certain number of β...
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 provides a fish-eye image dense stereo algorithm based on polar curve geometry. The method comprises the steps of: firstly, deriving out a polar curve analysis formula of a fish-eye stereo system according to a fish-eye camera imaging model; then based on the polar curve analysis formula, standardizing searching paths of corresponding points and selecting a correct window neighborhood to calculate initial matching cost; and finally utilizing a dynamic programming algorithm to optimize the initial matching cost and obtaining an initial parallax image of the fish-eye image. The searching ranges of the corresponding points are standardized by the algorithm, and the algorithm has the advantages that the matching speed is high, the matching precision of the fish-eye image is high, etc.
Description
technical field [0001] The invention relates to the technical field of image processing, in particular to a fisheye image dense stereo method based on polar curve geometry. Background technique [0002] A fisheye lens is an extremely wide-angle lens with a relatively short focal length, and its viewing angle is close to or equal to 180 degrees. One image can capture the entire hemisphere or even a wider field of view at one time. Although the large field of view imaging system composed of multiple cameras also has the above functions, it has problems such as high image processing complexity and imaging delay of the omnidirectional imaging system based on the rotating pan / tilt. Therefore, compared with a large field of view imaging system composed of multiple cameras, fisheye image processing has the advantages of simplicity and real-time performance, and is widely used in many fields that require a large field of view. For example: 3D reconstruction of urban environment, re...
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): G06T7/30
Inventor 李海滨宋涛贾璐熊文莉李雅倩侯培国
Owner YANSHAN UNIV