Simple tree dynamic programming binocular and stereo matching method based on invariant moment spatial information
A binocular stereo matching and dynamic programming technology, applied in the field of image processing, can solve problems such as unfavorable practical application, high mismatch rate, and complex algorithm, and achieve the effects of improving the quality of disparity maps, improving matching accuracy, and suppressing uneven illumination
Inactive Publication Date: 2014-04-09
NANJING UNIV OF SCI & TECH
View PDF6 Cites 8 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0003] Document 1 (Yanshan University. Liu Yingjie. Research on Stereo Matching Algorithm Based on Dynamic Programming and Belief Propagation [D]. Dissertation. 2011), proposed a global algorithm based on dynamic programming algorithm and belief propagation algorithm for matching, although the effect is good, However, the algorithm is complicated and the real-time performance is poor, which is not conducive to practical application.
Document 2 (Dalian University of Technology. Zhou Fang. Research and Implementation of Stereo Matching Algorithm in Binocular Vision [D]. Dissertation. 2013) proposed a method of combining DP algorithm with local linear filtering technology, which also exists The algorithm is complex, the real-time performance is poor, and the mismatch rate is high in the same depth area
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
[0020] Such as figure 1 As shown, the binocular stereo matching method of simple tree dynamic programming based on invariant moment space information includes the following steps:
[0021] Step 1. Use the first-order invariant moment formed by the second-order moment to convert the binocular grayscale image I captured by the camera to 1 and I 2 Convert to moment invariant space image L tex and R tex , the specific process is, respectively, with the grayscale image I 1 and I 2 The pixel to be transformed is the center, and the transformation window with a size of M×N traverses the grayscale image I 1 and I 2 For each pixel of , the first-order invariant moment Φ calculated in each transformation window 1 Instead of a grayscale image I 1 and I 2 The gray value of the pixel to be transformed, to obtain the moment invariant space image L tex and R tex ,
[0022] The first order invariant moment Φ 1 The calculation process is:
[0023] 1.1 Calculation in two dimension...
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 simple tree dynamic programming binocular and stereo matching method based on invariant moment spatial information. According to the method, a first invariant moment composed of a secondary moment is adopted for converting a binocular gray level image shot by a camera into an invariant moment spatial image; a three-dimensional matrix composed of the matching difference value of each parallax error serves as an initial parallax error matrix in a simple tree dynamic programming algorithm; a horizontal tree price energy function matrix and a vertical tree price energy function matrix are calculated through simple tree price energy function iteration; when the value of a price energy function is smallest, the corresponding parallax error is the accurate parallax error, and then a parallax graph is obtained. The simple tree dynamic programming binocular and stereo matching method based on the invariant moment spatial information effectively improves the quality of parallax graphs of areas affected by uneven illumination or disturbed by other noise, the areas of the same depth are smoother, and the computing speed of the algorithm is improved.
Description
technical field [0001] The invention belongs to the technical field of image processing, and in particular relates to a binocular stereo matching method based on simple tree dynamic programming of invariant moment space information. Background technique [0002] The binocular stereo matching algorithms used today are mainly divided into two categories: region-based algorithms and global-based algorithms. The region-based algorithm is simple and time-sensitive, but has a high error matching rate for discontinuous regions of the image. The global-based algorithm can solve the matching problem in uncertain regions, effectively solve the mismatching points, and improve the correct matching rate. [0003] Document 1 (Yanshan University. Liu Yingjie. Research on Stereo Matching Algorithm Based on Dynamic Programming and Belief Propagation [D]. Dissertation. 2011), proposes a global algorithm for matching based on dynamic programming algorithm and belief propagation algorithm. Alt...
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 Applications(China)
IPC IPC(8): G06T7/00
Inventor 张毅柏连发庞星韩静岳江金左轮祁伟赵壮
Owner NANJING UNIV OF SCI & TECH