A Method of Removing Mismatch Based on Angle Constraint
A deletion method and mismatching technology, applied in image data processing, instruments, calculations, etc., can solve problems such as high algorithm complexity and real-time performance degradation, and achieve good and fast results
Inactive Publication Date: 2016-03-30
BEIJING INSTITUTE OF TECHNOLOGYGY
View PDF3 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
However, this method requires a large number of inversion operations and a large number of matrix multiplication and square root operations, resulting in high algorithm complexity and reduced real-time performance
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
[0021] Attached below figure 2 Examples are also given to describe the present invention in detail.
[0022] The present invention is applicable to a registration method based on feature points, which is used to address possible mismatches after extracting feature points, performing feature point invariant feature description, determining similarity measurement criteria, and establishing preliminary matching point pairs. Feature point pairs, a method to delete wrong matching point pairs, avoid deviations in registration results, and improve the registration rate of feature points. This method improves the accuracy of corresponding feature point detection, estimates the transformation matrix faster and more accurately, and finally improves the accuracy and real-time performance of registration.
[0023] Such as figure 2 As shown, the method for deleting mismatches based on angle constraints in the present invention includes the following four steps: (1) finding the tan valu...
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 mismatching deleting method based on angle constraint. The method comprises the concrete process including the following steps that 1, the tan value of the coordinate angle of each matching point pair is solved by aiming at n matching point pairs; 2, the n matching point pairs are sequenced in the sequence of the tan values from large to small or from small to large; 3, the tan value of the i-th matching point pair and the tan value of the (i+1)th matching point pair are sequentially compared, when the difference between the two values is within the thresh, the two values are classified into one group, and m groups are totally obtained, wherein i is 1, 2 to (n-1), and the thresh is the preset threshold value; 4, the group k with the most members in the group in the m groups is found, and the matching point pairs beyond the group k are deleted. Compared with an RANSAC (random sample consensus) algorithm and the Mahalanobis distance algorithm, the complicated high-memory occupation operation such as matrix inversion and matrix multiplication is avoided, and the registration accuracy and the real-time performance are improved.
Description
technical field [0001] The invention belongs to the technical field of image registration, and in particular relates to an angle constraint-based error matching deletion method. Background technique [0002] The application of image registration is very extensive, such as pattern recognition, automatic navigation, medical diagnosis, computer vision and other fields. At present, a lot of research work has been carried out in the aspect of image registration, and many image registration methods have been proposed. [0003] Common registration methods are divided into two categories: feature-based registration methods, such as Harris corner method, SIFT method, etc.; region-based registration methods, such as mutual information, FMT, etc. Among them, the region-based registration method does not need to extract features, and is suitable for situations where the grayscale information is greater than the structural information; and the grayscale functions of the two images must ...
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): G06T7/00
Inventor 陈禾章学静马龙谢宜壮曾涛龙腾
Owner BEIJING INSTITUTE OF TECHNOLOGYGY