An Image Feature Point Matching Method Based on Hash Reconstruction
An image feature point and matching method technology, which is applied in the field of image feature point matching based on hash reconstruction, can solve the problems of low matching accuracy, large amount of calculation, and high matching complexity, so as to improve matching accuracy, reduce cost, and quickly The effect of feature point matching
Active Publication Date: 2019-07-19
北京小白世纪网络科技有限公司
View PDF5 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0003] One of the purposes of the present invention is to provide a method for matching image feature points based on hash reconstruction, so as to solve the problems of high matching complexity, large amount of calculation or low matching accuracy in existing image feature point matching methods
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
[0057] A kind of image feature point matching method based on hash reconstruction provided by the present invention specifically comprises the following steps:
[0058] (1) learn a hash function for the scale-invariant feature transform (SIFT) feature learning of the checked feature points of the checked image by iterative quantization (Iterative Quantization, ITQ) hashing method, specifically including the following steps:
[0059](1.1) Carry out PCA dimensionality reduction to the feature matrix X of the feature point of the checked image, obtain PCA dimensionality reduction matrix T;
[0060] (1.2) Utilize the PCA dimensionality reduction matrix T obtained in the previous step to obtain the characteristic matrix V after PCA dimensionality reduction is carried out to the characteristic matrix X of the feature points checked, V=TX;
[0061] (1.3) Build a hash function, which is:
[0062]
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 an image feature point matching method based on hash reconstruction. Points and checked feature points are screened for candidate feature points by distance measurement; hash reconstruction is performed on candidate feature points and query feature points; similarity comparisons are performed on candidate feature points and query feature points after hash reconstruction, and finally matching features are obtained point. An image feature point matching method based on hash reconstruction provided by the present invention improves the matching accuracy through a two-step matching method; in the process of matching the reconstructed features obtained through hash reconstruction, the original features do not need to be stored, reducing The cost of storage and indexing is reduced, so as to achieve fast feature point matching.
Description
technical field [0001] The invention belongs to the technical field of computer multimedia, in particular to an image feature point matching method based on hash reconstruction. Background technique [0002] Image description methods based on image local features (such as SIFT features) are widely used in image retrieval. In order to improve the retrieval accuracy, it is necessary to directly and quickly match the feature points. However, the current feature point matching methods are in terms of matching accuracy and matching speed. It's hard to find a better balance. Existing image feature point matching methods generally have the problems of high matching complexity, large amount of calculation, or low matching accuracy. Contents of the invention [0003] One of the objectives of the present invention is to provide an image feature point matching method based on hash reconstruction to solve the problems of high matching complexity, large amount of calculation or low ma...
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): G06F16/583G06K9/46G06K9/62
CPCG06F16/583G06V10/462G06F18/22
Inventor 邢爽杜强聂方兴
Owner 北京小白世纪网络科技有限公司