Binary Descriptor Construction Method Based on Simple Sampling Method and Three-valued Strategy
A binary description and construction method technology, applied in character and pattern recognition, instruments, calculations, etc., can solve problems such as unstable performance and poor sub-description ability, achieve accurate and stable methods, improve matching performance, and overcome instability Effect
Active Publication Date: 2019-05-17
HENAN POLYTECHNIC UNIV
View PDF5 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0011] Aiming at the shortcomings of the existing image binary descriptors such as poor description ability and unstable performance, the present invention proposes a binary descriptor construction method based on a simple sampling method and a three-valued strategy, which mainly includes the following steps:
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] Such as figure 1 Shown is the flow chart of the binary descriptor construction method based on the simple sampling method and the ternary strategy of the present invention, which mainly includes the following steps: collecting two images of the same scene from different perspectives and inputting them to the computer, converting the color image into a grayscale image And perform Gaussian smoothing, use the Harris corner detection algorithm to extract feature points on the two images, obtain the point pair sampling mode, calculate the main direction of the feature point, obtain the sampling point pair in the feature point sampling area, and perform the sampling point Smoothing, selecting 256 sampling point pairs from 400 sampling point pairs, using 256 sampling point pairs to construct a binary descriptor for each feature point, and matching feature points based on the binary descriptor. The specific implementation details of each step are as follows:
[0023] Step S1: Col...
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 binary descriptor construction method based on the simple sampling method and the ternary strategy of the present invention includes: collecting two images from different viewing angles of the same scene and inputting them into the computer, converting the color image into a grayscale image and performing Gaussian smoothing processing, using Harris The corner detection algorithm extracts feature points on the two images, calculates the main direction of the feature points, obtains the sampling point pairs in the feature point sampling area and smooths the sampling points, and selects 256 sets of sampling points from 400 sets of sampling point pairs Right, use 256 sets of sampling point pairs to construct a binary descriptor for each feature point, and perform feature point matching based on the binary descriptor. The method provided by the invention is simple in calculation, has better matching performance, and is suitable for image feature matching tasks of mobile terminal equipment.
Description
Technical field [0001] The invention relates to the field of feature point matching in image processing, in particular to the construction of binary descriptors in digital images and a method for feature point matching. Background technique [0002] Feature matching is an important issue in the field of image processing and computer vision. Feature point matching technology is widely used in many scenes such as target recognition, target tracking, and scene stitching. The basic principle of image feature description and matching is to select a local area centered on a feature point and construct a matching descriptor according to the texture information in the area to perform feature matching. The mainstream feature point matching technology is based on floating-point descriptors. Representative floating-point matching descriptors include SIFT[1], SURF[2] and DAISY[3]. With the application of mobile smart devices, binary descriptors with small storage space and high processing e...
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): G06K9/62
CPCG06V10/757
Inventor 王志衡李璐刘艳李广武刘红敏霍占强贾利琴姜国权王静
Owner HENAN POLYTECHNIC UNIV