A method and device for orb feature extraction and matching
A feature extraction and feature matching technology, applied in image analysis, image enhancement, instruments, etc., can solve the problem of consuming large memory and CPU resources, reduce development difficulty, ensure the accuracy of feature extraction and matching, and improve implementation performance. Effect
Active Publication Date: 2022-01-07
西安因诺航空科技有限公司
View PDF4 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0005] The purpose of the present invention is to provide a method and device for ORB feature extraction and matching, which solves the problem of consuming a large amount of memory and CPU resources in the prior art for identifying the same target image
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
[0052] The present invention is described in further detail below in conjunction with accompanying drawing:
[0053] The present invention provides a method for ORB feature extraction and matching, which can be applied to the SLAM front end. The present invention includes the following steps:
[0054] Step 1: Collect multiple frames of images of the same scene at different times and store them.
[0055] Step 2: Convert multiple frames of original images of the same scene at different times into images containing only grayscale information, perform FAST corner point and HARRIS corner point extraction on each frame of grayscale image after conversion, and extract each frame of grayscale image Calculate the centroid of each pixel in the image, and perform Gaussian blurring on each frame of grayscale image, and then output the key information of feature points of each frame of image and the image after Gaussian blurring. The key information of feature points includes the extracted F...
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 a method and device for ORB feature extraction and matching. Firstly, the original image is converted into a grayscale image, FAST corner point, HARRIS corner point extraction, centroid calculation and Gaussian blur are performed, and key information of feature points and Gaussian blur are output. image. The image is down-sampled to obtain key information of multiple sets of feature points and a Gaussian blurred image. Then calculate the descriptor information corresponding to the feature points, perform feature matching on the descriptor information of the two frames of images, and obtain the ORB feature matching result. Finally, the original image and the ORB feature matching result synchronized with the original image are output. The problem of consuming a large amount of memory and CPU resources in the prior art for identifying the same target image is solved. The invention can increase the speed of recognizing the same target image, and guarantees real-time performance under the premise of ensuring the accuracy of feature extraction and matching.
Description
technical field [0001] The invention belongs to the technical field of image registration, in particular to an ORB feature extraction and matching method and device. Background technique [0002] Image registration is the process of matching and superimposing two or more images acquired by imaging devices at different times, angles or environments. In the fields of target detection, map reconstruction, etc., in order to obtain the matching relationship between different images, a large number of real-time calculations need to be performed on the images captured by the camera, and the feature points with high recognition are smaller than the number of pixels. The order of magnitude, the calculation speed is faster, and a more accurate matching relationship between different images can be obtained. [0003] At different times, angles and environments, the same target image has a certain correlation, but because of the shooting angle, scale, brightness and other issues, it is ...
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): G06T7/33
CPCG06T7/33G06T2207/20016G06V10/44
Inventor 王黎
Owner 西安因诺航空科技有限公司