A Matching Method for Straight Line Segment of Close-range Image
A matching method and technology of straight line segments, applied in image analysis, image enhancement, graphics and image conversion, etc., can solve problems such as weak consistency and difficulty in obtaining matching results
Active Publication Date: 2020-02-14
LIAONING TECHNICAL UNIVERSITY
View PDF4 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
However, in the matching process, the parallel support domains of straight segments used to construct the target straight segment on the reference image and the candidate straight segment descriptor on the search image are directly centered on the two. For images with scale changes or perspective changes, the consistency of information contained in the two linear support domains is weak, so it is difficult for this descriptor to obtain reliable matching results for images with scale changes or perspective changes
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
[0067] An embodiment of the present invention will be described in detail below in conjunction with the accompanying drawings.
[0068] Such as figure 1 As shown, a kind of close-range image line segment matching method of the present invention comprises the following steps:
[0069] Step 1: Input the two close-range images of the reference image and the search image and the coordinates of the same point of the two close-range images;
[0070] The set of points with the same name is in, and are the coordinates of the same-name point of the reference image and the search image respectively, k is the index number of the same-name point, k=1, 2, ..., K, and K is the number of the same-name point;
[0071] Step 2: Extract straight line segments from the reference image and the search image respectively;
[0072] Use the straight line extraction algorithm to extract straight line segments from the reference image and the search image respectively, and remove the lines whose...
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
A method for matching straight line segments of close-range images, belonging to the technical field of close-range photogrammetry and computer vision; comprising: extraction of straight line segments; determination of a set of candidate straight line segments for target straight line segments; establishment of parallel support domains for target straight line segments, and based on the target straight line segment The parallel support domain establishes the plane support domain of the candidate straight line segment; calculates the Euclidean distance between the feature descriptor of the target straight segment and the feature descriptor of the candidate straight line segment; determines the straight line segment of the same name of the target straight segment according to the Euclidean distance, and obtains A collection of straight line pairs with the same name; the present invention uses epipolar constraints to determine the parallel support domains of candidate straight line segments, and constructs descriptors on this basis to enhance the effectiveness of descriptors; perform regional affine transformation on the parallel support domains of target straight line segments and candidate straight line segments , to achieve reliable matching of straight-line segments of images of different scales, improve the limitations of existing descriptors, and apply to the matching of straight-line segments of different types of close-range images such as translation, rotation, blur, and illumination.
Description
technical field [0001] The invention belongs to the technical field of close-range photogrammetry and computer vision, and in particular relates to a method for matching straight line segments of close-range images. Background technique [0002] Image matching refers to using a certain matching algorithm to find features with the same name on two or more images covering the same area or the same feature, and to establish the corresponding relationship between features with the same name on different images. In man-made target scenes, straight line features contain richer semantic and structural information. Therefore, using the straight line information on the image to reconstruct the 3D model of artificial ground objects has important research significance in the fields of close-range photogrammetry and computer vision. is its key core technology. [0003] At present, scholars at home and abroad have proposed many straight line matching methods. The research focus of exis...
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): G06T3/00G06T7/30
CPCG06T2207/10004G06T3/147
Inventor 王竞雪张雪熊俊
Owner LIAONING TECHNICAL UNIVERSITY