Accelerated structure from motion method based on GPS auxiliary information
A technology of auxiliary information and structural methods, applied in image data processing, instruments, 3D modeling, etc., can solve time-consuming and complex problems, and achieve the effect of accelerated process, easy calculation, and simple method
Active Publication Date: 2017-12-08
QINGDAO RES INST OF BEIHANG UNIV +1
View PDF8 Cites 3 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Some scholars have proposed image matching methods based on graphs and image matching methods based on vocabulary trees. Although they can guide the image matching process, the entire processing process was relatively complicated and time-consuming.
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
[0025] The present invention will be further described below in conjunction with the accompanying drawings and specific embodiments.
[0026] figure 1 It is a flow chart of the present invention based on the acceleration of GPS auxiliary information to recover the structure method from motion; the factual process of the present invention is mainly divided into the following steps:
[0027]Step 1: numbering the input image and extracting the GPS position information in the input image at the same time,
[0028] (1.1) First calculate a threshold d,
[0029]
[0030] (1.2) Then calculate the image pair to be matched;
[0031] The image i, j satisfying the following conditions is the image pair(i, j) to be matched;
[0032] (wherein, the value of N is taken as 5);
[0033] Step 2: Feature extraction and feature matching of images
[0034] Extract the SIFT features of all the images, and use the cascade hash method to calculate the feature matching for the above-mentioned...
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 relates to an accelerated structure from motion method based on GPS auxiliary information. The method comprises the steps that step one, input images are numbered, GPS position information in the input images is extracted and image pairs to be matched are computed; step two, feature extraction is performed on the images and feature matching is performed between the image pairs to be matched determined in the step one, and the matched result is used for three-dimensional point cloud computing of the step four and camera attitude estimation of the step three; step three, further filtering is performed on the feature matching result of the step two so as to be used for computing the camera attitude; and step four, the three-dimensional point of the target scene is computed by using the matching result of the step two and the camera attitude of the step three. The image pairs to be matched are determined by using the GPS position information so that redundant computation of the image matching phase can be avoided and the whole reconstruction process can be accelerated, and the acceleration effect is more obvious when the number of the images is high.
Description
technical field [0001] The present invention belongs to the field of image modeling, virtual reality technology, computer vision and computer graphics and image processing, and specifically uses the two-dimensional pixel information of the image of the target scene to recover the three-dimensional image of the scene through theories such as multi-viewpoint geometry and image matching. The structure, including optimizing the shooting position and orientation of the camera, feature point tracking, and the calculation and optimization of the 3D point cloud of the scene, is mainly used for the automatic reconstruction of large outdoor scenes. Background technique [0002] In computer vision, the 3D reconstruction theory of multi-view geometry has been developed relatively maturely, and the relevant basic theories and algorithms are introduced in detail in the document Multiple View Geometry in Computer Vision 2nd. Restoring structure from motion (Structure from Motion, SFM) has ...
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 Applications(China)
IPC IPC(8): G06T7/73G06T17/00
CPCG06T17/00G06T2207/10028G06T7/73
Inventor 齐越杜文祥
Owner QINGDAO RES INST OF BEIHANG UNIV