Flight data-based unmanned aerial vehicle non-measurement type camera calibration method
A technology of flight data and drones, applied in the field of computer programs, can solve problems such as multi-manual intervention and complex experimental environments, and achieve the effects of ensuring accuracy, high degree of automation, and improving stability
Active Publication Date: 2017-06-20
深圳飞马机器人科技有限公司 +1
View PDF2 Cites 5 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0004] In order to solve the above-mentioned technical problems, the present invention provides a non-quantitative unmanned aerial vehicle based on flight data for the problem that the existing unmanned aerial vehicle non-measurement camera calibration method requires a relatively complicated experimental environment and requires more manual intervention. Calibration method of measuring camera to achieve the purpose of improving the degree of automation of calibration and simplifying the calibration process
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
[0011] The technical content, structural features, achieved goals and effects of the present invention will be described in detail below in conjunction with the accompanying drawings, and the embodiments will be described in detail below with reference to the accompanying drawings.
[0012] This embodiment provides a UAV non-measurement camera calibration method based on flight data. The method uses the Speeded Up Robust Features (SURF) feature extraction algorithm to calculate the feature points and features of all images captured by UAV aerial photography. Descriptor; according to the airborne GPS data, build a Delaunay triangulation network to obtain the topological relationship between the images and the graph distance; for images whose graph distance is less than 4, calculate the distance between two feature descriptors, and perform Feature point matching; use the Random Sample Consensus (RANSAC) algorithm to calculate the fundamental matrix between the images, and at the ...
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 flight data-based unmanned aerial vehicle non-measurement type camera calibration method. The method is applied to unmanned aerial vehicle non-measurement type camera calibration. The method includes the following steps that: the feature points and feature descriptors of all images obtained through the aerial photography of an unmanned aerial vehicle are calculated; feature point matching is performed according to airborne GPS data; wrongly-matched feature points are eliminated; connection strength between the images is calculated; connected graphs with the largest strength value and connected graphs with the second largest value are found out, and the two kinds of connected graphs are adopted as candidate calibration images and verification images respectively; and a self-calibration method is adopted calculate the calibration parameters of a camera by using the machined feature points of the candidate calibration images, and the verification images are adopted to verify the rationality of the calibration parameters. According to the problems of requirements for complicated ground test fields or equipment and low automation degree of an existing calibration method, the invention provides the novel method applied to unmanned aerial vehicle non-measurement type camera calibration. With the method adopted, calibration accuracy can be ensured, and the automation level of a calibration process can be greatly improved.
Description
technical field [0001] The invention belongs to the field of computer programs, and in particular relates to a non-measurement camera calibration method for an unmanned aerial vehicle based on flight data. Background technique [0002] The low-altitude remote sensing system based on the drone is not limited by the site, and the data acquisition is real-time and efficient. It can quickly obtain geospatial data from multiple angles. It has been widely used in large-scale topographic map surveying, land and ecological environment surveys, dynamic monitoring and Evaluation, digital city and major engineering construction and other fields. Non-measuring cameras have been widely used in UAVs because of their low price, small size, light weight, and flexible use. However, non-measuring cameras are not specially designed for photogrammetry after all, and there are defects such as large lens distortion and unstable inner orientation elements. The traditional non-measurement camera ...
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 IPC IPC(8): G06T7/80G06T7/33G06T3/00
CPCG06T3/00G06T2207/10004G06T2207/30244
Inventor 高广支晓栋徐斌王邦松高宁
Owner 深圳飞马机器人科技有限公司