A Visual Detection Method of Spatial Arc Pose

Abstract

The invention relates to a visual inspection method for spatial arc positions and postures. A spatial arc is mapped to a virtual imaging plane gamma 1 of a virtual camera {C1} to build a mathematical conversion model between an actual image Ima of the spatial arc in an actual camera coordinate system {C} and a virtual image I'ma mapped in the virtual camera {C1}. The visual inspection method has high convergence, radial distortion correction can be performed before feature sampling points in the images are mapped, mapping transformation errors are decreased, detection precision is improved, three-dimensional matching links are omitted, the measuring method is simple and reasonable, and required hardware cost is low.

Description

technical field [0001] The invention relates to a method for detecting the pose of a space arc, in particular to a method for visually detecting the pose of the arc in space when the radius of the space arc is known. Background technique [0002] Holes and planes are the basic features of mechanical parts. Holes and planes are orthogonal to form a spatial geometric circle, and a spatial geometric circle is a shape with obvious characteristics and easy identification. It has incomparable advantages in image processing such as straight lines. advantage. Through the visual pose detection of the spatial geometric circle on a certain plane of mechanical parts, the spatial position and attitude of mechanical parts can be obtained indirectly, which has a wide range of applications in robot visual positioning, target tracking, and visual obstacle avoidance. For the visual pose detection of spatial geometric circles, many scholars have proposed many effective methods, such as litera...

AI Technical Summary

Problems solved by technology

For the visual pose detection of spatial geometric circles, many scholars have proposed many effective methods, such as literature 1 [HEIKKIL AJ, SILVENO.A four-step camera calibration procedure with implicit image correction[C].IEEE Computer Society Conference on Computer Vision and Pattern Recognition,1997:1106-1112] proposed to solve the space circle pose information by setting a circle with multiple lattice distributions in space and coupling multiple camera parameters together. This method has a large amount of calculation and is difficult to quickly Obtain the exact position of the center point of a single circle; Document 2 [Zhou Fuqiang, Zhang Guangjun, Jiang Jie, etc. Non-contact high-precision measurement method of spatial circle geometric parameters [J]. Journal of Instrumentation, 2004, 25(5): 604-607 】Proposed a method to obtain the corresponding matching point of the space circle in the binocular stereo vision according to the epipolar constraints, and then project it into the three-dimensional space to obtain the actual coordinates of the edge

Although the stereo matching algorithm has been extensively studied by many scholars and many effective measures have been proposed, the stereo matching algorithm itself is an ill-conditioned algorithm. Generally, it needs to establish an energy function, minimize the energy sum function and some constraints, and adopt the optimization theory method to solve the equation, the calculation process is more complicated; literature 3 [KIM J S, GURDJOS P. Geometric and algebraic constraints of projected concentric circles and their applications to camera calibration [J]. IEEE Trans Pattern Analysis and Machine Intelligence, 2005, 27 (4) :637-642.] It is proposed to take concentric circles as the target, use concentric circles to solve the straight line at infinity, and then find the exact center position of the circle, but combining two disjoint quadratic curves to solve the straight line at infinity increases the complexity of the solution and the amount of calculation; Document 4 [Xing Dekui, Da Feipeng, Zhang Hu. Research and application of circular target precision positioning method [J]. Instrumentation Journal. 2009,30(12):2593-2598] for concentric circular targets, A method to quickly and accurately obtain the true projection point of the center of the circular target in the image is proposed by using the invariance of the cross ratio

Images