Differential geometry visual identification method and system for spatial curved surface features

Abstract

The invention discloses a differential geometry visual identification system for spatial curved surface features, and the system comprises a Frenet camera calibration system which comprises a camera nonlinear model based on a movable frame theory, and a camera internal and external parameter solving method based on the relative position relation between a camera and a calibration plate; a curved surface image acquisition system which comprises a binocular stereoscopic vision measurement system consisting of two cameras and is used for acquiring a spatial stereoscopic real object image; and a curved surface characteristic parameter calculation module which is used for calculating curved surface characteristics based on a differential geometry tool and analyzing geometrical properties of a spatial curved surface and a singular curved surface through curvature characteristics. According to the invention, the curved-surface movable orthogonal frame field and the curved-surface basic formula are applied to the field of pattern recognition, so that on one hand, the complexity of camera calibration is reduced on the premise of ensuring the calibration precision; and on the other hand, a new way is provided for analyzing the geometrical characteristics of the curved surface by a machine vision tool.

Description

technical field [0001] The invention relates to the field of visual recognition, in particular to a method and system for differential geometric visual recognition of spatial curved surface features. Background technique [0002] With the development of optical technology and computer technology, the realization of binocular stereo vision is becoming more and more simple and common. Camera calibration and distortion correction are the basis of the visual recognition system, and their accuracy and real-time performance determine the accuracy of the measurement results. From 3D data to geometric properties of target objects is an important method for target surface description and target matching classification, and it has great significance. For a spatial surface, its bending direction, degree, and concave-convex degree are determined by its characteristic parameters such as normal curvature, Gaussian curvature, and average curvature, and these characteristic parameters are ...

AI Technical Summary

Problems solved by technology

The classic Zhang Zhengyou calibration method can accurately calibrate the camera, but the position of the calibration plate needs to be changed at least 3 times during the calibration process. This calibration method of manually changing the position of the calibration plate increases the complexity of visual inspection on the industrial assembly line; At present, through the analysis of the local geometric properties of the surface through three-dimensional data, most of them adopt the method of fitting the surface parameter equation and directly calculating the characteristic parameters of the surface, so as to achieve the purpose of quantitatively analyzing the local characteristics of the surface within the allowable error range.

However, for free-form surfaces and composite surfaces, surface parameter equations are often difficult to obtain or the expression form is too complicated, which leads to complicated calculation process of characteristic parameters.

Images