Research On Multi-target Surgical Positioning System Based On Stereo Vision

The robot-assisted surgical navigation system is of great significance to improve the success rate of surgery,the surgical incision of complications,the improvement of surgical methods and the accuracy of predictive positioning.Optical navigation systems are most widely used in surgical navigation systems because of their high tracking accuracy,large visual range,and suitable for coordinate matching conversion.They are usually composed of three parts: a visual positioning module,a visualization module and a surgical positioning instrument.At present,the mainstream optical positioning products on the market are mostly binocular.When a camera is blocked,the optical positioning system will lose the target.In order to reduce the possibility of light blocking,this article builds a multi-eye near infrared based on stereo vision Surgical positioning system.The main research contents of this article include:Construction of multi-view visual positioning system.Based on edge detection and centroid method,multi thread technology is used to improve the refresh frequency of optical positioning system.The registration,import and matching of any surgical tool can be realized by this system.At the same time,it provides an experimental platform for error analysis and temperature compensation.Research on error of multi-eye vision positioning system.Analyze the impact of the various structural parameters of the system on the positioning accuracy,and reduce errors on the hardware platform through reasonable layout,while meeting the needs of the work space.Due to the existence and anisotropy of the system error,there will be a point jumping phenomenon at the same landmark point after some cameras are blocked.This paper uses the neural network nonlinear fitting method to predict the image coordinates of the blocked point to improve the stability of the system.Research on the influence mechanism of temperature on the error of multieye vision positioning system.The change of camera temperature will cause the deviation of the imaging plane and the deformation of the body,which will have a huge impact on the multi-eye vision surgical positioning system.By studying the influence mechanism of temperature on the multi-eye vision positioning system,nonlinear compensation is performed to improve the accuracy of the system under the responsible environment.Finally,in response to the problems and methods proposed in this paper,the stability and accuracy of the optical positioning system are experimentally verified and quantitatively analyzed in a real environment.The results show that the multi vision positioning system has high accuracy,has certain anti occlusion ability,and the preheating time of the system is shorter.