为了实现精确放疗过程中精确定位和肿瘤位置的精确测量,建立了基于双目视觉的医疗机器人摆位测量系统,并对系统所采用的摄像机标定方法、标记点识别及其三维坐标计算、摆位系统的位置验证等算法进行研究。建立基于双目视觉的医疗机器人摆位测量系统;提出了一种对摄像机采用基于平面棋盘格和立体标定模板相结合的摄像机标定新方法;采用Roberts梯度算子对图像分割的方法,识别标记点中心并计算其三维坐标;通过比较基于双目视觉计算和三维坐标测量仪器测量的各个标记点三维坐标的摆位误差,实现放疗过程中位置验证和精确摆位。实验结果表明:摄像机的标定精度为36.5×10-3mm和各个标记点的三维坐标平均偏差为δX=O.573mm、δY=0.495mm、δZ=0.430mm,测量方法可获得较高的标定精度和摆位精度,能满足精确放疗对高精度摆位系统的临床需求。
In order to realize the precise positioning of radiation patients and the accurate measurement of tumor location in the process of precise radiation therapy, a medical robot positioning system based on binocular vision is established. First, the medical robot positioning measurement system is developed based on binocular vision; Secondly, based on the Tsai camera model and the influence of lens distortion, a new camera calibration method based on planar and three-dimensional checkerboard calibration tem- plates is proposed to get the camera calibration parameters; Thirdly, the center point positions of mark- ers are identified and their three-dimensional coordinates are calculated based on the robert gradient op- erator method for image segmentation. Finally, the measurement errors of all markers by binocular vision and the three-dimensional coordinate measuring instrument are used to achieve the position-validating for radiotherapy patients. The experimental results show that the camera calibration accuracy is 0. 036 5 mm and three-dimensional coordinates' average deviations of each marker are δX=O.573mm、δY=0.495mm、δZ=0.430mm, The measurement method can improve the accuracy of calibration and positioning, and meet the clinical requirements for high-precision positioning system in precise radiotherapy.