二阶累积三维模型轮廓块间混淆分离方法  

Separation Method of 3D Model Silhouette Confusion Block Based on Two Order Cumulant

作　　者：蔡百川[1] 梅容芳[1] 梅建国 

机构地区：[1]宜宾职业技术学院电子信息与控制工程系,四川宜宾644003 [2]重庆市卫生技工学校,重庆401320

出　　处：《科技通报》2015年第4期25-27,共3页Bulletin of Science and Technology

摘　　要：在三维模型视景仿真建模中,常因视觉切换和光线强度衰减,产生轮廓块间混淆,需要进行混淆分离算法设计。提出一种基于二阶累积泰勒展开的三维模型轮廓块间混淆分离方法,进行高精度多轮廓三维立体模型视景建模,对于遍历后的建模点进行模型重构,实现三维多轮廓立体模型建模。构建多轮廓三维模型视景建模视点切换运动方程,得到边缘轮廓特征检测结果。提取轮廓块间的混淆像素值二阶累积量特征,采用二阶累积量泰勒展开算法,对三维图像模型的轮廓边缘块间混淆进行分离的改进。仿真结果表明,采用该算法,混淆分离效果较好,三维图像模型的视景仿真效果真实度较高,轮廓信息保真度好,展示了算法的优越性能。The visual simulation modeling in 3D model, often because of visual switching and the light intensity attenuation,generating contour blocks need to be confused, separation algorithm design is necessary. The separation method of a confusion of 3D model contour block expansion two order cumulant based on Taylor＇s room, high precision contour three-dimensional model visual modeling, model reconstruction for modeling point traversing the multi contour, 3D stereo model. The visual modeling viewpoint switching motion equation of 3D model multi contour is constructed, and the edge feature detection is obtained. Two order cumulant features extraction of contour blocks the confusion between the pixels in value, using the two order cumulant Taylor expansion algorithm, to improve the contour edge separation of 3D image model confusion. The simulation results show that, the algorithm has good separation effect of confusion, visual simulation effect has true degrees,it has higher performance in 3D image model construction, contour information fidelity is good, it demonstrates the superior performance of the algorithm.

关 键 词：二阶累积量 泰勒展开 三维模型 混淆分离 

分 类 号：TP399[自动化与计算机技术—计算机应用技术]