车载3D全景模型及纹理映射算法  

3D model and texture mapping algorithm for vehicle-mounted panorama system

作　　者：刘志[1] 张常年[1] 黄冬[1] 

机构地区：[1]北方工业大学电子信息工程学院,北京100144

出　　处：《计算机工程与设计》2017年第1期172-176,共5页Computer Engineering and Design

基　　金：北京市教委面上基金项目(KM201610009003)

摘　　要：为解决典型的车载全景辅助驾驶系统只能呈现车身俯视角度的平面图像,存在视角单一、视野范围受限及操作性较差等缺点,研究如何利用平面图像构造具有良好3D视觉效果的全景图,提出一种简单高效的网状三维模型以及模型的纹理映射算法。网状三维模型由底面、球形连接面、环形柱面3部分组合构成,利用基于面积等比映射原理的改进算法实现纹理映射。实验结果表明,该模型可以较好地呈现车辆四周的3D全景,实现多角度的视角切换。The traditional vehicle-mounted panoramic system can only present a planar panoramic image at a top-down view point, which has some disadvantages, such as only one view point supported, limited scope, poor maneuverability and so on. To solve this problem, by mapping a planar panorama image to a 3D one, a simple and efficient three-dimension model for content presen- ting was proposed. The proposed model was composed of a planar undersurface, a spherical junction-surface and an annular cylindrical surface. A texture mapping algorithm was designed for the model based on the principle of equal-area. Experimental results show the constructed 3D panorama has better visual effects than the traditional planar one. It supports the function of arbitrary viewpoint switching.

关 键 词：3D全景模型 纹理映射 视觉盲区 车载全景 像素缓存 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]