基于空间六点不变量的道路立体目标识别方法  

Road Stereo Target Recognition Method Based on Space Six Point Invariants

在线阅读下载全文

作  者:张雅茜 韩斌[1] ZHANG Yaqian;HAN Bin(School of Computer,Jiangsu University of Science and Technology,Zhenjiang 212003)

机构地区:[1]江苏科技大学计算机学院,镇江212003

出  处:《计算机与数字工程》2022年第2期300-304,共5页Computer & Digital Engineering

摘  要:城市道路中常设置具有3D效果的平面路障或标志物,其具有高度的立体性和真实性,导致行人和辅助驾驶系统误判而造成严重事故,因此需要对道路立体目标进行识别,以获得真实路面情况。常见的射影不变量如交比是基于共面五点计算的,存在局限性,论文提出一种基于空间点元素的几何不变量计算方法,把空间元素的共点和共线用具有物理意义的量来表示,通过合理搭建把空间元素巧妙转化为共面关系。该几何不变量只依赖点的空间坐标,与投影视角、摄像机参数等无关。在三维造型软件SolidWorks上进行模拟实验,并用该计算方法对真实道路上的真假3D物体进行了测验,最终实测实验验证依据空间六点计算得到的几何量可用于道路立体目标的识别。In urban roads,plane barriers or signs with 3 D effect are often set up,which have a high degree of stereoscopic and authenticity,resulting in serious accidents caused by misjudgment of pedestrians and auxiliary driving system. Therefore,it is necessary to identify the stereoscopic objects of the road to obtain the real road conditions. Common projective invariants such as cross ratio are calculated based on coplanar five points,which has limitations. In this paper,a geometric invariants calculation method based on spatial point elements is proposed. The coplanarity and collinearity of spatial elements are represented by quantities of physical significance,and the spatial elements are cleverly transformed into coplanar relations by reasonable construction. The geometric invariants only depend on the spatial coordinates of the points and are independent of the projection angle and camera parameters. The simulation experiment is carried out on SolidWorks,which is a three-dimensional modeling software. The calculation method is used to test the real and false 3 D objects on the real road. Finally,the actual measurement experiment verifies that the geometric quantity calculated according to the space six points can be used for the recognition of the three-dimensional objects on the road.

关 键 词:射影不变性 几何不变量 交比 空间点元素 

分 类 号:TP391.4[自动化与计算机技术—计算机应用技术]

 

参考文献:

正在载入数据...

 

二级参考文献:

正在载入数据...

 

耦合文献:

正在载入数据...

 

引证文献:

正在载入数据...

 

二级引证文献:

正在载入数据...

 

同被引文献:

正在载入数据...

 

相关期刊文献:

正在载入数据...

相关的主题
相关的作者对象
相关的机构对象