摘要
针对基于机器视觉的室内环境感官设计模型的剔除精度低问题,提出基于三维视觉的室内环境感官设计建模研究。先重构室内环境三维视觉空间点,再运用欧氏距离计算空间点之间的矢量关系,实现三维视觉空间点匹配。完成上述操作后,运用ElasticFusion算法构建室内环境感官设计模型。由此,完成基于三维视觉的室内环境感官设计建模。最后进入实验部分,分别测试两个模型的剔除精度,在数据库中挑选同一场景的10个空间点,运用两个模型分别计算这10个空间点的深度值。实验结果显示,所提模型求取的10个空间点深度值均为0,符合剔除要求;而运用基于机器视觉的室内环境感知设计模型求取的七个空间点深度值有7个均大于0,不符合剔除要求。由此可知,所提模型剔除精度高。
In allusion to the problem of the low elimination precision of the indoor environment sensory design model based on 3D vision,an indoor environment sensory design modeling based on 3D(three dimensions)vision is proposed.The 3D visual spatial points of indoor environment are reconstructed,and then the Euclidean distance is used to calculate the vector relationship between spatial points to achieve the matching of 3D visual spatial point.After the above operations,the ElasticFusion algorithm is used to build the sensory design model of indoor environment.The indoor environment sensory design modeling based on 3D vision is completed.In the experimental part,the elimination precision of the two models is tested respectively,10 spatial points in the same scene are selected from the database,and two models is used to calculate depth values of 10 spatial points respectively.The experimental results show that the depth values of 10 spatial points calculated by the proposed model are all 0,which meets the elimination requirements.The depth values of 7 spatial points calculated by the indoor environment sensory design model based on machine vision are all greater than 0,which does not meet the elimination requirements.It can be seen from this that the proposed model has high elimination precision.
作者
汤喜辉
TANG Xihui(Henan University of Urban Construction,Pingdingshan 467036,China)
出处
《现代电子技术》
2021年第8期169-172,共4页
Modern Electronics Technique
基金
国家自然科学基金项目(50902075)
河南省科技厅项目(172400410081)。
关键词
室内环境
感官建模
三维视觉
空间点重构
深度值计算
对比实验
indoor environment
sensory modeling
3D vision
spatial point reconstitution
depth value calculation
contrast experiment