基于Kinect的室内空间特征多层次布局仿真  

Kinect-Based Simulation of Multi-Level Layout of Interior Spatial Features

作　　者：李彦平[1] 张璐[1] LI Yan-ping;ZHANG Lu(College of Information,Shanxi Agricultural University,Taigu Shanxi 030800,China)

机构地区：[1]山西农业大学信息学院,山西太谷030800

出　　处：《计算机仿真》2021年第5期357-361,共5页Computer Simulation

基　　金：2020年度山西省高等学校哲学社会科学研究项目(2020W338)。

摘　　要：传统的室内空间特征多层次布局仿真方法在面对复杂的空间环境时,很难在保证布局合理的情况下,保证仿真数据的精度。为此,提出基于Kinect的室内空间特征多层次布局仿真。使用Kinect设备采集室内空间特征深度数据,经过修复和滤波后,将处理后的数据转换为空间坐标系下的数据,构建空间特征数据的几何模型,采用遗传算法计算出最大迭代次数,获取模型最优解,完成室内空间特征多层次布局仿真。以模型整体偏差以及空间特征数据聚类两方面作为评价指标,仿真结果表明:提出的基于Kinect的室内空间特征多层次布局仿真方法数据聚类效果好、整体偏差小,该方法的数据精度较高。Generally, the traditional simulation method of indoor space feature multi-level layout cannot maintain the accuracy of simulation data. Therefore, this paper proposed a Kinect-based multi-level layout simulation of indoor space features. The indoor spatial feature depth data was collected via the use of Kinect equipment, and then the data was repaired and filtered, and then the processed data were converted into the data under the spatial coordinate system, thus founding the geometric model of the spatial feature data. According to the genetic algorithm, the maximum number of iterations was calculated. The optimal solution of the model was obtained to complete the multi-level layout simulation of indoor space characteristics. The simulation experiment was carried out with the overall deviation of the model and spatial feature data clustering as evaluation indexes. The results show that the method designed in this paper has an excellent clustering effect, minor deviation, and high data precision.

关 键 词：遗传算法 布局方案 多层次布局 空间布局 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]