应用科技
應用科技
응용과기
YING YONG KE JI
2015年
3期
60-64
,共5页
林宝聚%马思群%张伟龑%李健%陶然%李恩实
林寶聚%馬思群%張偉龑%李健%陶然%李恩實
림보취%마사군%장위엄%리건%도연%리은실
动车组%虚拟现实%LOD技术%边折叠方法%AABB算法
動車組%虛擬現實%LOD技術%邊摺疊方法%AABB算法
동차조%허의현실%LOD기술%변절첩방법%AABB산법
electric multiple units%virtual reality%LOD technology%edge collapse method%AABB algorithm
多细节层次模型( LOD)技术已广泛应用于计算机仿真、虚拟现实等多个领域。动车组虚拟现实系统中,由于复杂模型零件个数众多,结构复杂,为了尽可能展示逼真的虚拟场景,运用LOD理念进行模型简化具有重要意义。现有的研究成果多着眼于模型网格和纹理结构的简化,较少考虑模型本身所具有的外部形状特征。本文提出了一种分步式模型简化算法,先通过简化的边折叠方法进行预处理,再运用基于包围盒的AABB算法进一步简化,并且通过引入阈值来控制生成不同精度的模型。实验证明,处理后的模型不但达到了较高的简化率,而且具有较高的真实感。
多細節層次模型( LOD)技術已廣汎應用于計算機倣真、虛擬現實等多箇領域。動車組虛擬現實繫統中,由于複雜模型零件箇數衆多,結構複雜,為瞭儘可能展示逼真的虛擬場景,運用LOD理唸進行模型簡化具有重要意義。現有的研究成果多著眼于模型網格和紋理結構的簡化,較少攷慮模型本身所具有的外部形狀特徵。本文提齣瞭一種分步式模型簡化算法,先通過簡化的邊摺疊方法進行預處理,再運用基于包圍盒的AABB算法進一步簡化,併且通過引入閾值來控製生成不同精度的模型。實驗證明,處理後的模型不但達到瞭較高的簡化率,而且具有較高的真實感。
다세절층차모형( LOD)기술이엄범응용우계산궤방진、허의현실등다개영역。동차조허의현실계통중,유우복잡모형령건개수음다,결구복잡,위료진가능전시핍진적허의장경,운용LOD이념진행모형간화구유중요의의。현유적연구성과다착안우모형망격화문리결구적간화,교소고필모형본신소구유적외부형상특정。본문제출료일충분보식모형간화산법,선통과간화적변절첩방법진행예처리,재운용기우포위합적AABB산법진일보간화,병차통과인입역치래공제생성불동정도적모형。실험증명,처리후적모형불단체도료교고적간화솔,이차구유교고적진실감。
Level of detail ( LOD) technology is widely used in computer simulation, virtual reality and other fields. In the virtual reality system of electric multiple units ( EMUs) , complex models with complex structures have many spare parts, so simplifying the model by LOD technology is important in showing realistic virtual scenes. Most of re?searches currently only focus on simplifying the grid and texture of the model while ignoring the characteristics of the external shape of the model. In this paper, a method to simplify models by fractional steps is proposed. This method contains two steps as follows:the first step is to preprocess the model by the edge collapse method;the sec?ond step is to use the axis?aligned bounding box ( AABB) algorithm based on bounding box to do further simplifica?tion. Different precision models are generated by the introduction of thresholds. The experiment results show that the model which has been processed not only achieves a high simplification rate, but also has high fidelity.