三维视觉类电子出版物中冗余光源的判别——基于K-L展开式  

Discrimination of Redundant Light Sources in 3D Visual Electronic Publications——Based on K-L Expansion

作　　者：黄洁 董刚 HUANG Jie;DONG Gang(S&T Information Institute,Shandong Academy of Agricultural Sciences,Jinan Shandong 250100,China;Shandong Agriculture and Engineering University,Jinan Shandong 250100)

机构地区：[1]山东省农业科学院科技信息研究所,山东济南250100 [2]山东农业工程学院,山东济南250100

出　　处：《山东农业工程学院学报》2019年第11期48-53,共6页The Journal of Shandong Agriculture and Engineering University

摘　　要：近年来,随着计算机图形学技术的进步,电子出版物中越来越多的出现各种三维视觉类的新鲜东西,如:虚拟现实(VR)、增强现实(AR)、实景地图服务等等。一些三维模型设计中存在光源过多的问题,这就可能造成视觉污染。识别和消除多余的光源,是一种有效的消除此类视觉污染的方法。本文基于K-L展开式进行特征提取,对识别冗余光源的方法进行了初步的探究,明确了此类方法的有效性。并对一些常用计算方法进行了改进,提高了运行性能,加强了结果的可信性。In recent years, with the advancement of computer graphics technology, more and more new things of three-dimensional vision appeared in electronic publications, such as virtual reality(VR), augmented reality(AR), real-world map services and so on. There are too many light sources in some three-dimensional model design, which may cause visual pollution. Identifying and eliminating redundant light sources is an effective method to eliminate such visual pollution. In this paper, based on K-L expansion for feature extraction, the method of identifying redundant light sources is preliminarily explored, and the effectiveness of this method is clarified. Some common calculation methods are improved to improve the operation performance and enhance the credibility of the results.

关 键 词：视觉污染 电子出版物 光源 K-L变换 AHP算法 

分 类 号：TP311[自动化与计算机技术—计算机软件与理论]