基于经验值ξ因子小样本图像二维至立体视图实现  

Implementation of Small Sample Images Two-dimensional to Stereo View Based on Experience ValuexξFactor

作　　者：张海军 陈映辉[1,3] 钟向阳 卓汉逵[4] ZHANG Hai-jun;CHEN Ying-hui;ZHONG Xiang-yang;ZHUO Han-kui(Guangdong Provincial Key Laboratory of Conservation and Precision Utilization of Characteristic Agricultural Resources in Mountainous Areas,Meizhou 514015,China;School of Computing,Jiaying University,Meizhou 514015,China;School of Mathematics,Jiaying University,Meizhou 514015,China;School of Computing,Sun Yat-sen University,Guangzhou 510006,China)

机构地区：[1]广东省山区特色农业资源保护与精准利用重点实验室,广东梅州514015 [2]嘉应学院计算机学院,广东梅州514015 [3]嘉应学院数学学院,广东梅州514015 [4]中山大学计算机学院,广州510006

出　　处：《嘉应学院学报》2024年第3期9-18,共10页Journal of Jiaying University

基　　金：国家自然科学基金项目(61171141,61573145);广东省自然科学基金重点项目(2014B010104001,2015A030308018);广东省普通高等学校人文社会科学省市共建重点研究基地(18KYKT11);广东省嘉应学院自然科学基金重点项目(2017KJZ02);广东省嘉应学院自然科学基金项目(2021KJY05);教育部产学合作协同育人项目(201802153047);2019年广东省教育厅高校特色创新项目(2019KTSCX169)。

摘　　要：面向小样本目标,给出了一种基于经验值ξ因子小样本图像二维至立体视图实现的方法.1.目标小样本原始特征点提取,基于二次元等方法构建射映、旋转、缩放和平移等变换阵;2.获得基于小样本的构建因子旋转变换量和平移变换量;3.基于变换把平面点逆映射成三维空间点;4.设定经验值ξ因子、无限逼近创建贴片;5.经迭代构建最优目标.实验结果表明,本方法对小样本重建计算量低、效率较高、重建视图可视性好.For small sample targets,a method for the implementation of small sample images two-dimensional to stereo view based on experience valuexξfactor was proposed.Firstly,the original feature points of small samples were extracted,and the transformation arrays of projection,rotation,scaling and translation were constructed based on the quadratic element methods.Secondly,construction factor rotation transformation quantity and translation transformation quantity based on small samples were obtained.Thirdly,based on the transformation,the plane points were inversely mapped into three dimensional space points.Fourthly,the experience valuexξfactor was set and the surface patch was created by infinite approximation.Finally,the optimal goal was constructed by iterating.The experimental results showed that the proposed method had low computational cost,high efficiency and good visibility of reconstructed view for small samples.

关 键 词：小样本 射影 自标定 贴片 经验值 稀疏稠密化 

分 类 号：TP3[自动化与计算机技术—计算机科学与技术]