基于多重HMM的脱机手写汉字识别

介绍了一种基于多重隐马尔克夫模型(Multiple HMM——MHMM)的手写体汉字识别新方法。该方法首先提取基于区域投影变换形成的边界链码特征,对每个汉字建立4个HMM,通过等比重综合方法将4个分类器的计算结果进行综合,从而得到识别结果。实验结果证明该方法较传统的HMM具有更高的抗干扰能力和更高的识别率。     

实对称矩阵空间的直交投影

在实对称矩阵空间建立了内积和范数 ,在由所有的半正定矩阵所组成的闭凸子集上定义了一种直交投影 ,并且讨论了它的一些性质和一个应用 .     

Banach空间上的投影算子及应用

将HiIbert空间上的投影算子[1-2]推广到Banach空间,并讨论了它的性质、运算及在线性代数中的应用。     

轴测投影中各夹角参数之间的关系

空间直角坐标系的三个轴互相垂直。把坐标系向轴测投影面投影时，坐标轴与轴测投影面的垂线的三个夹角，坐标轴与投影方向的三个夹角以及投影方向与轴测投影面的夹角，都直接影响到轴测投影的结果，因此讨论各个夹角之间的关系是必要的。本文讨论了相关七个夹角参数之间的关系。     

深入调研加强规划提高学科建设质量和效益

本文在调查研究的基础上，提出了学院学科建设与发展规划的设想，并对学院学科建设的保障措施进行了探讨。     

概念整合对幽默言语的阐释力

本文从认知的角度,利用概念整合网络及图表形式清晰地将幽默效果的获取这一动态的过程呈现出来,一方面丰富了国内关于幽默言语的研究,有助于人们对幽默言语的欣赏,另一方面也有助于人们对概念整合相对于概念隐喻理论的优势的了解。     

The Project ‘Art for Learning Art’ in Contemporary Art Museums

Projection‐based augmented reality and virtual reality are used in a visual arts‐based educational project in contemporary art museums from an a/r/tographic perspective. The project ‘Art for Learning Art’ (in Spanish, Arte para aprender arte), at the Museo CajaGRANADA in Granada (Spain) has been developed in collaboration with the University of Granada since 2013. We have employed creative, educational and research methodologies inspired by exhibited works in art museums to encourage visual feedback of visitors participating in collaborative installations. Two such experiences also were produced at the Tate Liverpool Gallery and Museum in March 2018; utilising the methodology of mediation through projection‐based augmented reality and virtual reality, which introduces facets of visual and physical experience that alter the whole experience for the museum's public. By putting on virtual reality headsets, and playing with physical movements, we generate images and change the projections in the museum using projection‐based augmented reality to disrupt the way the public typically moves in the museum. The purpose of the developed interrelations with artworks in the Museo CajaGRANADA and the Tate Liverpool Gallery and Museum was to create collaborative digital images by playing with select artworks exhibited in the museums’ collections. We use this kind of mediation in art museums to develop a visual understanding to provoke learning about art through art creation in a contemporary way. The results are extraordinary as images; they are collaborative artworks, which connect visually with the artworks in the exhibition.     

影像辅助LiDAR点云的地物提取方法研究

从Li DAR点云中准确地提取地物是Li DAR数据处理过程中的一项关键工作。针对Li DAR点云数据在滤波分类和点云精确提取分类方面的不足,提出了一种新的影像辅助下Li DAR点云地物提取的方法。首先,对Li DAR点云按高程值重采样生成高程投影影像并自动分割,同时从高分辨率正射影像提取特征影像,然后将特征影像和由Li DAR生成的分割影像进行集成分析,利用影像提取结果辅助Li DAR点云错分、漏分纠正和点云精确分类处理,最后通过试验分析,验证了该方法的可行性与有效性。     

“器官投影”说——技术发展的内在逻辑

工具的创造体现着人类智慧的结晶,工具的演化历史更负载着技术的内涵。人是技术创造的尺度,“器官投影”理论是工具演化的本源。本文从工具发展的各个历史阶段扫“器官投影”学说相结合来透析技术自身演变的逻辑性,即技术是怎样成为人体器官的结构与功能在外界的投影。