主题学习和稀疏表示的MODIS图像超分辨率重建

针对MODIS图像分辨率受传感器限制和噪声干扰,且分辨率局限在一定水平等问题,提出一种采用主题学习和稀疏表示的MODIS图像超分辨率重建方法,该方法通过双边滤波将MODIS图像的平滑及纹理部分分离,并将纹理部分看成是由若干"文档"组成的训练样本;运用概率潜在语义分析提取"文档"的潜在语义特征,从而确定"文档"所属的"主题"。在此基础上,针对每个主题所对应的图像块,采用改进的K-SVD方法训练若干适用于不同主题的高低分辨率字典对,从而可以运用这些字典对,通过稀疏编码实现测试图像相应主题块的超分辨率重建。实验结果表明,重建图像在视觉效果和PSNR等指标上均优于传统方法。     

基于机器学习的南京市旅游地个性及其文化景观表征

旅游地具有物质和精神双重属性。鉴于旅游地精神属性在量化表达和测量方面的困难,以往旅游地研究较少涉及。旅游地个性概念的提出为旅游地精神属性研究打开了思路。由于开展时间不长,当前旅游地个性研究大多基于营销学中的品牌个性研究思路展开,缺乏对地方适用性的考量,且尚未深入到对个性表征形式及表征机制的探讨。基于此,从地理学视角出发,选取城市旅游地作为研究对象,着眼地方特性,系统构建城市旅游地个性量表,在此基础上,开展针对案例地城市南京的个性测量和分析,并以城市文化景观作为表征媒介,深入探究城市旅游地个性的景观表征形式及表征机制。研究发现：城市旅游地个性主要包含态度、气质、性格、能力四大特征;城市文化景观是城市旅游地个性的重要表征媒介,包含现代空间景观、传统空间景观、生态景观、生活景观、社会景观五大维度;以上景观维度及其所辖景观元素在表征城市旅游地个性方面发挥不同作用。研究过程及结论为包括旅游地在内的地方精神属性研究提供了一种概念框架和方法体系,是对精神层面人地关系认识上的丰富。机器学习这一智能化探索型数据分析手段的运用为处理地理学中普遍存在的高维非线性噪声数据提供了有力的方法支撑。     

Calibrating a Land Parcel Cellular Automaton (LP-CA) for urban growth simulation based on ensemble learning

The reliability of raster cellular automaton (CA) models for fine-scale land change simulations has been increasingly questioned, because regular pixels/grids cannot precisely represent irregular geographical entities and their interactions. Vector CA models can address these deficiencies due to the ability of the vector data structure to represent realistic urban entities. This study presents a new land parcel cellular automaton (LP-CA) model for simulating urban land changes. The innovation of this model is the use of ensemble learning method for automatic calibration. The proposed model is applied in Shenzhen, China. The experimental results indicate that bagging-Naïve Bayes yields the highest calibration accuracy among a set of selected classifiers. The assessment of neighborhood sensitivity suggests that the LP-CA model achieves the highest simulation accuracy with neighbor radius r = 2. The calibrated LP-CA is used to project future urban land use changes in Shenzhen, and the results are found to be consistent with those specified in the official city plan.     

Mobile Learning in a Human Geography Field Course

This article reports on reusable mobile digital learning resources designed to assist human geography undergraduate students in exploring the geographies of life in Dublin. Developing active learning that goes beyond data collection to encourage observation and thinking in the field is important. Achieving this in the context of large class sizes presents several challenges. Combining in situ learning with spatially accurate historical and contemporary multimedia, a set of location-aware digital mobile tools, or mediascapes, was developed. How scaffolding can be achieved in such a context, focusing on the development of students’ observational, enquiry, and thinking skills in the field was explored.     

基于核半监督判别分析的高光谱影像特征提取

针对高光谱影像特征提取中地物类别训练样本获取代价较高的情况,在线性判别分析的基础上,结合核方法和半监督学习理论,提出了一种基于核半监督判别分析(KSDA)的高光谱影像特征提取方法。该方法同时利用少量已知类别和大量未知类别样本数据进行模型的学习和训练。通过OMIS高光谱影像数据实验表明:在少量已知类别训练样本的条件下,经KSDA特征提取的样本数据在特征空间中能更好地聚集成团,且类别之间的距离较大,增加了类别之间的可分性,得到了较高的分类精度;同时,提取的特征影像能够较好地区分各种地物类别。     

基于用户点击数据的细粒度图像识别方法概述

近年来,细粒度图像识别逐渐成为计算机视觉领域的研究热点.由于不同类别图像间的视觉差异小、语义鸿沟问题严重,传统的基于视觉特征的细粒度图像识别性能往往不尽人意.针对这些挑战,目前许多学者都在研究基于用户点击数据的图像识别.本文围绕点击数据在图像识别中数据预处理、特征提取和模型构建3大模块中的应用,总结了已有的基于点击数据的识别算法及最新的研究进展.     

Empowering learning through design: a comparative analysis of ICM course development for web-based and classroom delivery

This article reviews the development of two courses focussed on meeting Integrated Coastal Management capacity building needs: an undergraduate degree course, and a web-based distance learning course. Influential factors in the development of each course are identified and compared. Important considerations are found to be methods of quality assurance, course content, the temporal and spatial context of teaching and learning, level of learner empowerment, and the role of the teacher. It is concluded that in order to understand the role of such courses in capacity building, coastal educators need to share experiences and encourage reflection on the effectiveness of capacity building efforts.     

一种多层神经网络的学习算法

提出一种基于序贯预测误差方法（ＳＰＥ）的多层神经网络（ＭＮＮ）的学习算法。经模拟计算，它比传统的基于最陡下降方法的误差反传（ＳＤＢＥＰ）算法具有更好的收敛性能。并对这两种算法进行了模拟计算的比较．