An adaptive unimodal subclass decomposition (AUSD) learning system for land use and land cover classification using high-resolution remote sensing

Land use and land cover classification is an important application of remote-sensing images. The performances of most classification models are largely limited by the incompleteness of the calibration set and the complexity of spectral features. It is difficult for models to realize continuous learning when the study area is transferred or enlarged. This paper proposed an adaptive unimodal subclass decomposition (AUSD) learning system, which comprises two-level iterative learning controls: The inner loop separates each class into several unimodal Gaussian subclasses; the outer loop utilizes transfer learning to extend the model to adapt to supplementary calibration set collected from enlarged study areas. The proposed model can be efficiently adjusted according to the variability of spectral signatures caused by the increasingly high-resolution imagery. The classification result can be obtained using the Gaussian mixture model by Bayesian decision theory. This AUSD learning system was validated using simulated data with the Gaussian distribution and multi-area SPOT-5 high-resolution images with 2.5-m resolution. The experimental results on numerical data demonstrated the ability of continuous learning. The proposed method achieved an overall accuracy of over 90% in all the experiments, validating the effectiveness as well as its superiority over several widely used classification methods.     

唐山市区土地利用的Landsat 8影像分层分类

针对土地利用遥感分类方法多样、分类精度高低不一等问题,该文以土地利用变化明显的唐山市路南区、路北区为研究区域,并以中分遥感影像Landsat 8OLI为信息源,在对地类样本进行可分离性分析的基础上,建立研究区土地利用分层分类体系。通过监督分类实验,选择分类效果最好、分类精度最高的最大似然分类器进行地类初分;通过绘制归一化植被指数(NDVI)、归一化建筑指数(NDBI)、两指数差值(NDVI-NDBI)的曲线及地类光谱特征曲线,建立决策树分类规则,进行地类再分。该方法可以较好地完成多种土地利用二级地类的划分,有助于提高中分影像土地利用分类效率。     

CA-Markov模型的清江土地利用变化研究

针对经济发展类型面临转型的山区土地利用动态变化和预测问题,该文以清江中下游为例,提出了采用CA-Markov模型的研究方法。结合2000、2008和2015年的Landsat遥感影像,考虑地形、交通、行政中心和当前土地利用约束等影响因子,模拟出2015年预测情况与解译结果相比,Kappa系数达0.810 9,表明模型可信度较高。在此基础上预测2022年的土地利用情况,结果表明,2022年延续了当前发展模式,即由农业经济主导转变为旅游生态经济主导,耕地的重要性降低,城镇化率增加,但耕地的面积仍占据着总面积的23.23%,建设用地面积仅占总面积的2.62%,说明该区域的城镇化水平依然较低。该结果揭示了清江流域土地利用动态变化趋势。     

20世纪80年代以来全球耕地变化的基本特征及空间格局

本文基于全球1982-2011年土地利用/覆被的矢量数据,分析了20世纪80年代以来全球耕地变化的基本特征及空间格局。结果表明：① 20世纪80年代以来,全球耕地面积增加了528.768×10⁴ km²,增加速率为7.920×10⁴ km²/a,呈不显著增加趋势,全球耕地面积以20世纪80年代增速最快。20世纪80年代以来,北美洲、南美洲、大洋洲耕地面积呈显著增加趋势,分别增加了170.854×10⁴ km²、107.890×10⁴ km²、186.492×10⁴ km²,增加速率分别为7.236×10⁴ km²/a、2.780×10⁴ km²/a、3.758×10⁴ km²/a;亚洲、欧洲、非洲耕地面积为减少趋势,分别减少了23.769×10⁴ km²、4.035×10⁴ km²、86.76×10⁴ km²,减少速率分别为-5.641×10⁴ km²/a、-0.813×10⁴ km²/a、 -0.595×10⁴ km²/a。② 20世纪80年代以来,全球增加的耕地主要由草地、林地转化,分别占53.536%、26.148%。新增耕地面积主要分布在非洲南部及中部、澳大利亚东部和北部、南美洲东南部、美国的中部及阿拉斯加、加拿大中部、俄罗斯西部及芬兰北部、蒙古北部等区域。非洲南部的博茨瓦纳为全球耕地增加比例最高区域,增加了80%~90%。③ 20世纪80年代以来,全球耕地换化为其他用地共计1071.946×10⁴ km²,全球减少的耕地主要转化为了草地、林地,分别占比为57.482%、36.000%;全球减少耕地主要分布在非洲中部的苏丹南部、美国中南部、俄罗斯南部及欧洲南部的保加利亚、罗马尼亚、塞尔维亚和匈牙利等国,减少最大的区域为非洲南部,减少了60%。④ 各大洲耕地均表现出向高纬扩张的趋势,全球多数国家表现出新增耕地扩张而原有耕地减少的特点。     

Making a Cultural Landscape: The Case of Dongshan River Basin,Taiwan

Developing a cultural landscape establishes a regional identity, creates value for residents, and has potential to diversify a community’s economic base due to tourism development. Changes in local landscape morphology lead to multidimensional challenges in terms of the environmental consciousness and cultural identity of residents. The study examined residents’ perception of changes in their local landscape regarding its visual form, the meanings they constructed with the landscape, and their functional experience of living and working in it. Findings revealed that in its agricultural era, the visual landscape, meanings, and experience consisted mainly of physical elements connected with residents’ small-scale agricultural and fishery production. In the tourism era, the visual landscape, meanings, and experience were connected to recreational settings and iconic landmarks associated with the tourism industry. Both agricultural and tourism development occurred in ways that complemented each other and built a cultural landscape that connected the past with the present.     

GIS-based dynamic modelling and analysis of flash floods considering land-use planning

Flood inundation is a common natural disaster and a growing development challenge for many cities and thousands of small towns around the world. Soil features have frequently altered with the rapid development of urbanised regions, which has led to more frequent and longer duration of flooding in urban flood-prone regions. Thus, this paper presents a geographic information system (GIS)-based methodology for measuring and visualising the effects on urban flash floods generated by land-use changes over time. The measurement is formulated with a time series in order to perform a dynamic analysis. A catchment mesh is introduced into a hydrological model for reflecting the spatial layouts of infrastructure and structures over different construction periods. The Geelong Waurn Ponds campus of Deakin University is then selected as a case study. Based on GIS simulation and mapping technologies, this research illustrates the evolutionary process of flash floods. The paper then describes flood inundation for different built environments and presents a comparison by quantifying the flooding extents for infrastructure and structures. The results reveal that the GIS-based estimation model can examine urban flash floods in different development phases and identify the change of flooding extents in terms of land-use planning. This study will bring benefits to urban planners in raising awareness of flood impact and the approach proposed here could be used for flood mitigation through future urban planning.     

岳阳市土地利用结构变化与土地利用绩效的关联度分析

利用信息熵、灰色关联等分析方法对岳阳市2009—2015年土地利用结构信息熵变化规律、土地利用绩效指数以及它们之间的内在联系进行系统分析。结果表明:1时间上,岳阳市土地利用结构信息熵持续增长,土地利用变化主要表现为城镇村及工矿用地、交通运输用地、耕地的大幅增加以及园地、林地、水域及水利设施用地等地类的减少;2空间上,岳阳市各地区信息熵存在较大差异,中心城区信息熵明显高于其他区域,且经济发展水平与信息熵值呈正相关;3岳阳市土地利用绩效指数逐年上升,其中经济绩效对综合绩效的带动作用最明显,公平绩效则相对滞后;4与土地利用绩效相关性最显著的是城镇村及工矿用地、交通运输用地,但关联系数都较低。     

基于变化向量分析（CVA）的土地利用／覆盖变化动态监测（Ⅱ）—变化类型的确定方法

以光谱直接比较为基础的变化向量分析法是一种非常有效的土地利用／覆盖变化动态监测方法，在双窗口变步长阈值搜寻方法确定变化和非变化像元的基础上，提出了参考图像分类并结合变化向量方向余弦最小距离分类的变化类型确定方法，同时应用该方法在北京市海淀区进行了实验研究，得到了较为理想的结果。变化类型的判断精度达到70％以上，显示了新方法的优越性和技术可行性。     

以土地利用/土地覆被变化为中心的土地科学研究进展

土地利用 /土地覆被变化已成为土地科学、地球系统科学、全球环境变化及可持续发展研究中的核心领域。其研究的根本目标是实现土地资源的有效管理与可持续利用 ,这不仅是解决资源与环境问题的根本立足点 ,也是经济增长与区域可持续发展过程的重要前提。针对国际土地科学研究中最为活跃的几个前沿领域 :土地利用 /土地覆被变化、土地资源系统分析与评价、土地质量指标体系、生态环境效应及持续土地利用 ,结合近年来国内外有关的主要研究计划和项目 ,本文对以土地利用 /土地覆被变化为中心的土地科学研究进展进行了综述。     

三峡库区土地持续利用潜力与途径模式

建立了量化评价模型,对三峡库区土地资源持续利用的潜力及其差异性进行了评价与分析。指出库区土地利用总体上处于弱持续发展的阶段,而且地区差异悬殊、面临的障碍因素多。因此,未来库区土地利用优化决策的关键在于遵照因地制宜和分类指导的原则,正确处理生态建设、移民安置和经济协调发展的关系。主要的途径模式包括植树造林、坡地改梯田、优化结构和产业突破等四个方面。