基于植被分区的多特征遥感智能分类

引用本文

于菲菲, 曾永年, 徐艳艳, 郑忠, 刘朝松, 王君, 何晋强. 基于植被分区的多特征遥感智能分类[J]. 国土资源遥感, 2014,26(1): 63-70
YU Feifei, ZENG Yongnian, XU Yanyan, ZHENG Zhong, LIU Zhaosong, WANG Jun, HE Jinqiang. Intelligent remote sensing classification of multi-character data based on vegetation partition[J]. REMOTE SENSING FOR LAND & RESOURCES,2014,26(1): 63-70 复制到剪切板

Permissions

《国土资源遥感》编辑部

基于植被分区的多特征遥感智能分类

于菲菲^1,², 曾永年^1,², 徐艳艳^1,², 郑忠^1,², 刘朝松^1,², 王君^1,², 何晋强^1,²

1.中南大学地球科学与信息物理学院,长沙 410083

2.中南大学空间信息技术与可持续发展研究中心,长沙 410083

曾永年(1959-),男,教授,博士生导师,主要从事环境遥感与地理信息系统应用、土地利用/覆盖变化监测与模拟方面的研究。Email:ynzeng@mail.csu.edu.cn。

第一作者简介: 于菲菲(1987-),女,硕士研究生,主要从事遥感信息提取及土地利用变化模拟研究。Email:feifeiyu_2012@163.com。

收稿日期: 2013-02-26

基金: 国家自然科学基金项目(编号: 41171326,41201383和41201386)资助

摘要

为了有效地提取大范围地形复杂区域的土地利用/土地覆盖遥感信息,以位居青藏高原与黄土高原过渡地带的青海东部地区为研究区,研究基于蚁群智能优化算法(ant colony intelligent optimization algorithm,ACIOA)的土地利用/土地覆盖遥感智能分类。首先选用TM图像、DEM、坡度和坡向数据作为分类的特征波段; 然后利用归一化植被指数NDVI对实验区数据进行植被分区; 最后利用ACIOA算法进行分类规则挖掘,并依据分类规则进行土地利用/覆盖信息的提取。研究表明,基于植被分区的多特征蚁群智能分类的总体精度为88.85%, Kappa=0.86,优于传统的遥感图像分类方法,为大范围地形复杂区域的土地利用/土地覆盖遥感信息提取提供了有效的方法。

关键词: 蚁群智能优化算法(ACIOA); 植被分区; 多特征; 遥感分类

中图分类号:TP75S127 文献标志码:A 文章编号:1001-070X(2014)01-0063-08 doi: 10.6046/gtzyyg.2014.01.12

Intelligent remote sensing classification of multi-character data based on vegetation partition

YU Feifei^1,², ZENG Yongnian^1,², XU Yanyan^1,², ZHENG Zhong^1,², LIU Zhaosong^1,², WANG Jun^1,², HE Jinqiang^1,²

1.School of Geosciences and Geomatics, Central South University, Changsha 410083, China

2.Center for Geomatics and Sustainable Development Research, Central South University, Changsha 410083, China

Abstract

In order to effectively extract land use/land cover remote sensing information in a wide range of terrain complex area,the authors, taking the transition zone between Tibetan Plateau and the Loess Plateau in eastern Qinghai as the study area,studied the intelligent remote sensing classification of land use/land cover by using ant colony intelligent optimization algorithm(ACIOA)in this paper. Firstly,TM image,digital elevation model,slope and aspect data were selected as characteristic bands for classification. Secondly,the study area was divided into two parts using the normalized difference vegetation index(NDVI)so as to reduce the influence of different objects with the same spectrum. Finally,the classification rules were excavated using ACIOA,by which regional land use/cover information was extracted. The results show that the ACIOA classification of multi-character data based on vegetation partition is superior to the traditional remote sensing classification. The overall accuracy of the classification and the Kappa coefficient of ACIOA with multi-character data based on vegetation partition is 88.85% and 0.86 respectively. Therefore,this study provides an effective way for extracting land use/land cover information in large-area complex terrain.

Keyword: ant colony intelligent optimization algorithm(ACIOA); vegetation partition; multi-character data; remote sensing classification

Show Figures

0 引言

土地利用/土地覆盖变化(land use and land cover change, LUCC)研究是全球变化研究的热点之一^{[1, 2, 3, 4]}。中国西部地区是全球变化的敏感区域, 地形复杂, 水土流失和土地荒漠化严重, 生态环境极其脆弱^{[5, 6]}。为有效治理西部地区的生态环境, 近年来开展了大规模“ 退耕还林还草” 工程。与此同时, 随着西部大开发战略的深入实施, 城镇化和工业化进程得到加速, 建设用地占用耕地情况较为突出。因此, 中国西部地区土地利用/覆盖发生了较为显著的变化^{[7, 8, 9, 10]}。为了保证西部地区经济的可持续发展和土地资源的合理有效利用, 需要准确、快速地掌握和分析LUCC的信息。

中国西部地区的自然条件复杂, 遥感技术作为快速提取LUCC信息的有效手段, 已在该区土地资源调查和测绘制图中发挥了重要作用^{[11, 12]}, MODIS和NOAA等中、低分辨率遥感数据已被成功应用于该地区土地荒漠化监测^{[13, 14, 15, 16]}。然而, 目前在较大范围区域开展的资源与环境遥感应用中仍以目视解译为主, 费时费力, 且难以保证遥感图像解译的时效性。因此, 如何从海量遥感数据中快速、智能化地提取感兴趣信息是亟待解决的问题。随着群体智能理论的快速发展, 群体智能方法已被引入到遥感信息提取中, 并成为新的研究热点^{[17, 18, 19]}。蚁群智能优化算法(ant colony intelligent optimization algorithm, ACIOA)是模仿自然界的蚂蚁群体觅食过程中沿最短路径行进的生物学行为发展起来的一种群体智能优化方法。ACIOA算法作为一种“ 自上而下” 的群体智能算法, 不需要统计分布的先验知识, 具有很强的自学习能力、鲁棒性、自适应性和正反馈机制等优点, 在海量遥感图像数据的信息提取方面表现出极大的优势。许多学者先后用ACIOA算法进行了遥感图像分类规则的挖掘研究^{[17, 18, 19, 20, 21, 22]}。本文针对大范围复杂区域LUCC信息的提取问题, 采用基于植被分区的ACIOA算法挖掘遥感分类规则, 实现对LUCC信息的有效提取。

本文以平安县为实验区, 探索利用ACIOA算法的综合分类方法实现LUCC遥感信息的准确自动提取, 以期为大范围地形复杂地区的LUCC信息提取提供有效方法。

1 实验区概况与数据源

1.1 实验区概况

实验区选择地处青海东部湟水中游的平安县, 该区域位于青藏高原与黄土高原的过渡地带, 处于E101° 49'~102° 10', N36° 15'~36° 34'之间。全县东西长23 km, 南北宽33.6 km, 土地总面积742.89 km²。区内地形复杂, 沟壑纵横, 海拔在2 066~4 166.7 m之间。该区域属典型的干旱、半干旱地区, 人口密度较大, 经济相对发达, 是以西宁市为中心的青海东部城市群以及新型工业化发展的重点区域, 也是青海省重要的农业区^[23]。有川水、浅山和脑山3大农业类型区; 土地利用类型以林地和牧草地为主, 耕地相对较少, 建设用地所占比例较低。其中, 建设用地与水浇地主要分布在重要交通沿线、湟水谷地及其主要支流河谷地带, 而旱地、林地、牧草地和未利用地主要分布在自然条件较差的浅山和脑山地区, 土地利用空间分布差异明显。

1.2 数据及预处理

采用的遥感数据为TM图像, 来源于中国科学院对地观测与数字地球科学中心数据服务平台(www.ceode.cas.cn), 成像日期为2009年9月21日, 轨道号132/35, 空间分辨率30 m。实验选用了TM1(0.45~0.52 μ m), TM2(0.52~0.60 μ m), TM3(0.63~0.69 μ m), TM4(0.76~0.90 μ m), TM5(1.55~1.75 μ m)和TM7(2.08~2.35 μ m)波段的图像。

此外, 采用DEM、坡度和坡向数据作为TM图像的辅助特征波段; 采用第二次全国土地利用调查数据(该数据是采用2007— 2009年的高分辨率遥感图像经解译、矢量化和野外调查获得的1︰1万比例尺数据, 具有较高精度), 以及接近同期的部分QuickBird高分辨率遥感数据, 作为遥感分类训练样本和分类结果精度验证的地面参考数据; 将实验区土地利用调查数据按本文分类系统进行地类合并; 采用分层随机采样方法选取了852个验证样本, 用于本文方法的精度验证。

TM图像的预处理包括: ①几何纠正, 首先以土地调查数据为参考, 在TM图像上选取地面控制点, 其次构建几何畸变纠正的数学关系, 并进行图像重采样; ②辐射校正, 采用FLAASH大气校正模型进行辐射校正; ③图像裁剪, 依据土地利用调查数据中的平安县行政区矢量数据对图像进行裁剪, 得到实验区范围内的TM5(R)4(G)3(B)假彩色合成图像(图1)。

	Figure Option View Download New Window
	图1 实验区TM5(R)4(G)3(B)假彩色合成图像Fig.1 False color image composed of TM5(R)4(G)3(B) of the study area

2 蚁群智能优化算法

蚁群智能优化算法(ACIOA)具体过程如图2所示, 它是一种基于群体智能的启发式仿生学优化算法, 该算法模拟了自然界中蚂蚁觅食时的路径选择行为^[18]。研究发现, 蚂蚁总能在食物源和蚁穴间选择一条最短的路径, 这是因为蚂蚁在已走过的路径上释放了一种特殊的挥发性物质— — 信息素, 信息素浓度的高低能够引导蚂蚁的路径选择; 当某路径上的信息素浓度高于其他路径时, 蚂蚁更倾向于选择此路径。当蚂蚁开始觅食时, 在未走过的交叉路口处会随机选择一条路径; 随着路径上信息素的不断挥发和积累, 较短路径上的信息素浓度不断增

	Figure Option View Download New Window
	图2 蚁群智能优化算法基本原理示意图Fig.2 Sketch diagram of basic principle of ACIOA

加, 而较长路径上的信息素不断消失。这样, 更多的蚂蚁选择较短的路径; 与此同时, 短路径的信息素浓度继续上升, 于是形成一种正反馈机制; 最终, 所有的蚂蚁都会选择距离最短的路径。

假设蚂蚁在觅食过程中存在2条距离不等的路径ABD和ACD, 各有30只蚂蚁从蚁穴和食物源出发。在t=0时刻, 到达路口A和D时分别有15只蚂蚁从2条路径经过; 因ACD路径比ABD路径短, 故ACD路径上的蚂蚁先到达目的地, 此时, ACD路径的信息素浓度高于ABD路径; 经过t+1时刻, 越来越多的蚂蚁选择路径ACD, 同时该路径信息素的浓度不断增加; 最终, 所有的蚂蚁都选择蚁穴— ACD— 食物源这条较短的路径。

将蚂蚁觅食的这种寻优过程应用于群体智能优化领域, 即产生了ACIOA算法。

3 研究方法

3.1 土地利用/土地覆盖分类体系

参考土地利用调查数据的土地利用分类, 同时考虑实验区遥感图像的实际可解译能力、环境条件及土地利用现状等因素, 根据“ 全国遥感监测土地利用/覆盖分类体系” , 构建了适合于实验区的LUCC遥感信息提取的分类体系(表1)。

表1 实验区的土地利用/土地覆盖分类体系 Tab.1 Land use / cover classification system of the study area

3.2 技术流程

本文采用基于植被分区的多特征数据ACIOA智能分类方法, 具体技术路线如图3所示。

	Figure Option View Download New Window
	图3 技术流程图Fig.3 Technical flow chart

3.3 遥感分类

3.3.1 多特征数据

遥感数据与非遥感数据的复合利用能弥补单一遥感数据的不足, 发挥非遥感数据的辅助作用, 有效地提高遥感数据的分类精度。本文的实验区域位于地形复杂、海拔差异大的中国西部, DEM数据以及由DEM数据生成的坡度和坡向数据能有效地反映植被垂直分带情况与局地生长环境, 是提高遥感分类精度的有效辅助数据^[24]。因此, 本文以TM图像、DEM、坡度和坡向数据作为分类的特征波段。

3.3.2 植被分区

根据干旱、半干旱地区的植被特征, 将实验区分为以植被为主的区域和以非植被为主的区域, 以减少整体分类中异物同谱现象的影响。归一化植被指数(normalized difference vegetation index, NDVI)作为植物生长状态和植被空间分布密度的最佳指示因子, 与植被分布密度呈线性关系, 是遥感估算植被覆盖度研究中常用的植被指数。为此, 采用NDVI进行植被分区。利用预处理后的TM图像生成NDVI, 统计分析训练样本的NDVI值, 确定植被或非植被的划分阈值NDVI=0.18, 以此将实验区划分为以植被为主的区域和以非植被为主的区域。同时, 以土地利用调查数据为参照, 采用分层随机采样的方法采集训练样本, 以植被和非植被为主的实验区域分别选取了872个和432个训练样本。

3.3.3 分类规则智能挖掘

采用ACIOA算法分别对上述2个区域挖掘遥感分类规则。该算法以训练样本的各波段离散值为属性节点, 以各目标地类为类节点, 通过寻优搜索属性节点和类节点的连线, 挖掘相应的遥感分类规则。规则挖掘的步骤包括:

1)波段离散化。在挖掘分类规则前, 采用基于信息熵的离散化方法, 对训练样本的数据进行离散处理^[25]。

2)初始规则构造。为缩短蚂蚁搜索的时间, 构造与问题相关的启发式函数, 引导蚁群的搜索。规则构造时采用赌轮机制选择属性节点, 直到该路径中包含了所有属性; 然后选择1个类节点而且只能有1个类节点, 至此1条完整初始分类规则形成。

3)规则剪枝。完成初始分类规则构造后, 每一个属性节点都被选为规则的1个条件项; 通过规则剪枝, 移除降低分类规则有效性的属性节点, 简化分类规则, 最大化地提高分类规则的有效性。

4)信息素更新。当所有蚂蚁构造的初始规则经过规则剪枝后, 更新所有属性节点的信息素浓度, 直至剩余的训练样本数小于预设样本数。以上分类规则的自动挖掘是在C#环境下, 通过编程实现的。

4 实验结果与分析

4.1 实验结果

1)分别采用ACIOA算法、支持向量机(SVM)和最大似然(ML)方法进行土地利用/覆盖分类, 结果如图4所示。

	Figure Option View Download New Window
	图4 基于植被分区的多特征数据土地利用/覆盖分类结果Fig.4 Land use/cover classification results of multi-character data based on vegetation partition

2)在不进行植被分区的情况下, 利用多特征数据, 采用不同算法得到土地利用/覆盖分类结果(图5)。

	Figure Option View Download New Window
	图5 多特征数据的土地利用/覆盖分类结果Fig.5 Land use/cover classification results of multi- character data

3)在不进行植被分区的情况下, 仅使用TM图像, 采用不同算法得到土地利用/覆盖分类结果(图6)。

	Figure Option View Download New Window
	图6 TM图像的土地利用/覆盖分类结果Fig.6 Land use/cover classification results of TM image

4.2 精度分析

本文选用混淆矩阵作为上述3种不同情况下分类结果的精度评价方法。该方法是国内外广泛使用的遥感分类精度评价方法, 具体评价指标有总体精度、Kappa等, 分别从不同的分析角度评价分类结果的精度。其中, Kappa不随类别样本点数量的不同而改变, 能够更加客观地评价分类精度。

1)在不进行植被分区的情况下, 仅用TM图像, 采用ACIOA算法的总体分类精度为83.22%, Kappa=0.79, 比SVM和ML方法的总体分类精度分别提高了0.59%和2.82%, Kappa分别提高了0.02和0.04(表2)。由此可见, 与SVM和ML方法相比, 基于遥感图像的ACIOA分类方法有一定的优势。

表2 基于TM图像分类精度 Tab.2 Classification accuracy based on TM image

2)在不进行植被分区的情况下, 利用多特征数据图像, 采用ACIOA算法的分类总体精度为85.09%, Kappa=0.81(表3), 比SVM和ML方法的分类总体精度分别提高了1.99%和3.17%, Kappa

表3 多特征数据图像分类精度 Tab.3 Classification accuracy based on multi-character data image

分别提高了0.03和0.04; 比仅用TM图像的分类总体精度提高了1.87%, Kappa提高了0.02。由此可见, 在地形复杂的大范围区域, 多特征数据能够有效地提高遥感数据的分类精度, 并且基于多特征数据的ACIOA分类方法优于SVM和ML方法。

3)在植被分区的情况下, 利用多特征数据图像, 采用ACIOA算法分类的总体精度88.85%, Kappa=0.86, 比SVM和ML方法的分类总体精度分别提高了2%和5.52%, Kappa分别提高了0.03和0.08(表4); 比仅用TM图像和多特征数据图像分类总体精度分别提高了5.63%和3.76%, Kappa分别提高了0.07和0.05。由此可见, 在地形复杂的大范围区域, 基于植被分区的多特征数据遥感分类方法更为有效地提高了土地利用/覆盖遥感信息提取精度, 而且基于植被分区的多特征ACIOA分类方法明显优于SVM和ML方法。

表4 基于植被分区多特征图像分类精度 Tab.4 Classification accuracy based on multi-character data in vegetation partition

5 结论

为有效提取大范围地形复杂地区的土地利用/覆盖遥感信息, 以青藏高原与黄土高原过渡地带的青海东部地区为研究区, 进行了ACIOA分类方法的研究。结果表明:

1)利用遥感图像、DEM、坡度和坡向复合的多特征数据, ACIOA分类的总体精度达到85.09%, 比仅用TM图像ACIOA分类的总体精度提高了1.87%, 并优于基于多特征数据的SVM和ML分类结果。

2)为减小异物同谱现象对整体分类的影响, 根据干旱、半干旱地区的植被特征, 采用植被分区的多特征ACIOA分类的总体精度达到88.85%, 比不分区的ACIOA分类的总体精度提高了3.76%, 并优于基于植被分区的多特征SVM和ML分类结果。

3)本文提出的基于植被分区的多特征数据ACIOA智能分类方法与策略, 能有效地提高大范围地形复杂地区土地利用/覆盖遥感信息提取的精度, 在大范围地形复杂的区域土地利用/覆盖遥感信息提取中具有较大的应用潜力。

The authors have declared that no competing interests exist.

参考文献

View Option

[1]	史培军, 宫鹏, 李晓兵, 等. 土地利用/覆盖变化研究的方法与实践[M]. 北京: 科学出版社, 2000. Shi P J, Gong P, Li X B, et al. The method and practice of land use/cover change[M]. Beijing: Science Press, 2000. [本文引用:1]
[2]	李秀彬. 全球环境变化研究的核心领域——土地利用/土地覆被变化的国际研究动向[J]. 地理学报, 1996, 51(6): 553-558. Li X B. A review of the international researches on land use/land cover change[J]. Acta Geographica Sinica, 1996, 51(6): 553-558. [本文引用:1] [CJCR: 2.937]
[3]	陈佑启, Verburg P H. 中国土地利用/土地覆盖的多尺度空间分布特征分析[J]. 地理科学, 2000, 20(3): 197-202. Chen Y Q, Verburg P H. Multi-scale spatial characterization of land use/land cover in China[J]. Scientia Geographica Singca, 2000, 20(3): 197-202. [本文引用:1]
[4]	宋翔, 颜长珍, 朱艳玲, 等. 黄河源区土地利用/覆被变化及其生态环境效应[J]. 中国沙漠, 2009, 29(6): 1049-1055. Song X, Yan C Z, Zhu Y L, et al. Land use/cover change and associated effects on eco-environment in source region of Yellow River[J]. Journal of Desert Research, 2009, 29(6): 1049-1055. [本文引用:1] [CJCR: 2.099]
[5]	程国栋, 张志强, 李锐. 西部地区生态环境建设的若干问题与政策建议[J]. 地理科学, 2000, 20(6): 503-510. Cheng G D, Zhang Z Q, Li R. On some issues of the ecological construction of west China and proposals for policy[J]. Scientia Geographica Sinica, 2000, 20(6): 503-510. [本文引用:1] [CJCR: 2.779]
[6]	刘彦随, 陈百明. 中国可持续发展问题与土地利用/覆被变化研究[J]. 地理研究, 2002, 21(3): 324-330. Liu Y S, Chen B M. The study framework of land use/cover change based on sustainable development in China[J]. Geographical Research, 2002, 21(3): 324-330. [本文引用:1]
[7]	曾永年, 吴孔江, 靳文凭, 等. 青海高原东北部耕地变化及驱动力分析[J]. 干旱区资源与环境, 2012, 26(8): 89-92. Zeng Y N, Wu K J, Jin W P, et al. The arable land changes and its driving factors in eastern Qinghai Plateau[J]. Journal of Arid Land Resources and Environment, 2012, 26(8): 89-92. [本文引用:1] [CJCR: 0.861]
[8]	曾永年, 靳文凭, 何丽丽, 等. 青海高原东部农业区土地利用遥感分类制图[J]. 农业工程学报, 2012, 28(16): 225-231. Zeng Y N, Jin W P, He L L, et al. Land use mapping using remote sensing for eastern part of Qinghai Plateau[J]. Transactions of the Chinese Society of Agricultural Engineering, 2012, 28(16): 225-231. [本文引用:1] [CJCR: 1.299]
[9]	刘纪远, 张增祥, 庄大方, 等. 20世纪90年代中国土地利用变化时空特征及其成因分析[J]. 地理研究, 2003, 22(1): 1-12. Liu J Y, Zhang Z X, Zhuang D F, et al. A study on the spatial-temporal dynamic changes of land -use and driving forces analyses of China in the 1990s[J]. Geographical Research, 2003, 22(1): 1-12. [本文引用:1]
[10]	何书金, 王秀红, 邓祥征, 等. 中国西部典型地区土地利用变化对比分析[J]. 地理研究, 2006, 25(1): 79-86. He S J, Wang X H, Deng X Z, et al. Analysis on influencing factors of land use change in three typical areas of western China[J]. Geographical Research, 2006, 25(1): 79-86. [本文引用:1]
[11]	吴笛, 吴满意. 青海省第二次全国土地调查中遥感影像目视解译的研究[J]. 科技信息, 2010(7): 430-431. Wu D, Wu M Y. The second national land survey in Qinghai Province in the visual interpretation of remote sensing images[J]. Science and Technology Information, 2010(7): 430-431. [本文引用:1] [CJCR: 0.1452]
[12]	李维森. 新技术在国家西部1︰50 000地形图空白区测图工程中的应用[J]. 测绘科学, 2010, 35(6): 8-11. Li W S. New technology applied in the project of west China topographic mapping at 1: 50 000 scale[J]. Science of Surveying and Mapping, 2010, 35(6): 8-11. [本文引用:1] [CJCR: 0.778]
[13]	沙占江, 曾永年, 马海州, 等. 遥感和GIS支持下的龙羊峡库区土地沙漠化动态研究[J]. 中国沙漠, 2000, 20(1): 51-54. Sha Z J, Zeng Y N, Ma H Z, et al. Dynamic monitoring of desertification with RS and GIS in Longyangxia reservoir area[J]. Journal of Desert Research, 2000, 20(1): 51-54. [本文引用:1] [CJCR: 2.099]
[14]	曾永年, 冯兆东. 沙质荒漠化遥感监测与环境影响研究进展[J]. 山地学报, 2005, 23(2): 218-227. Zeng Y N, Feng Z D. Advances in sand y desertification detecting and its environmental impacts[J]. Journal of Mountain Science, 2005, 23(2): 218-227. [本文引用:1] [CJCR: 0.736]
[15]	刘爱霞, 王长耀, 王静, 等. 基于MODIS和NOAA/AVHRR的荒漠化遥感监测方法[J]. 农业工程学报, 2007, 23(10): 145-150. Liu A X, Wang C Y, Wang J, et al. Method for remote sensing monitoring of desertification based on MODIS and NOAA/AVHRR data[J]. Transactions of the Chinese Society of Agricultural Engineering, 2007, 23(10): 145-150. [本文引用:1] [CJCR: 1.299]
[16]	高志海, 魏怀东, 丁峰. TM影像荒漠化解译与成图技术研究[J]. 遥感技术与应用, 2002, 17(6): 293-298. Gao Z H, Wei H D, Ding F. Study on interpretation and cartography technology of desertification by TM image[J]. Remote Sensing Technology and Application, 2002, 17(6): 293-298. [本文引用:1] [CJCR: 0.749]
[17]	戴芹, 刘建波, 刘士彬. 群智能方法在遥感信息提取中的应用分析[J]. 计算机工程与应用, 2011, 47(4): 13-16. Dai Q, Liu J B, Liu S B. Analysis of remote sensing information extraction using swam intelligence method[J]. Computer Engineering and Applications, 2011, 47(4): 13-16. [本文引用:2] [CJCR: 0.457]
[18]	刘小平, 黎夏, 何晋强, 等. 基于蚁群智能的遥感影像分类新方法[J]. 遥感学报, 2008, 12(2): 253-262. Liu X P, Li X, He J Q, et al. Classification of remote sensing images based on ant colony optimization[J]. Journal of Remote Sensing, 2008, 12(2): 253-262. [本文引用:3] [CJCR: 0.992]
[19]	戴芹, 刘建波. 基于蚁群优化分类规则挖掘的遥感图像分类研究[J]. 计算机工程与应用, 2008, 44(15): 12-14. Dai Q, Liu J B. Research on remote sensing image classification using ant colony optimization(ACO) based classification rule mining algorithm[J]. Computer Engineering and Applications, 2008, 44(15): 12-14. [本文引用:2] [CJCR: 0.457]
[20]	肖菁, 梁燕辉. 基于改进Ant-miner算法的分类规则挖掘[J]. 计算机工程, 2012, 38(17): 162-165. Xiao J, Liang Y H. Classification rule mining based on improved ant-miner algorithm[J]. Computer Engineering, 2012, 38(17): 162-165. [本文引用:1] [CJCR: 0.492]
[21]	Zhang H N, Huang F, Guo J C, et al. Automatic classification of remote sensing image using ant colony clustering algorithm[C]//Proceedings of the 2nd International Congress on Image and Signal Processing. Tianjin: IEEE Press, 2009: 1-4. [本文引用:1]
[22]	Li R, Meia X, Liu J Y. New method of remote sensing image classification based on ant colony[C]//Proceedings of the 6th IEEE Joint International Information Technology and Artificial Intelligence Conference. Chongqing: IEEE Press, 2011: 375-379. [本文引用:1]
[23]	曾永年, 刘茂国, 何丽丽, 等. 青海东部河湟地区建设用地时空变化分析[J]. 国土与自然资源研究, 2011, 34(4): 26-28. Zeng Y N, Liu M G, He L L, et al. Spatial-temple analysis on buildup land changes in Huangshui and Yellow River basins in eastern Qinghai[J]. Territory and Natural Resources Study, 2011, 34(4): 26-28. [本文引用:1]
[24]	杜明义, 武文波, 郭达志. 多源地学信息在土地荒漠化遥感分类中的应用研究[J]. 中国图象图形学报, 2002, 7(7): 740-743. Du M Y, Wu W B, Guo D Z. Research on multi-source geographic information based classification of desertification[J]. Journal of Image and Graphics, 2002, 7(7): 740-743. [本文引用:1] [CJCR: 0.758]
[25]	谢宏, 程浩忠, 牛东晓. 基于信息熵的粗糙集连续属性离散化算法[J]. 计算机学报, 2005, 28(9): 1570-1574. Xie H, Cheng H Z, Niu D X. Discretization of continuous attributes in rough set theory based on information entropy[J]. Chinese Journal of Computers, 2005, 28(9): 1570-1574. [本文引用:1] [CJCR: 1.796]

2000

0.0

... 0 引言土地利用/土地覆盖变化(land use and land cover change,LUCC)研究是全球变化研究的热点之一^[1,2,3,4] ...

1996

0.0

2.937

... 0 引言土地利用/土地覆盖变化(land use and land cover change,LUCC)研究是全球变化研究的热点之一^[1,2,3,4] ...

2000

0.0

陈佑启, Verburg P

中国土地利用/土地覆盖的多尺度空间分布特征分析[J]. 地理科学, 2000, 20(3): 197-202.

Chen Y

, Verburg P

Multi-scale spatial characterization of land use/land cover in China[J]. Scientia Geographica Singca, 2000, 20(3): 197-202.

This paper analyzed the determining factors Of land use and its spatial dependency in China, and through statistical and multi-scale GIS techniques investigated the spatial patterns of land use. Correlation and regression analysis was used to identify the most important explanatory variables from a large set of candidate determining factors. We found that the distribution of all land use types in China will be best described by a combination of different biophysical and socio-economic factors. Specific attention was given to the influence of scale and the spatial distribution of arable land. The results will provide some reference for the land use planning and spicy-making of sustainable utilization of land resources.

研究了我国土地利用影响因素及其空间规模相关性,采用数理统计与GIS手段从多个规模尺度上探讨了我国土地利用的空间分布特征.其结果对全国以及各级政府进行土地利用规划、制定相应的土地开发与利用政策等具有一定的参考意义.

... 0 引言土地利用/土地覆盖变化(land use and land cover change,LUCC)研究是全球变化研究的热点之一^[1,2,3,4] ...

2009

0.0

2.099

宋翔, 颜长珍, 朱艳玲, 等. 黄河源区土地利用/覆被变化及其生态环境效应[J]. 中国沙漠, 2009, 29(6): 1049-1055.

Song

, Yan C

, Zhu Y

, et al. Land use/cover change and associated effects on eco-environment in source region of Yellow River[J]. Journal of Desert Research, 2009, 29(6): 1049-1055.

Through interpreting TM image in 1989 and ETM+ image in 2005, the land use/cover database of the source region of the Yellow river had been established, and then the land use/cover dynamic information was derived by overlaying the database of land use/cover. Based on the land use/cover database, an eco-environment quality index model was used to establish the quantitative relation between land use/cover type and its eco-environment effect. The results show as follows: ①The change of land use/cover was slow and moderate, and two contrary changing trends of degradation and improvement in eco-environment coexisted. In the large region scale, the eco- environment quality kept relatively stable in the period of 1989\_2005, but changed significantly in the small scale. ②Vegetation was the major factor that affected the eco-environment quality in the source region of the Yellow River, followed by grassland factor. ③The eco-environment quality index model can not only be used to study the eco-environment quality in the region scale, but also to find the main driving force of affecting the eco-environment quality.

应用1989年TM影像和2005年ETM+影像提取黄河源区两期土地利用/覆被数据,并利用GIS的空间叠加分析对两期土地利用图进行土地利用动态信息的提取。在建立黄河源区土地利用数据库的基础上,采用生态环境质量指数模型建立各土地利用类型与生态环境质量之间的数量关系,定量的分析土地利用变化对生态环境质量的影响。结果表明:①黄河源区1989-2005年间,土地利用变化缓慢而匀速,存在着生态环境的恶化和改善两种相反的趋势。在大尺度范围内,生态环境质量相对稳定,有轻微的恶化,但是在研究区局部范围内,生态环境质量变化显著;②植被是影响黄河源区生态环境质量的主要因素,其中草地对环境质量的影响处于主导地位;③利用生态环境质量指数模型不仅可以从整体上了解研究区生态环境质量,也可以发现影响区域环境质量的主要驱动力。

... 0 引言土地利用/土地覆盖变化(land use and land cover change,LUCC)研究是全球变化研究的热点之一^[1,2,3,4] ...

2000

0.0

2.779

程国栋, 张志强, 李锐. 西部地区生态环境建设的若干问题与政策建议[J]. 地理科学, 2000, 20(6): 503-510.

Cheng G

, Zhang Z

, Li

On some issues of the ecological construction of west China and proposals for policy[J]. Scientia Geographica Sinica, 2000, 20(6): 503-510.

West China goes across 10 provinces,with an area of about 5.43?106km 2 and 2.85 ?10 8 population ,which accounts for 56.6% of the total area and 22.8% of the total population of China respectively, West China hasa vast area and is the most vulnerable ecological area. The natural geomorphology of Northwest China and Southwest China is widely different. The vulnerably ecological characteristics of West China are severe water and soil erosion, continually developed desertification, low vegetable coverage, increasingly degenerated grasslands, lack of water resource in Northwest China, frequently occurred landslide and mud-rock flow, and so on. Ecological construction is very important to the sustainable development of West China region.Based on the assessment of the eco-environmental status of West China, the issues that existed in the ecological construction and the urgent scientific, social and economic problems that needed to be solved are analyzed in this paper. The aim of ecological construction in West China is also put forward in the paper. In the end, the authors propose some proposals for the policy of the ecological construction in West China in the first half of the 21st century.

西部地区是我国生态环境的脆弱地区,生态建设是西部地区开发建设必须首先研究和解决的一个重大问题。西部地区生态环境十分复杂,西北地区和西南地区的自然生态环境迥异、生态退化问题千差万别。对西部地区的生态环境的基本现状进行了评价,提出了生态环境建设的目标,分析了生态环境建设中面临的亟待研究和解决的科学、技术和社会经济问题,并提出了生态环境建设的若干政策建议。并特别指出,应充分发挥科技对生态建设的支撑作用,科学实施生态建设工程;必须采取基于不同生态经济类型区的自然生态环境特点和生态退化问题的分类指导的生态建设策略;建立西部地区生态安全体系,开展西部地区生态建设和环境保护立法,以实现西部地区生态环境与经济社会的协调可持续发展。

... 中国西部地区是全球变化的敏感区域,地形复杂,水土流失和土地荒漠化严重,生态环境极其脆弱^[5,6] ...

2002

0.0

刘彦随, 陈百明. 中国可持续发展问题与土地利用/覆被变化研究[J]. 地理研究, 2002, 21(3): 324-330.

Liu Y

, Chen B

The study framework of land use/cover change based on sustainable development in China[J]. Geographical Research, 2002, 21(3): 324-330.

Being the core and frontier issue in the global environmental change and sustainable development disciplines, studies on land use/cover change (LUCC) has gained increasingly attention. China, restricted by its large population and less land, is one of the developing countries facing serious ecological and environmental problems in the world. Many problems of sustainable development confronting China have inherent and inevitable associations with LUCC. Therefore, systematic LUCC researches, aiming at the implementation of sustainable development strategies in China, have great significance in both academic and practical respects. Their relations between the problems of sustainable development and LUCC, and the research objects, contents, investigative methods and key issues of LUCC based on sustainable development in China are briefly discussed in this paper. The interrelationship between LUCC and sustainable development in China is very complex. Generally speaking,the linkage can be shown as the balance between land water resources demand and supply. Thus, the orientation and content of LUCC and sustainable development researches should be developed along with the critical issues for sustainable development driving force of human being scenarios of LUCC optimized decision making for sustainable development. Meanwhile, different situations under different temporal and spatial dimensions should be incorporated into the researches. The key issues to be studied are how to reveal the land supply and demand relationships under the multiple driving forces, as well as its changing extents and influences. Upon them, the framework of scientific decision making and comprehensive management for LUCC in China, which is confined by the sustainable development relationship between food and resources supply, ecological protection and economic growth, is proposed.

土地利用 /覆被变化 (LUCC)研究 ,已成为全球环境变化和可持续发展领域前沿的核心问题。中国是世界上人多地少、生态与环境问题较为突出的发展中国家 ,目前中国可持续发展所面临的许多问题 ,都与土地利用及其变化有着内在的、必然的联系。因此 ,面向可持续发展战略的实施 ,开展土地利用 /覆被变化的系统研究 ,具有重要的科学价值与战略意义。本文重点对土地利用 /覆被变化与可持续发展重点问题的关系 ,以及面向可持续发展问题的LUCC研究目标、主要内容和技术方法等进行了讨论

... 中国西部地区是全球变化的敏感区域,地形复杂,水土流失和土地荒漠化严重,生态环境极其脆弱^[5,6] ...

2012

0.0

0.861

曾永年, 吴孔江, 靳文凭, 等. 青海高原东北部耕地变化及驱动力分析[J]. 干旱区资源与环境, 2012, 26(8): 89-92.

Zeng Y

, Wu K

, Jin W

, et al. The arable land changes and its driving factors in eastern Qinghai Plateau[J]. Journal of Arid Land Resources and Environment, 2012, 26(8): 89-92.

Along with the economic developmemt and ecological environment constraction of western China,arable land was reduced in recent years.Taking agricultural areas in eastern Qinghai Plateau as study area,we discussed the arable land changes and driving force.The results indicated that arable land was reduced from 1996 to 2005.The population,GDP and fixed assets investment were the driving factors of socio-ecnomic development.The adjustment of agricultural strcture and ecological restoration were the driving factors of policy control.

以青海高原东部农业区为研究区域,以1996-2005年土地利用现状调查与变更调查统计数据及统计年鉴提供的各类社会经济数据为依据,采用定性与定量相结合的方法,分析耕地变化过程与特点,探寻耕地变化驱动机制。研究结果表明,1996-2005年海东地区耕地面积总体呈下降趋势;耕地变化的主要驱动因素包括社会经济发展、政策调控驱动力两方面。其中,社会经济发展驱动力因子主要有人口、GDP、固定资产投资。政策调控驱动力因子主要有农业结构调整和生态退耕因素。

... 因此,中国西部地区土地利用/覆盖发生了较为显著的变化^[7,8,9,10] ...

2012

0.0

1.299

曾永年, 靳文凭, 何丽丽, 等. 青海高原东部农业区土地利用遥感分类制图[J]. 农业工程学报, 2012, 28(16): 225-231.

Zeng Y

, Jin W

, He L

, et al. Land use mapping using remote sensing for eastern part of Qinghai Plateau[J]. Transactions of the Chinese Society of Agricultural Engineering, 2012, 28(16): 225-231.

In order to effectively monitor ecological effects of the complex area, taking the eastern agricultural area in Qinghai-Tibet Plateau and the Loess Plateau transition zone as case study area, rapid extraction methods of land use information was studied using remote sensing technology in the large complex area. Firstly, according to the study areas characteristics, a classification system for analyzing land use change was designed. Secondly, a comprehensive remote sensing classification method was adopted based on the geographical division in eastern agricultural area of Qinghai plateau. Result showed that the proposed method was suitable for land use information extraction of large complex area, the Kappa coefficient was 0.71 with 0.12 higher than that of the conventional method. The classification mapping indicated that farmland was the main in the structure of land utilization, the proportion of slope farmland from 6° to 25° to total cultivated land was 63.4% and water and soil conservation should be intensified for slope farmland from 6° to 25°. This study provides an effective way for land use remote sensing information extraction in a wide range of complex areas and provides a reference for the scientific management and planning of agricultural land resources in the eastern part of Qinghai Plateau.

为有效地监测复杂区域生态环境治理成效，该文以青藏高原与黄土高原过渡地带的青海高原东部农业区为例，研究大范围复杂地区土地利用遥感自动、快速提取方法。首先构建适于研究区土地利用变化分析的分类系统；其次，采用地理分区综合遥感分类法对青海高原东部农业区进行土地利用遥感分类制图。研究结果表明，上述方法适宜大范围复杂地区土地利用信息的有效提取，Kappa系数0.71，较未分区的分类结果高0.12。分类制图结果表明，研究区土地利用结构以农用地为主，其中＞6°～25°的坡耕地占耕地总量的63.33%，应加大这部分坡耕地水土流失的防治。该研究为大范围复杂地区土地利用遥感信息提取提供了有效的方法，为青海高原东部农业区土地资源的科学管理与规划提供了参考。

... 因此,中国西部地区土地利用/覆盖发生了较为显著的变化^[7,8,9,10] ...

2003

0.0

刘纪远, 张增祥, 庄大方, 等. 20世纪90年代中国土地利用变化时空特征及其成因分析[J]. 地理研究, 2003, 22(1): 1-12.

Liu J

, Zhang Z

, Zhuang D

, et al. A study on the spatial-temporal dynamic changes of land -use and driving forces analyses of China in the 1990s[J]. Geographical Research, 2003, 22(1): 1-12.

Supported by the key knowledge innovation projects,i.e., a preliminary study on the theories and techniques of the remotely sensed temporal-spatial information and digital Earth; and a study on the integration of national resources and environment and data sharing, the authors have set up a spatial-temporal information platform by the integration of the corresponding scientific and research achievements during the periods of the 8th- and 9th-Five Year Plan, which comprehensively reflected the features of land-use change, designed a series of technical frameworks on the spatial-temporal database construction based on remote sensing techniques, e.g., the construction of remotely sensed database and land-use spatial database of the mid-1980s, the mid-1990s and the end of the 1990s, which laid a foundation for the dynamic monitoring of land-use change and the corresponding studies. In this paper,the authors have analyzed comprehensively the features of land-use change in the 1990s, revealed the spatial-temporal change of land use supported by remote sensing and GIS technologies as well as analyzed the geophysical and socio-economic driving factors.The findings are as follows: the arable land has been increased in total amount, the balance of decrease in the south and increase in the north was resulted from the reclamations of grassland and forest land. On the whole, the forest land area had a process of decrease, and the decreased area was mainly distributed in the traditional forest areas. Areas with plentiful precipitation and heat in the south, however, had distinct effects of reforestation. The rural-urban construction land had a situation of persistent expansion, and the general expansion speed has been slowed down during the last five years of the 1990s with the exception of the Western China where the expansion speed has been accelerated. The land use change in China in the 1990s had distinct temporal and spatial differences due to two main reasons, which were policy control and economic driving. Hereby, conclusions and proposals brought forward by the authors were as follows: the spatial diversity rules of the modern land use change in China must be fully considered in the future land use planning. At the same time, the pertinence of physical geographical zones must be considered during the planning of eco-environment construction. And, based on the increasingly maturity of the infrastructure, the traditional thoughts on planning and management of resources must be shifted so as to fully realize the optimized allocation of land resources at regional scale.

在土地利用变化时空信息平台的支持下,本文对我国20世纪80年代末到90年代末的土地利用变化过程进行了全面分析,揭示了我国10年来土地利用变化的时空规律,分析了这些规律形成的主要政策、经济和自然成因。研究表明,20世纪90年代,全国耕地总面积呈北增南减、总量增加的趋势,增量主要来自对北方草地和林地的开垦。林业用地面积呈现总体减少的趋势,减少的林地主要分布于传统林区,南方水热充沛区造林效果明显。中国城乡建设用地整体上表现为持续扩张的态势。90年代后5年总体增速减缓,西部增速加快。20世纪90年代我国的土地利用变化表现出明显的时空差异,政策调控和经济驱动是导致土地利用变化及其时空差异的主要原因。据此,本文提出在今后的全国土地利用规划中,应充分考虑我国现代土地利用变化的区域分异规律。同时,在生态环境恢复与建设规划中也应强调自然地理地带的针对性,同时要改变传统的资源规划与管理思路,在基础设施日益完备的条件下,最大程度地发挥跨区域土地资源优化配置的综合优势

... 因此,中国西部地区土地利用/覆盖发生了较为显著的变化^[7,8,9,10] ...

2006

0.0

何书金, 王秀红, 邓祥征, 等. 中国西部典型地区土地利用变化对比分析[J]. 地理研究, 2006, 25(1): 79-86.

He S

, Wang X

, Deng X

, et al. Analysis on influencing factors of land use change in three typical areas of western China[J]. Geographical Research, 2006, 25(1): 79-86.

Three typical study cases were analyzed based on rural socio-economic statistical data,land survey and remote sensing image interpretation and 136 household questionnaires since 1980.All the three study areas are situated in the interlock area of farming and pasturing along the Great Wall,including Yulin city of Shaanxi province,Jiuquan district of Gansu province in Hexi Corridor of arid Northwest China,and Deyang city in hilly areas of Sichuan Basin. The results showed that land use changes have characteristics of integrity and regionality,which are not only restricted by natural geo-bio-physical mechanisms but also affected by other factors such as society,economy,technology and history.Western China has a long history of land reclamation and varies in types of land cover patterns.Dramatic changes in land use of western China have taken place since the beginning of the 21st century due to unique physiographical characteristics,and some complicated social,economic and political factors.These factors are closely related with land resources utilization relative to agriculture,forestry,and animal husbandry as well as sustainable development relative to urbanization and land degradation.Among these,human factors constitute the main driving forces influencing the region's land use change,including population growth,urbanization,fixed assets investment,development of industry,mining and transportation,economic restructuring,development of science and technology,and policy management etc.

作者选择中国西部长城沿线农牧交错带的陕西省榆林市榆阳区、西北干旱区河西走廊的甘肃省酒泉市肃州区和西南四川盆地丘陵区的四川省德阳市中江县为3个典型研究区域,对20世纪80年代以来的农村社会经济统计资料、土地概查、土地详查、遥感图像判读以及136个典型农户调查进行分析发现:土地利用的变化既受自然的地球生物物理驱动力的制约,又受社会、经济、技术和历史等因素的影响,具有很强的综合性和地域性。西部地区具有悠久的土地开发历史,其土地覆盖类型多种多样。由于其特殊的自然地理特征和社会、经济、政治等驱动因素的复杂多变,本区的土地覆盖发生了很大的变化,这些变化同农业、林业、牧业等与土地资源利用有关的产业发展以及城市化和土地退化等可持续发展问题密切相关。影响西部地区土地利用变化的驱动因素主要是人口增长与城镇化、固定资产投入与工矿交通产业发展、经济结构调整、科技发展和政策管理等人文因素。

... 因此,中国西部地区土地利用/覆盖发生了较为显著的变化^[7,8,9,10] ...

0.0

0.1452

吴笛, 吴满意. 青海省第二次全国土地调查中遥感影像目视解译的研究[J]. 科技信息, 2010(7): 430-431.

, Wu M

The second national land survey in Qinghai Province in the visual interpretation of remote sensing images[J]. Science and Technology Information, 2010(7): 430-431.

在第二次全国土地调查的农村土地调查工作中,遥感影像被大范围的采用,对于遥感影像内业解译兢成为了发挥遥感影像作用的重要环节.本文就遥感影像在内业解译过程中的部分问题与在青海省黄南藏族自治州的实地调查相结合,对内业解译应解决的问题,以及在完成外业勘查后对内业解译能产生指导作用的相关方法予以说明.

... 中国西部地区的自然条件复杂,遥感技术作为快速提取LUCC信息的有效手段,已在该区土地资源调查和测绘制图中发挥了重要作用^[11,12],MODIS和NOAA等中、低分辨率遥感数据已被成功应用于该地区土地荒漠化监测^{[13,14,15,16]} ...

2010

0.0

0.778

李维森. 新技术在国家西部1︰50 000地形图空白区测图工程中的应用[J]. 测绘科学, 2010, 35(6): 8-11.

Li W

New technology applied in the project of west China topographic mapping at 1: 50 000 scale[J]. Science of Surveying and Mapping, 2010, 35(6): 8-11.

国家西部1:50000地形图空白区测图工程(简称西部测图工程)是填补我国西部1:50000基础地理信息空白区的国家重大测绘专项,涉及区域地形、气候、交通等条件异常恶劣,常规测绘手段难以完成全部测图任务。集成航天遥感、数字航空摄影、航空航天合成孔径雷达、全球定位和地理信息系统等现代地理空间信息技术,综合利用多种已经发展成熟的和新型的遥感影像测图技术,是西部测图工程顺利实施关键所在。文章对西部测图工程中使用的这些测绘新技术、新方法进行了较为详细的阐述,这些测绘新技术、新方法形成西部测图工程的创新技术体系,为西部测图工程提供了具有自主知识产权的测图软硬件系统,促进测绘整体技术水平的提高、产业进步与升级。

2000

0.0

2.099

沙占江, 曾永年, 马海州, 等. 遥感和GIS支持下的龙羊峡库区土地沙漠化动态研究[J]. 中国沙漠, 2000, 20(1): 51-54.

Sha Z

, Zeng Y

, Ma H

, et al. Dynamic monitoring of desertification with RS and GIS in Longyangxia reservoir area[J]. Journal of Desert Research, 2000, 20(1): 51-54.

地处高寒干旱与半于旱自然环境的龙羊峡库区生态环境脆弱,加之不合理的人为经济活动影响加剧了沙漠化正过程,使本区土地沙漠化普遍发生,威胁到水库的生产、发展,严重影响着库区经济持续稳定地发展.通过多时相、多波段TM影像(1987、1996),结合实地考察,以IDRISI、ARC/INFO、ARCVIEw等为核心软件,以PⅡ微机为主平台实现遥感与GIS一体化,建立了易操作可运行的龙羊峡库区地理信息系统,在系统支持下对库区土地沙漠化进行动态监测研究.表明土地沙漠化现象日趋严重,严重沙漠化土地面积日益扩大,平均以21 km2·a-1的速度递增,沙漠化正过程迅速发展.造成沙漠化普遍发展的主导因子是自然因素,但人为因素也不容忽视,不合理的经济活动加剧了沙漠化正过程发展.

2005

0.0

0.736

曾永年, 冯兆东. 沙质荒漠化遥感监测与环境影响研究进展[J]. 山地学报, 2005, 23(2): 218-227.

Zeng Y

, Feng Z

Advances in sand y desertification detecting and its environmental impacts[J]. Journal of Mountain Science, 2005, 23(2): 218-227.

荒漠化、全球变暖和生物多样性的减少是人类面临的最主要的环境问题,其中沙质荒漠化又是困扰当前人类社会最严重的环境一社会经济问题.由此,就沙质荒漠化的遥感监测及环境影响研究方面进行了评述与展望.自1977年联合国荒漠化大会以来,通过20多年的努力与研究实践,沙质荒漠化遥感监测已建立了一套以遥感目视解译为主的经验性指标体系,但目前尚未形成一个完全适应于遥感监测的指标体系;沙质荒漠化遥感监测方法从定性走向定量,从主观走向客观.但仍缺乏有效的沙质荒漠化遥感监测方法与技术路线;沙质荒漠化对全球变化的响应的研究相对薄弱,如何准确预测未来气候变化对沙质荒漠化过程的影响是值得深入开展的一项研究工作;沙质荒漠化对气候变化的影响研究多以数值摸拟为主,其中沙质荒漠化对区域气候产生根本性影响的作用机制的研究有待进一步深化;沙质荒漠化对碳循环的影响研究近年来得到重视,但沙质荒漠化对陆地生态系统碳循环的影响研究仍比较薄弱,沙质荒漠化土地土壤碳流失及固碳潜力是有待进一步研究的重要问题.

2007

0.0

1.299

刘爱霞, 王长耀, 王静, 等. 基于MODIS和NOAA/AVHRR的荒漠化遥感监测方法[J]. 农业工程学报, 2007, 23(10): 145-150.

Liu A

, Wang C

, Wang

, et al. Method for remote sensing monitoring of desertification based on MODIS and NOAA/AVHRR data[J]. Transactions of the Chinese Society of Agricultural Engineering, 2007, 23(10): 145-150.

Desertification is one of the most serious ecological and environmental problems in the west of China. Understanding the distribution and development trend of desertification provides researchers important scientific basis for desertification control and rehabilitation. In this paper the authors proposed a method suitable for large-scale desertification monitoring using remote sensing techniques. First, five desertification indexes(MSAVI, FVC, Albedo, LST and TVDI) suitable for large-scale desertification monitoring using remote sensing technique were selected. In terms of the desertification climate types, the potential extent of desertification in China was respectively divided into four categories: dry sub-humid areas, semi-arid areas, arid areas, high and cold areas. Second, different desertification index systems were built for each area. Then based on analysis and comparison of current retrieval algorithms, the authors utilized a suitable algorithm on large scale to retrieve five desertification indexes with ten-day NOAA/AVHRR data set in 1995 and 16-day MODIS data set in 2001 in China. By assessing the classification accuracies of three types of classifiers, the authors selected decision tree classifier for desertification monitoring. Supported by desertification index system and the database of desertification indexes, the desertification status in China in 1995 and 2001 was classified by decision tree classifier, and analysis of desertification changes from 1995 to 2001 was also completed. Statistical result showed that the speed of desertification development was faster than that of rehabilitation, there was a trend of development as a whole and improvement locally in desertificated areas in China. The desertification monitoring results confirmed the practicability of the method founded in this paper for desertification monitoring.

土地荒漠化是中国西部最严重的生态环境问题，荒漠化遥感监测技术是掌握荒漠化发展趋势的重要手段。基于空间分辨率为1 km的NOAA/AVHRR和MODIS遥感数据，首先依据湿润指数法确定了荒漠化范围和气候分区。然后选取出了可以反映不同荒漠化特征，并且能够通过中低分辨率的NOAA/AVHRR和MODIS遥感影像反演出来的5个荒漠化遥感监测指标，通过精度评价分析，确定了最佳指标组合。由于MODIS和NOAA/AVHRR数据的影像特征存在较大差异，针对每一气候分区，建立了分别适用于MODIS和NOAA/AVHRR数据的两套遥感指标分类体系。通过不同分类方法的比较，确定了最佳分类方法－决策树分类法。通过对中国1995年和2001年的荒漠化动态变化状况进行了分析，结果表明，本文提出的荒漠化遥感监测方法，不仅能够成功地进行大尺度的荒漠化遥感监测，而且能够取得较好的监测效果。

2002

0.0

0.749

高志海, 魏怀东, 丁峰. TM影像荒漠化解译与成图技术研究[J]. 遥感技术与应用, 2002, 17(6): 293-298.

Gao Z

, Wei H

, Ding

Study on interpretation and cartography technology of desertification by TM image[J]. Remote Sensing Technology and Application, 2002, 17(6): 293-298.

:There are primary desertification types in Wuwei city that are wind erosion, water erosion and grassland retrogress at middle and low mountain. Through geometrical and projection system transform,standard false color composite, spatial enhance to image, link with land use/cover type, establish interpret symbol of TM image, and to extract information of desertification. It use different model piece of ARC/INFO software adequately, to interpret image, and compile and edit map, and statistic all kind of desertification information data. The results showed that desertification is very severe in Wuwei city, and total desertified lands have 217.07hm2, and occupy 83.33% of total area, among them most is Minqin county.

以TM影像为数据源,系统探讨了影像荒漠化信息提取的图像处理与荒漠化土地类型成图的技术方法。将武威市三地区的荒漠化土地类型分为风蚀荒漠化、水蚀荒漠化和中低山草场退化等类型。通过对图像几何和投影系统转换,图像标准假彩色合成,影像空间增强处理,并结合划分的土地利用/覆被类型,建立TM影像目视解译标志,进行荒漠化信息的提取。借助于ARC/INFO软件的不同模块,进行图像判读、编辑制作图形和统计各类荒漠化信息数据。研究表明,武威市荒漠化现象十分严重,三地区共有各类荒漠化土地217.07万hm2,占其总面积的83.33%,其中以民勤县荒漠化土地面积最大。

2011

0.0

0.457

戴芹, 刘建波, 刘士彬. 群智能方法在遥感信息提取中的应用分析[J]. 计算机工程与应用, 2011, 47(4): 13-16.

Dai

, Liu J

, Liu S

Analysis of remote sensing information extraction using swam intelligence method[J]. Computer Engineering and Applications, 2011, 47(4): 13-16.

Remote sensing data as an important spatial data source play an indispensable role in many areas.Promoting by the development of remote sensing information acquiring technology and expanding application fields，remote sensing information extraction method has also gained great achievements.As the continuing increase in artificial intelligence algorithms and successfully applications，the intelligent algorithms are gradually used in remote sensing information extraction for more efficient processing result.After making a thorough study of remote sensing information research advances，this paper analyzs the potential and advantage of swarm intelligent method that can be used in the research field.And applying particle swarm optimization（PSO） algorithm to remote sensing image classification，this paper demonstrates and implements the technical processing steps of remote sensing image classification，and gains ideal results.So，swarm intelligence algorithm can provide a new efficient method for remote sensing information extraction.

遥感数据作为重要的空间数据源，在众多领域发挥着不可或缺的作用。遥感信息获取技术的不断发展与遥感数据应用领域的不断扩展，促进了遥感信息提取方法的不断进步。随着人工智能算法不断被提出及成功应用，遥感信息提取领域也在逐步引入智能算法实现高效的信息提取。在对遥感信息提取方法的研究进展进行深入分析的基础上，剖析了群智能方法应用于遥感信息提取领域的潜力与优势。并应用微粒群优化方法进行遥感数据的分类，实现了微粒群优化方法应用于遥感数据分类的技术流程，取得了很好的实验结果。因此，群智能方法能够为遥感信息提取领域提供一种新的有效智能处理方法。

... 随着群体智能理论的快速发展,群体智能方法已被引入到遥感信息提取中,并成为新的研究热点^[17,18,19] ...

... 许多学者先后用ACIOA算法进行了遥感图像分类规则的挖掘研究^{[17,18,19,20,21,22]} ...

2008

0.0

0.992

刘小平, 黎夏, 何晋强, 等. 基于蚁群智能的遥感影像分类新方法[J]. 遥感学报, 2008, 12(2): 253-262.

Liu X

, Li

, He J

, et al. Classification of remote sensing images based on ant colony optimization[J]. Journal of Remote Sensing, 2008, 12(2): 253-262.

This paper presents a bottom-up approach to mi prove the classification performance for remote sensing applications. Top-down approaches, such as statistical classifiers, have inherited lmi itations in dealingwith complicated relationships in classification. Forexample, data correlation between bandsofremote sensing mi ageryhas caused problems in generating satisfactory classification with statisticalmethods. In this paper, ant colony optmi ization (ACO) based on swarm intelligence is used to mi prove classification performance. Actually, ACO is a complex multi-agent system, in which agentswith smi ple intelligence can complete complex tasks through cooperation such as classification problems.Ants guide their route selection based on pheromone, which is accumulated from the collectivemovements of individual ants. In thisway, an ant learns from the pastexperience ofothers. Ant-Miner is different from decision tree approaches.The entropymeasure is a local heuristicmeasure, which considers only one attribute at a tmi e, and so it is sensitive to attribute correlation problems. Whereas inAnt-Miner, pheromone updating tends to cope betterwith attribute correlation,since pheromone updating is directly based on the performance of the rule as awhole. Thus, Ant-Miner should have great potential in mi proving remote sensing classification. In this study, an Ant-Miner program for discovering classification rules is developed for the classification of remote sensing mi ages. In theAnt-Miner program, the route search by an ant colony is to find the best links between attribute nodes and class nodes. An attribute node corresponds to a band value of remote sensing mi ages. An attribute node can only be selected once andmustbe associated with a class node. Each route corresponds to a classification rule, and discovering a classification rule can be regarded as searching for an optmi al route.To enableACO to effectively classify remote sensing mi agery of very large data sets, original band values are sliced into a number of intervals by using a discretization technique. The ACO method is more explicit and comprehensible than mathematical equations. Our study in Guangzhou city indicates that the ant colony-based classifier yields better accuracy than conventionalmaxmi um likelihood classifiers and decision tree classifiers. The overall accuracy of theACO method is 88.6%, with aKappa coefficientof0.861. The decision treemethod has an accuracy of85.4% and aKappa coefficientof 0.822. Themaxmi um likelihoodmethod has an accuracy of83. 3% and aKappa coefficient of0. 796. The results clearly support the conclusion that themethod explored in thispapercan bemore effective than conventional classificationmethods.

智能式遥感分类是遥感研究的新热点.提出了一种基于蚁群智能规则挖掘(ant-miner)的遥感影像分类新方法.遥感数据各波段之间存在较强的相关性,这种相关性往往会导致分类产生误差.而ant-miner算法中的信息素是基于规则整体性能的,信息素的动态更新能有效地处理相关性较强的数据,所提供的正反馈信息能纠正启发式函数缺陷所造成的错误.因此,蚁群智能算法应用于遥感分类具有一定的优势.将该方法用于广州市地区的遥感影像,取得了较好的分类结果.并与See5.0决策树方法及最大似然方法(MLH)进行了对比研究,实验结果表明,蚁群智能算法分类精度比后两者的分类精度更高.

... 随着群体智能理论的快速发展,群体智能方法已被引入到遥感信息提取中,并成为新的研究热点^[17,18,19] ...

... 许多学者先后用ACIOA算法进行了遥感图像分类规则的挖掘研究^{[17,18,19,20,21,22]} ...

... 2 蚁群智能优化算法蚁群智能优化算法(ACIOA)具体过程如图2所示,它是一种基于群体智能的启发式仿生学优化算法,该算法模拟了自然界中蚂蚁觅食时的路径选择行为^[18] ...

2008

0.0

0.457

戴芹, 刘建波. 基于蚁群优化分类规则挖掘的遥感图像分类研究[J]. 计算机工程与应用, 2008, 44(15): 12-14.

Dai

, Liu J

Research on remote sensing image classification using ant colony optimization(ACO) based classification rule mining algorithm[J]. Computer Engineering and Applications, 2008, 44(15): 12-14.

Ant Colony Optimization（ACO） as one of the main algorithms in swarm intelligence has been applied successfully to optimization problems in many research areas，but it is still a new research topic in remote sensing data processing.The ACO algorithm has self-organization，cooperation，communication，and other intelligent merits，and without prior knowledge of statistical distribution parameters，so it has many potential advantages in the remote sensing data processing research area.This paper introduces the theory and algorithm process of the ACO for mining classification rules applied to remote sensing image classification.Selecting CBERS image located in Beijing area as experimental data，this paper use the ACO based rule mining algorithm for constructing the classification rules，then use these rules to classify the example data，and compared the classification result with Maximum Likelihood Classification（MLC） method，the experiment results show that ACO based classification rule mining algorithm has provided a new method for remote sensing image classification.

蚁群优化算法作为群智能理论的主要算法之一，已经成功应用在众多研究领域的优化问题上，但是在遥感数据处理领域还是一个新的研究课题。蚁群优化具有自组织、合作、通信等智能化优点，对数据无需统计分布参数的先验知识，因此在遥感数据处理领域具有很大的潜在优势。介绍了将蚁群优化分类规则挖掘算法应用到遥感图像分类研究领域的理论与算法流程。并采用北京地区的CBERS遥感数据作为实验数据，通过蚁群优化算法构造分类规则，对选择的遥感数据进行了分类实验，并和最大似然分类方法进行对比，实验结果表明，蚁群优化分类规则挖掘算法为遥感图像的分类提供了一种新方法。

... 随着群体智能理论的快速发展,群体智能方法已被引入到遥感信息提取中,并成为新的研究热点^[17,18,19] ...

... 许多学者先后用ACIOA算法进行了遥感图像分类规则的挖掘研究^{[17,18,19,20,21,22]} ...

2012

0.0

0.492

肖菁, 梁燕辉. 基于改进Ant-miner算法的分类规则挖掘[J]. 计算机工程, 2012, 38(17): 162-165.

Xiao

, Liang Y

Classification rule mining based on improved ant-miner algorithm[J]. Computer Engineering, 2012, 38(17): 162-165.

In order to improve the classification rule accuracy of the classical Ant-miner algorithm, this paper proposes an improved Ant-miner algorithm for classification rule mining. Heuristic function with sample density and sample proportion is constructed to avoid the bias caused by the same probability in Ant-miner. A pruning strategy with mutation probability is emploied to expand the search space and improve the rule accuracy. An evaporation coefficient in Ant-miner’s pheromone update formula is added to slow down the convergence rate of the algorithm. Experimental results on UCI datasets show that the proposed algorithm is promising and can obtain higher predication accuracy than the original Ant-miner algorithm.

为提高基于传统Ant-miner算法分类规则的预测准确性，提出一种基于改进Ant-miner的分类规则挖掘算法。利用样例在总样本中的密度及比例构造启发式函数，以避免在多个具有相同概率的选择条件下造成算法偏见。对剪枝规则按变异系数进行单点变异，由此扩大规则的搜索空间，提高规则的预测准确度。在Ant-miner算法的信息素更新公式中加入挥发系数，使其更接近现实蚂蚁的觅食行为，防止算法过早收敛。基于UCI标准数据的实验结果表明，该算法相比传统Ant-miner算法具有更高的预测准确度。

... 许多学者先后用ACIOA算法进行了遥感图像分类规则的挖掘研究^{[17,18,19,20,21,22]} ...

2009

0.0

... 许多学者先后用ACIOA算法进行了遥感图像分类规则的挖掘研究^{[17,18,19,20,21,22]} ...

2011

0.0

... 许多学者先后用ACIOA算法进行了遥感图像分类规则的挖掘研究^{[17,18,19,20,21,22]} ...

2011

0.0

曾永年, 刘茂国, 何丽丽, 等. 青海东部河湟地区建设用地时空变化分析[J]. 国土与自然资源研究, 2011, 34(4): 26-28.

Zeng Y

, Liu M

, He L

, et al. Spatial-temple analysis on buildup land changes in Huangshui and Yellow River basins in eastern Qinghai[J]. Territory and Natural Resources Study, 2011, 34(4): 26-28.

应用土地利用现状调查与变更调查数据,分析了青海东部河湟地区建设用地结构、规模数量及空间分布随时间变化的规律与特点.研究结果表明,建设用地规模迅速扩张,城镇建设用地与农村居民点用地同步增长；建设用地集约节约水平不高,建设用地外延扩展占用耕地现象较为突出.青海高原东部经济发展所带来的建设用地的扩张与耕地的减少给区域土地资源的规划与管理提出了挑战.

... 该区域属典型的干旱、半干旱地区,人口密度较大,经济相对发达,是以西宁市为中心的青海东部城市群以及新型工业化发展的重点区域,也是青海省重要的农业区^[23] ...

2002

0.0

0.758

杜明义, 武文波, 郭达志. 多源地学信息在土地荒漠化遥感分类中的应用研究[J]. 中国图象图形学报, 2002, 7(7): 740-743.

Du M

, Wu W

, Guo D

Research on multi-source geographic information based classification of desertification[J]. Journal of Image and Graphics, 2002, 7(7): 740-743.

土地荒漠化是当今全球面临的重大环境问题之一,它的发生、发展及其逆转是气候、环境和人类社会经济活动综合作用的结果.荒漠化土地分类是土地荒漠化研究中至关重要的一个环节,由于地形粗糙度和植被覆盖率对土地荒漠化具有重要的影响,但遥感图象很难对地表地形粗糙度进行定量描述,因此引入了数字高程模型(DEM)对研究区地形粗糙度进行表达和采用植被指数(NDVI)对地面植被覆盖率进行描述,并对由DEM生成的地面坡度(sLOPE)图象、植被指数图象与原始遥感图象进行了信息融合,从而大大丰富了遥感图象的荒漠化信息.通过对融合图象的n-维散度分析,不同类型、不同程度的土地荒漠化样本的分离度大幅度提高.分类结果也证明,信息融合可大大提高分类精度.

... 本文的实验区域位于地形复杂、海拔差异大的中国西部,DEM数据以及由DEM数据生成的坡度和坡向数据能有效地反映植被垂直分带情况与局地生长环境,是提高遥感分类精度的有效辅助数据^[24] ...

2005

0.0

1.796

谢宏, 程浩忠, 牛东晓. 基于信息熵的粗糙集连续属性离散化算法[J]. 计算机学报, 2005, 28(9): 1570-1574.

Xie

, Cheng H

, Niu D

Discretization of continuous attributes in rough set theory based on information entropy[J]. Chinese Journal of Computers, 2005, 28(9): 1570-1574.

该文提出了一种新的粗糙集连续属性离散化算法.首先对每一个候选断点定义了信息熵,以此作为对断点重要性的量度,在此基础上给出了断点选择的粗糙集连续属性离散化算法.最后采用多组数据对此算法的性能进行了检验,并与其它算法做了对比实验.实验结果表明此算法是有效的,而且当候选断点个数增多时仍有很高的计算效率.

... 在挖掘分类规则前,采用基于信息熵的离散化方法,对训练样本的数据进行离散处理^[25] ...