吉林大学学报(地球科学版) ›› 2019, Vol. 49 ›› Issue (5): 1477-1485.doi: 10.13278/j.cnki.jjuese.20180208

• 地球探测与信息技术 • 上一篇    下一篇

基于自适应可变滤镜的地类变化预测模型

柳长源1,2, 刘鹏1, 毕晓君2   

  1. 1. 哈尔滨理工大学电气与电子工程学院, 哈尔滨 150080;
    2. 哈尔滨工程大学信息与通信工程学院, 哈尔滨 150001
  • 收稿日期:2018-08-07 发布日期:2019-10-10
  • 作者简介:柳长源(1971-),男,副教授,主要从事模式识别与图像处理、机器学习技术研究,E-mail:liuchangyuan@hrbust.edu.cn
  • 基金资助:
    国家自然科学基金项目(51779050);黑龙江省自然科学基金项目(F2016022)

Land Use Change Prediction Model Based on Adaptive Variable Filter

Liu Changyuan1,2, Liu Peng1, Bi Xiaojun2   

  1. 1. School of Electrical and Electronic Engineering, Harbin University of Science and Technology, Harbin 150080, China;
    2. College of Information and Communication Engineering, Harbin Engineering University, Harbin 150001, China
  • Received:2018-08-07 Published:2019-10-10
  • Supported by:
    Supported by National Natural Science Foundation of China (51779050) and Natural Science Foundation of Heilongjiang Province (F2016022)

摘要: 随着土地开发建设规模不断扩大,土地利用情况也在逐年发生变化,准确预测未来土地利用的发展趋势,可以为本地区的土地利用规划提供依据,提升本地区的土地利用效率。传统方法一般采用CA_Markov、ANN以及CA_ANN模型进行预测,存在训练时间长、预测精度不足和缺乏说服力等问题。本文针对上述问题,结合元胞自动机以及人工神经网络模型,建立一种自适应可变滤镜网络模型,针对特定大小区域内的土地类别数目,创建多类数据集来训练不同参数的多个神经网络,可以成功预测未来土地变化的情况,这样就避免了训练单一网络时数据对网络权值的抵消。相比于传统模型中效果最好的CA_ANN模型,本文建立的自适应可变滤镜网络模型不仅总体精度提高了1%~3%,各种地类转化精度提高了12.82%~33.33%,模型预测时间也缩减了49.47%。

关键词: 遥感图像, 土地利用预测, 人工神经网络, 元胞自动机, 自适应可变滤镜

Abstract: With the continuous expansion of land development and construction scale, land use conditions are also changing year by year. Accurate forecasting of future land use development trends can provide a basis for regional land use planning and improve the efficiency of regional land use. Traditionally,the methods of CA_Markov, ANN, and CA_ANN models are usually used for prediction; however,there are problems such as long training time, poor prediction accuracy, and lack of persuasiveness. Aiming at the above problems, the authors established an adaptive variable filter network model in combination with the cellular automaton and neural network models, and created multiple data sets based on the number of land use categories within a certain area for training of multiple neural networks with different parameters. This model can predict the future land change situation, thus avoid the cancellation of network weights when training a single network. Compared with the best model CA_ANN out of the traditional ones, the overall accuracy of this model is improved by 1%-3%, the accuracy of land conversion is improved by 12.82%-33.33%, and the model predicting time is reduced by 49.47%.

Key words: remote sensing image, land use forecast, artificial neural network, cellular automata, adaptive variable filter

中图分类号: 

  • TP753
[1] Sterk B,Leeuwis C,van Ittersum M K.Land Use Models in Complex Societal Problem Solving:Plug and Play or Networking[J]. Environmental Modelling and Software,2008,24(2):165-172.
[2] 张新荣,刘林萍,方石,等.土地利用、覆被变化(LUCC)与环境变化关系研究进展[J].生态环境学报,2014,23(12):2013-2021. Zhang Xinrong, Liu Linping, Fang Shi, et al. Research Progress on the Relationship Between Land Use and Cover Change (LUCC) and Environmental Change[J]. Journal of Eco-Environment,2014,23(12):2013-2021.
[3] 骈宇哲,姜朋辉,陈振杰,等.LUCC研究进展及其对干旱区生态环境的意义[J].水土保持研究,2015,22(5):358-364. Pian Yuzhe, Jiang Penghui, Chen Zhenjie, et al. Research Progress of LUCC and Its Significance to the Ecological Environment in Arid Area[J]. Research of Soil and Water Conservation, 2015,22(5):358-364.
[4] 张丽,杨国范,刘吉平.1986-2012年抚顺市土地利用动态变化及热点分析[J].地理科学,2014,34(2):185-191. Zhang Li, Yang Guofan, Liu Jiping. Analysis of Land Use Dynamic Changes and Hot Spots in Fushun City from 1986 to 2012[J]. Geographical Sciences,2014,34(2):185-191.
[5] Mazzocchi C,Corsi S,Sali G.Agricultural Land Consumption in Periurban Areas:A Methodological Approach for Risk Assessment Using Artificial Neural Networks and Spatial Correlation in Northern Italy[J].Applied Spatial Analysis and Policy,2017,10(1):3-20.
[6] Tauhid R,Faheemah T,Md R,et al.Temporal Dynamics of Land Use/Land Cover Change and Its Prediction Using CA_ANN Model for Southwestern Coastal Bangladesh[J]. Environmental Monitoring and Assessment,2017,189(11):10-12.
[7] Ganbold G,Chasia S.Comparison Between Possibilistic c-Means (PCM) and Artificial Neural Network (ANN) Classification Algorithms in Land Use/Land Cover Classification[J].International Journal of Knowledge Content Development & Technology,2017,7(1):57-58.
[8] 韩会然,杨成凤,宋金平.北京市土地利用空间格局演化模拟及预测[J].地理科学进展,2015,34(8):976-986. Han Huiran, Yang Chengfeng, Song Jinping. Simulation and Prediction of Land Use Spatial Pattern Evolution in Beijing[J]. Progress in Geography,2015,34(8):976-986.
[9] 金明一,钱坤.基于遥感数据的哈尔滨土地利用/覆被变化研究[J].测绘与空间地理信息,2016,39(12):110-111. Jin Mingyi,Qian Kun. Study on Land Use/Cover Change in Harbin Based on Remote Sensing Data[J]. Surveying and Mapping and Spatial Geography Information,2016,39(12):110-111.
[10] 李晓东,姜琦刚.吉林西部多时相遥感数据分类方案的构建及应用[J].吉林大学学报(地球科学版),2017,47(3):907-915. Li Xiaodong,Jiang Qigang. Construction and Application of Multi-Temporal Remote Sensing Data Classification Scheme in Western Jilin Province[J]. Journal of Jilin University (Earth Science Edition),2017,47(3):907-915.
[11] 修春亮,程林,宋伟.重新发现哈尔滨地理位置的价值:基于洲际航空物流[J].地理研究,2010,29(5):811-819. Xiu Chunliang,Cheng Lin, Song Wei. Rediscovering the Value of Harbin's Geographical Location:Based on Intercontinental Aviation Logistics[J].Geographical Research,2010,29(5):811-819.
[12] 朱利凯,蒙吉军.国际LUCC模型研究进展及趋势[J].地理科学进展,2009,28(5):782-790. Zhu Likai,Meng Jijun. Research Progress and Trend of International LUCC Models[J]. Progress in Geography,2009,28(5):782-790.
[13] 白穆,刘慧平,乔瑜,等.高分辨率遥感图像分类方法在LUCC中的研究进展[J].国土资源遥感,2010(1):19-23. Bai Mu, Liu Huiping, Qiao Yu,et al. Research Progress of High Resolution Remote Sensing Image Classification Methods in LUCC[J]. Remote Sensing for Land & Resources,2010(1):19-23.
[14] 周嵩山,李红波.元胞自动机(CA)模型在土地利用领域的研究综述[J].地理信息世界,2012,10(5):6-10. Zhou Songshan,Li Hongbo. A Survey of Cellular Automata (CA) Models in the Field of Land Use[J]. The World of Geographic Information,2012,10(5):6-10.
[15] 陆秋琴,牛倩倩,黄光球.记忆原理的元胞自动机优化算法及其收敛性证明[J].计算机科学,2013,40(4):249-255. Lu Qiuqin,Niu Qianqian,Huang Guangqiu. Cellular Automata Optimization Algorithm of Memory Theory and Proof of Its Convergence[J].Computer Science,2013,40(4):249-255.
[16] 杨俊,解鹏,席建超,等.基于元胞自动机模型的土地利用变化模拟:以大连经济技术开发区为例[J].地理学报,2015,70(3):461-475. Yang Jun,Xie Peng,Xi Jianchao,et al. Land Use Change Simulation Based on Cellular Automata Model:A Case Study of Dalian Economic and Technological Development Zone[J]. Acta Geographica Sinica,2015,70(3):461-475.
[17] 赛莉莉,王涛,陈坤,等.2000-2010年威海城市土地利用变化及马尔科夫预测[J].鲁东大学学报(自然科学版),2016,32(2):162-167. Sai Lili,Wang Tao,Chen Kun, et al. Urban Land Use Change and Markov Prediction in Weihai City from 2000 to 2010[J]. Journal of Ludong University (Natural Science),2016,32(2):162-167.
[18] 李志明,宋戈,鲁帅,等.基于CA_Markov模型的哈尔滨市土地利用变化预测研究[J].中国农业资源与区划,2017,38(12):41-48. LI Zhiming,Song Ge,Lu Shuai,et al. Research on Land Use Change Prediction in Harbin Based on CA_Markov Model[J]. China Agricultural Resources and Regional Planning,2017,38(12):41-48.
[19] 范晓锋.基于ANN_CA模型的珲春市土地利用格局模拟研究[D].长春:吉林大学,2016. Fan Xiaofeng. Simulation of Land Use Pattern in Hunchun City Based on ANN_CA Model[D]. Changchun:Jilin University,2016.
[1] 王明常, 张馨月, 张旭晴, 王凤艳, 牛雪峰, 王红. 基于极限学习机的GF-2影像分类[J]. 吉林大学学报(地球科学版), 2018, 48(2): 373-378.
[2] 王宇, 卢文喜, 卞建民, 侯泽宇. 三种地下水位动态预测模型在吉林西部的应用与对比[J]. 吉林大学学报(地球科学版), 2015, 45(3): 886-891.
[3] 陈永良,李学斌. 核概率距离聚类方法及应用[J]. 吉林大学学报(地球科学版), 2013, 43(1): 312-318.
[4] 陈永良, 李学斌, 林楠. 遥感图像像素级异常识别的一种方法[J]. J4, 2012, 42(3): 881-886.
[5] 王羽, 肖盛燮, 冯五一, 张元才, 于忆骅. 土质边坡失稳判别的CA-AA耦合模型[J]. J4, 2010, 40(1): 148-152.
[6] 秦胜伍,陈剑平. 隧道围岩压力的神经网络时间序列分析[J]. J4, 2008, 38(6): 1005-1009.
[7] 邱道宏,陈剑平,阙金声,安鹏程. 基于粗糙集和人工神经网络的洞室岩体质量评价[J]. J4, 2008, 38(1): 86-0091.
[8] 徐佩华,陈剑平,阙金声,仲志成,王 清. 基于人工神经网络的三峡水库库岸稳定性分级[J]. J4, 2007, 37(3): 564-0569.
[9] 林 玎,刘 伟,张治国. 自组织特征映射神经网络在厄尔尼诺事件检验中的应用[J]. J4, 2006, 36(04): 609-612.
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
No Suggested Reading articles found!