spatial differences and temporal diversities are taken account into GeoCA, it makes simulations with this kind of model much closer to actualit

Size: px

Start display at page:

Download "spatial differences and temporal diversities are taken account into GeoCA, it makes simulations with this kind of model much closer to actualit"

顶甭宋
5 years ago
Views:

1 Journal of Image and Graphics Vol. 15, No. 6 June, ) 1), 2), 3) 1) (, ) 2) (, ) 3) (, ), :,,, ;,,:,,,,,, : TP23 : A : (2010) A Partitioned & A synchronous CA Ba sed on Spa tia l Da ta M in ing 2) 1) KE Xinli 1), 2), 3), B IAN Fuling 1) ( Spatial Inform ation & D igital Engineer Research Center of W uhan University, W uhan ) ( Institute of Geographic Science & N atural Resources Research, Chinese Academ y of Sciences, B eijing ) 3) ( School of Resources & Environm ent Science, X ianning College, X ianning ) Abstract In traditional CA, general transfer rule and same speed are used to drive models. temp differences of geographical phenomenon, both transfer rules and transfer speed, are ignored. partitioned & asynchronous CA based on spatial data m ining is given in this paper. : ( ) ; ( 2006BAB15B02) ;( 2009CDB342) ; (2010Q130) : ;: : (1977 ),, , GIS E2mail: kevin. kxl@ gmail. com In this kind of CA, spatial2 To solve this p roblem, a In this model, cell space is departed by dual2constraint spatial cluster and general transfer rule is rep laced by partition transfer rule, asynchronous transfer speed is calculated using general grid and synchronous transfer speed is rep laced by asynchronous transfer speed. of geographical transfer rule and geographical transfer speed are taken into account in this kind of models. Spatial differences Taking land use change in Hangzhou as a case, partitioned & asynchronous CA based on spatial data m ining are imp lemented in land use. Result shows: comparing with traditional CA, partitioned & asynchronous CA based on spatial data m ining can get more accurate simulation result, and this kind of models can be used to simulate geographical phenomenon in a larger area for a comparatively longer time. A new viewpoint of GeoCA is given in partitioned & asynchronous CA. In this kind of Model,

2 spatial differences and temporal diversities are taken account into GeoCA, it makes simulations with this kind of model much closer to actuality. Yet theories and methods of partitioned & asynchronous CA is still in tentative research stage, there are many p roblem s, such as divided methods of cell space, methods of calculating weight of dual2constraint spatial cluster, calculation of cell transfer speed, getting of transfer rules, evaluation of CA result accuracy, and app lications of partitioned & asynchronous CA in simulating land use at larger area and during longer time and etc. should be discussed and analyzed. Keywords CA, spatial data m ining, partition, asynchronous 0 (CA),,, [ 1 ],, [ 229 ], GIS, GIS, GIS Jenerette CA, [ 10 ] L iu Andersson [ 11 ] Takeshi Tetsuya, [ 12 ] L iu Stuart CA, [ 13 ] Claudia Bayes, Bauru [ 14 ], ANN2CA [ 15 ], (ACO2CA ) [ 16 ], BP [ 17 ] (Agent) [ 18 ],,,,, ;,,,,,,, , 1,, ( ),,,,,

3 6 : Fig. 1 Working flow of partitioned asynchronous CA based on spatial data m ining,,, ;,,, :,, ;,,,, ,,,,,,,,, : S = {O 1, O 2,, O n },O i S, S, [ 19 ],,,, Euclid, Euclid [ 20 ] : D ij = w p ( x i - x j ) 2 + ( y i - y j ) 2 + w a w p + w a = 1 m k =1 w k = 1 m k =1 w k ( z ik - z jk ) 2 (1), D ij i j Euclid ; ( x i, y i ) ( x j, y j ) i j ; z ik, z jk i j k; m ; w p, w a Euclid ; w k w p = 1, w a = 0, D ij, ; w p = 0, w a = 1, D ij, ,

4 924 15,,,,,,,,,,, : v i = c i s i (2), v i i, c i i ( ), s i i ( ) (2) v i i, v i, v i,,, v i, (3): In terva l i = 1 v i (3), In terva l i v i, 0 v i 1, 1 In terva l i, Interval i ;, Interval i =,, M axi, [ 1, M axi ], M axi M axi (4) : Interval i = Interval i Interva l i M axi M axi Interva l i >M axi (4),,, Interval i,, ;, : GIS C Fig. 2 Structure of partitioned & asynchronous CA based on spatial data m ining

6 : 925, GIS GIS, GIS ; () ; C5. 0, C5. 0;, ; C5. 0, 2 2.

5 6 : 925, GIS GIS, GIS ; () ; C5. 0, C5. 0;, ; C5. 0, , km 2, km 2 8, 3, 2 3,,, ( 3) Landsat TM /ETM + : ERDAS Imagine 8. 7, ; A rcgis9. 2 DEM, A rcgis9. 2,, 3 Fig. 3 Research area

926 15 2. 2 2. 2. 1, 2000 2005, 2000 2005 ( ),4 : 1),,,,,,, ; 2), w p w a,, w p w a 4 Fig.

6 , , ( ),4 : 1),,,,,,, ; 2), w p w a,, w p w a 4 Fig. 4 Result of dual2constraint spatial cluster,, 5 5,, ( ),,, ( ) , , , ( ), CA 2000,

7 6 : Fig. 5 Comparing result of traditional spatial cluster and dual2constraint spatial cluster 2005, 7,, 2005,

8 , CA :, CA ;, Moran I ,,

9 6 : CA CA Tab. 1 Com par ing ma tr ix of trad itiona l CA 2005 / % ,,, 87127%, 91104%, 89149%,, CA Tab. 2 Com par ing ma tr ix of partitioned & a synchronous CA result 2005 / % ,, Moran I 3 3, CA Moran I CA 3 M oran I Tab. 3 Com par ing table of M ora l I ind ices 2005 CA CA Moran I Moran I, CA 3, ;,,,,,,,, ;,,:,,,,,,, w p w a, w p w a, w p w a,,,,

10 ( References) [ 1 ] L i Xia, Ye Jia an, L iu Xiaop ing, et al. Geographical Simulation System s: Cellular Automata and Multi2Agent Systems [ M ]. Beijing: Science Press, 2007: 1. [,,,. : [M ]. :, 2007: 1. ] [ 2 ] L iu Yaolin, L iu Yanfang, M ing Dongp ing. U rban expansion model based on cellular automata of decision2making with grey situation [ J ]. Geomatics and Information Science of W uhan University, 2004, 29 (1) : [,,. [ J ]. :, 2004, 29 (1) : ] [ 3 ] L iu Xiaop ing, L i Xia. Fisher discrim inant and automatically getting transition rule of CA [ J ]. Acta Geodaetica et Cartographica Sinica, 2007, 36 (1) : [,. Fisher [ J ]., 2007, 36 (1) : ] [ 4 ] Zhang Yan, Chen Yunhao, L i Jing. Study on the dynam ic city model based on cellular automata and population condition [ J ]. Journal of Image and Graphics, 2007, 12 (8) : [,,. [ J ]., 2007, 12 (8) : ] [ 5 ] Batty M, Xie Yichun, Sun Zhanli. Modeling urban dynam ics through GIS2based cellular automata [ J ]. Computers, Environment and U rban Systems, 1999, 23 (3) : [ 6 ] A lmeida C M D, Batty M, Antonio M V M, et al. Stochastic cellular automata modeling of urban land use dynam ics [ J ]. Computer, Environment and U rban Systems, 2003, 27 ( 5 ) : [ 7 ] Sun Tengda, W ang Jinfeng. M icroscop ic traffic simulation model based on road network grid [ J ]. Journal of System Simulation, 2007, 19 (12) : [,. [ J ]., 2007, 19 (12) : ] [ 8 ] Huang Guangqiu, W ang Xiaohai, Chen Huim ing. New method of computer simulation on fire spreading along underground tunnels system based on cellular automata [ J ]. Journal of System Simulation, 2007, 19 ( 1 ) : [,,. [ J ]., 2007, 19 (1) : ] [ 9 ] Yang Q ingsheng, L i Xia. Cellular automata for simulating land use changes based on support vector machine [ J ]. Journal of Remote Sensing, 2006, 10 (6) : [,. [ J ]., 2006, 10 (6) : ] [ 10 ] Jenerette G D. Analysis and simulation of land2use change in the central A rizona2phoenix region, USA [ J ]. 2001, 16 (7) : Landscape Ecology, [ 11 ] L iu X H, Andersson C. A ssessing the impact of temporal dynam ics on land use change modeling [ J ]. Computer, Environment and U rban System s, 2004, 28 (122) : [ 12 ] Takeshi A, Tetsuya A. Emp irical analysis for estimating land use transition potential functions2case in Tokyo metropolitan region [ J ]. Computer, Environment and U rban System s, 2004, 28 (122) : [ 13 ] L iu L, Stuart R P. Modeling urban development with cellular automata incorporating fuzzy2set approaches [ J ]. Environment and U rban System s, 2003, 27 (5) : Computers, [ 14 ] Claudia M A, BattyM. Stochastic cellular automata modeling of urban land use dynamics: Empirical development and estimation [ J ]. Computers Environment and U rban System s, 2003, 27 (5) : [ 15 ] L i Xia, Ye Jia an. Neural2network2based cellular automata for realistic and idealized urban simulation [ J ]. Acta Geographica Sinica, 2002, 57 (2) : [,. CA [ J ]., 2002, 57 (2) : ] [ 16 ] L iu X P, L i X, Yeh A G O. D iscovery of transition rules for geographical cellular automata by using ant colony optim ization [ J ]. Science in China: Series D, 2007, 50 (10) : [ 17 ] Hu Maogui, Fu Xiaoyang, Zhang Shuqing, et al. Land2cover simulation and forcast of melmeg wetland using cellular automata [ J ]. Resource Science, 2007, 29 ( 2) : [,,,. [ J ]., 2007, 29 (2) : ] [ 18 ] Yang Q ingsheng, L i Xia. Integration of multi2agent system s with cellular automata for simulation urban land expansion [ J ]. Sciatica Geographic Sinica. [,. [ J ]., 2007, 27 (4) : ] [ 19 ] Zhou Jiaogen. Study on Dual2constraint and A lterable Size Spatial Cluster and D istributed Isolated Points Detection [ D ]. W uhan: W uhan Unversity, [. [ D ].. :, ] [ 20 ] Jiao L im in, L iu Yaolin, Ren Zhouqiao. Spatial points clustering based on self2organizing neural networks and its application [ J ]. Geomatics and Information Science of W uhan University, 2008, 33 (2) : [,,. [ J ]. :, 2008, 33 (2) : ]

SVM OA 1 SVM MLP Tab 1 1 Drug feature data quantization table

SVM OA 1 SVM MLP Tab 1 1 Drug feature data quantization table 38 2 2010 4 Journal of Fuzhou University Natural Science Vol 38 No 2 Apr 2010 1000-2243 2010 02-0213 - 06 MLP SVM 1 1 2 1 350108 2 350108 MIP SVM OA MLP - SVM TP391 72 A Research of dialectical classification