کاربرد شبکه عصبی مصنوعی و رگرسیون حداقل مربعات معمولی در مدلسازی تغییرات کاربری سرزمین

نوع مقاله: مقاله پژوهشی

نویسندگان

1 دانشجوی کارشناس‌ارشد GIS و سنجش از دور، دانشکدة علوم زمین، دانشگاه شهید چمران اهواز، اهواز، ایران

2 دانشیار، گروه GIS و سنجش از دور، دانشکدة علوم زمین، دانشگاه شهید چمران اهواز، اهواز، ایران

3 مربی، گروه GIS و سنجش از دور، دانشکدة علوم زمین، دانشگاه شهید چمران اهواز، اهواز، ایران

4 دانشجوی دکترای، گروه جنگلداری و اقتصاد جنگل، دانشکدة منابع طبیعی، دانشگاه تهران، کرج، ایرا

چکیده

با توجه به اهمیت بالای اثر تغییرات کاربری سرزمین در آینده، لازم است الگوی رشد و تغییر کاربری‌ها قبل از اتخاذ هر گونه تصمیمی به مسئولان و تصمیم‌گیرندگان امور مربوط ارائه شود. هدف این پژوهش مدل‌سازی تغییرات کاربری سرزمین در منطقة کوهمره سرخی استان فارس با استفاده از روش رگرسیون حداقل مربعات معمولی برای پیش‌پردازش متغیرها و مدل‌سازی با استفاده از شبکة عصبی است. بدین منظور نقشه‌های کاربری سرزمین با استفاده از تصاویر لندست در سال‌های 1366، 1379 و 1391 تهیه شد. سپس، صحت‌سنجی نقشه‌ها و آشکارسازی تغییرات انجام شد. نتایج آشکارسازی تغییرات دورة اول (1366-1379) با ضریب کاپای 83% نشان داد بیشترین افزایش مساحت در ناحیة مرتع (24/4224 هکتار) و بیشترین کاهش مساحت در ناحیة جنگل (75/3953 هکتار) رخ داده است. بر مبنای این تغییرات و انتخاب بهترین ترکیب برای متغیرها، مدل‌سازی پتانسیل تبدیل کاربری برای سال 1391، با استفاده از روش شبکة عصبی پرسپترون چندلایه انجام شد. سپس، با روش زنجیرة مارکوف، نقشة کاربری سرزمین برای سال 1391 پیش‌بینی شد. نتیجة ماتریس خطا بین نقشة حاصل از مدل‌سازی و نقشة کاربری سرزمین سال 1391، ضریب کاپای 75% است. در مرحلة بعد، نتایج آشکارسازی تغییرات دورة دوم (1379-1391) با ضریب کاپای 88% نشان داد بیشترین افزایش مساحت در ناحیة مرتع (82/1871 هکتار)، همچنین بیشترین کاهش مساحت در ناحیة جنگل (05/3082 هکتار) رخ داده است. با توجه به تغییرات دورة دوم، نقشة کاربری سرزمین برای سال 1403 پیش‌بینی شد که بیشترین تغییر کاربری نسبت به سال 1391، در ناحیة کشاورزی آبی خواهد بود.

کلیدواژه‌ها


عنوان مقاله [English]

Application of artificial neural network and ordinary least squares regression in modeling land use changes

نویسندگان [English]

  • sara azizi ghalaty 1
  • Kazem Rangzan 2
  • Ayub Taghizadeh 3
  • Shahram Ahmadi 4
1 MSc. Student of GIS and Remote Sensing, Shahid Chamran University, Ahvaz, I.R. Iran
2 Associate Professor of GIS and Remote Sensing, Shahid Chamran University, Ahvaz, I.R. Iran
3 Instructor of GIS and Remote Sensing, Shahid Chamran University, Ahvaz, I.R. Iran
4 PhD. Candidate, College of Agriculture and Natural Resource, University of Tehran, Karaj, I.R. Iran
چکیده [English]

Owing to the vital effects of future land use changes, it is necessary to predict land use growth pattern before any decision making by the authorities and decision makers. Purpose of this research is to model land use change of Kohmare Scorch plain of Shiraz province using Ordinary Least Squares regression (OLS) for pre-processing variables and Modeling using neural networks. To perform this model, the land use maps using Landsat images of the years 1987, 2000 and 2012 were prepared. Next, the validation of classified images and change detection analysis performed. Results of change detection between 1987 and 2000 with accuracy of 83% kappa, shows the greatest increase in rangeland area (4224.24 ha) and the greatest decrease was on forest area (3953.75). Considering these changes, selection of the best combination of explanatory variables, potential land use changes for year 2012 was performed using multi-layer perceptron algorithm of artificial neural network. Next, using Markov chain method the land use map for 2012 was predicted. The error matrix for modeled land use map and that of Landsat image of year 2000 is 75%. Next, the revealed changes for the second period (2000-2012) with Kappa of 88% show greatest increase for rangeland area (1807.02ha). In contrast the greatest decrease was for forest (2132.82). Considering change detection at second period, land use for year 2024 was predicted and result shows that irrigated agriculture would have the greatest change.
 

کلیدواژه‌ها [English]

  • land use change
  • modeling
  • multi-layer perceptron neural network
  • ordinary Least Squares
 
[1]. Koomen, E., Stillwell J., Bakema A., and Scholten H.J. (2007). Modelling Land-Use Change, Progress and Applications, Springer, Dordrecht, the Netherlands.
[2]. Niyazi, Y., Ekhtesasi, M., and Hosseini, Z. (2010). Comparison of two method classification of maximum likelihood and artificial neural network to extract the land use map (Case Study: catchment of Ilam dam). Geography and Development Journal, 20: 132-119.
[3]. Ademola N., Braimoh K., and Onishi T. (2007). Spatial determinants of urban land use change in Lagos. Land Use Policy, 24(2): 502–515.
[4]. Ashraf, M.D., and Yasushi, Y. (2009). Land use and land cover change in Greater Dhaka, Bangladesh: Using remote sensing to promote sustainable urbanization. Applied Geography, 29: 390–401.
[5]. Alizadeh Rabi'i, A. (2008). Remot Sensing (Principles and Application), Publications of SAMT, Tehran.
[6]. Lu, D., Mausel, P., Brondizio, E., and Moran, E. (2004). Change detection techniques. International Journal of RemoteSensing, 25(12): 2365-2401.
[7]. Kamyab, H., Salman Mahini, A., Hosseini, S.M., and Gholamalifard, M. (2011). Application of neural network in modeling urban development (Case Study: Gorgan City). Human Geographical Journal, 76: 113-99.
[8]. Pijanowski B., Pithadia S., Shellito B.y., and Alexandridis K. (2005). Calibrating a neuralnetwork-based urban change model for two metropolitan areas of the UpperMidwest of the United States. International Journal of Geographical InformationScience, 19(2): 197–215.
[9]. Mas J.F., Puig H., Palacio J.L., and Sosa-Lopez A. (2004). Modelling deforestation using GIS and artificial neural networks. Environmental Modelling and Software, 19(5): 461–471.
[10]. Pontius R., and Malanson J. (2005). Comparison of the structure and accuracy of twoland change models. International Journal of Geographical Information Science, 19(2): 243–265.
[11]. Van Vliet, J., White, R., and Dragicevic, S. (2009). Modeling urban growth using a variablegrid cellular automaton. Computers Environment and Urban Systems, 33(1): 3543.
[12]. Pijanowski, B., Brown, D., Shellito, B., and Manik, G. (2002). Using Neural networks and GIS to forecast land use changes: a land transformation model. Computers, Environment and Urban Systems, 26(6): 553-575.
[13]. Vahidnia, M.H., Aleshikh, A.A., and Varshosaz, M. (2010). The assessment of multilayer perceptron networks with the existing model interpolation. Spatial Information Technology Engineering Journal, 1(1): 95-116.
[14]. Chau, K.W., and Cheng, C.T. (2002). Real-time prediction of water stage with artificial neural network approach. Lecture Notes in Artificial Intelligence, 2557: P.715.
[15]. Foody, G.M. (2000). Mapping land cover from remotely sensed data with a softed feed forwardneural network classification. Journal of Intelligent and Robotic Systems, 29(4) 443-449.
[16]. Kamyab, H., Salman Mahini, A., Hosseini, S. M., and Gholamalifard, M. (2010). Adopting approach based on information using logistic regression modeling urban development Gorgan city. Journal of Ecology, 36 (54): 96-89.
[17]. Zeyaeian, P., Shakiba, A., Matkan, A.A., and Sadeghi, A. (2009). Remote sensing (RS), geographic information systems (GIS) and cellular automata model (CA) as a tool for simulating urban land use change (case study: the Shahre kord City). Journal of Environmental Sciences, 7(1): 148-133.
[18]. Ma C., Zhang G.Y., Zhang X.C., Zhao Y.J., and Li H.Y. (2012). Application of Markov model in wetland change dynamics in Tianjin Coastal Area. China Procedia Environmental Sciences, 13: 252 – 262.
[19]. Pérez-Vega A, Mas J.F., and Ligmann-Zielinska A. (2012). Comparing two approaches to land use/cover change modeling and their implications for the assessment of biodiversity loss in a deciduous tropical forest. Environmental Modelling & Software, 29: 11-23.
[20]. Ghorbani, Kh. (2012). Geographically weighted regression: a method for mapping the samerainfall in Gilan province. Journal of Soil and Water, 3:752-743.
[21]. Wooldridge J.M. (2003). Introductory Econometrics: A Modern Approach. South-Western college Publishing, Mason, Ohio.
[22]. Singh V., Dubey A. (2012). Land use mapping using remote sensing & GIS techniques in Naina-Gorma Basin, Part of Rewa District. International Journal of Emerging Technology and Advanced Engineering, 2(11): 151-156
[23]. Zare Garizi, A., Bardi Sheykh, V., Saedaldein, A., and Salman Mahini, A. (2012). Application logistic regression method in modeling spatial pattern of vegetation change (Case Study: catchment of Chehel Chai of Golestan province). Journal of Geographical Space, 12 (37): 68-55.
[24]. Lu, D., and Weng, Q. (2007). A Survey of Image Classification Methods and Techniques for Improving Classification Performance. International Journal of Remote Sensing, 28(5): 823-870.
[25]. Alavipanah, S.K. (2009). Application of Remote Sensing in the Earth Sciences (Soil), University of Tehran press, Tehran.
[26]. Şatır O., and Berberoğlu S. (2012). Land Use/Cover Classification Techniques Using Optical Remotely Sensed Data in Landscape Planning, Dr. Murat Ozyavuz (Ed.), InTech, Turkey Published.
[27]. Verburg, P., Schot, P., Dijst, M., and Veldkamp, A. (2004). Land use change modelling: current practice and research priorities. GeoJournal, 61: 309-324.
[28]. Costanza R., and Rrut M. (2004). Using dynamic modeling to scope environmental problems and build consensus, GeoJourna, l61: 309–324.
[29]. Eastman, J.R., Van Fossen, M.E., and Solarzano, L.A. (2012). Transition potential modelingfor land cover change. In: Maguire, D., Goodchild, M., Batty, M. (Eds.), GIS,Spatial Analysis and Modeling. ESRI Press, Redlands, California.
[30]. Václavík, T., and Rogan, J. (2009). Identifyng trends in land Use/Land cover changes in thecontext of Post-Socialist Transformation in Central Europe. GIS Science andRemote Sensing, 49(1): 1-32.
[31]. Gontier, M., Mörtberg, U. and Balfors, B. (2009). Comparing GIS based habitat models forapplications in EIA and SEA. Environmental Impact Assessment Review, 30(1): 8-18.
[32]. Mas, J.F., and Flores, J.J. (2008). The application of artificial neural networks to the analysis of remotely sensed data (review article). International Journal of Remote Sensing, 29(3): 617-663.
[33]. Coppin, P., Jonckheere, I., Nackaerts, K. and Muys, B. (2004). Digital change detection methods in ecosystem monitoring. International Journal of Remote Sensing, 25 (9), 1565–1596.
[34]. Safyanian, A. (2009). Surveyofland use Changes in ESFAHAN city using vector change detection techniques during the years 1366 to 1377. Journal of Soil and Water Sciences, 13(49): 152-164.
[35]. Rabei’I, H., Zeyaeian, P. and Alimohamadi, A. (2004). Detecting and recovering Changes in land use and land cover in ESFAHAN city with the help of remote sensing and GIS. Geographical Research Quarterly, 84:41-54.
[36]. Gholamalifard, M., Jorabiyan Shoshtary, Sh., Hosseini Kahnoj, S.H., and Mirzaei, M. (2012). Modeling of land use changes in coastal of MAZANDARAN province using LCM in GIS. Journal of Environmental Studies, 38(4): 124-109.
[37]. Zhang G.P. (2003). Neural Networks in Business Forecasting, Idea Group Inc. 310p.
[38]. Huang W., Wang Sh, YuL., Bao Y., and Wang L. (2006). A New Computational Method of Input Selection for Stock Market Forecasting with Neural Networks.  Part IV, 308–315.
[39]. Stone M., and Brooks, R.J. (1990). Continuum regression: cross-validated sequentially constructed prediction embracing ordinary least squares, partial least squares and principal components regression. Journal of the Royal Statistical Society, 2: 237-269.
[40]. Ghabaei Sogh, M., Mosaedi, A., Hesam, M., and Hezarjaribi, A. (2010). Assessment effect of pre-processing parameters input to artificial neural networks (ANNs) using the step by step regression and Gamma test methods in order to faster estimate daily evapotranspiration. Journal of Soil and Water, 3: 624-610.
[41]. Farahani, H.A., Rahiminezha, A., Same, L., and Immannezhad, K. (2010). A Comparison of partial least squares (PLS) and ordinary least squares (OLS) regressions in predicting of couples mental health based on their communicational patterns. Procedia Social and Behavioral Sciences, 5: 1459–1463.
[42]. Eastman, J.R. (2009(. IDRISI Taiga Guide to GIS and Image Processing. Clark-Labs, Clark University, Worcester,MA.
[43]. Eastman, J.R., (2006). IDRISI Andes. Tutorial. Clark-Labs, Clark University, Worcester,MA.
[44]. Pistocc hi, A., Lu zi L., and Napolitano P. (2002). The use of predictive modeling techniques for optimal exploitation of spatial databases: a case study in landslide hazard mapping with expert system-like methods. Environmental Geology, 41(765):1-24.
[45]. Chuvieco E., (2002). Teledetección ambiental: La observación de la Tierra desde especial. Editorial Ariel. Barcelona, Espuma.
[46]. Bishop C.M., (1995). Neural Networks for Pattern Recognition. Oxford University Press, Oxford.
[47]. Samanta, B., Bandopadhyay, S., and Ganguli, R., (2006). Comparative Evaluation of Neural Network Learning Algorithms for Ore Grade Estimation. Mathematical Geology, 38: 175-197.
[48]. Salmanmahiny, A. and Kamyab, H. (2012). Applied Remote Sensing and GIS with Idrisi.Mehrmahdis publisher. 2nd Edition.
[49]. Fan, F., Wang, Q., and Wang, Y. (2007). Land use and land cover change in Guangzhou, Chaina, from 1998 to 2003, based on land sat TM/ETM+ imagery, Sensors, 7: 1323-1342.
[50]. Fatemi, S.B., and Rezai, Y. (2012). Principles of Remot Sensing. Publication of Azadeh, Tehran, 288P.
[51]. Rosenshein, L., Scott, L., and Pratt, M. (2011). Finding a Meaningful Model, ArcUser, 40-45.
[52]. Babaei Eghdam, F., Esmaei Ouri, A., and Heidari Sariyan, V. (2011). Modeling the spatial pattern of land use SAREEN city in 1400 year using CLUE_S model. Geographical Research, 26(4): 93-116.
[53]. Bakhtiyarifar, M., Mesgari, M.S., Kaeimi, M., and Chaharghani, A. (2011). Modeling of land use change using Multi Criteria Decision Methods and GIS. Journal of Ecology, 37(58): 57-43.