عنوان مقاله [English]
Topography is a factor controlling the spatialdistribution of soil moisture, vegetation, soil salinity, soil texture andso on. It has an important role in changing thecharacteristics of the soil and hydrological processes. In recent yearsthe topographyhave been used as an important factor forpredicting the properties of soil, climate, geology, etc. According tothe importance of topography to extract different information, use ofsatellite images with high spatial resolution seems very necessary. Digitalelevation models (DEM) have become a widely used tool andproduct in the last 20 years. They provide a snapshot of the landscape and landscape features while also providingelevation values. They have allowed us to better visualize andinterrogate topographic features. In addition to increasing the spatial resolution, information of the digital elevation model (DEM) that isthe most important issues in quantitative geomorphology have increased. In orderto increase the spatial resolution several modelshave been proposed. Among the models, the attraction model as the newest modelhas very high accuracy. The sub-pixel attraction models convertthe pixel towards sub-pixels based on the fraction valuesin neighboring pixels that can be attracted only by centralpixel. Based on this approach only a maximum of eightneighboring pixels can be selected for the attraction. In themodel, other pixels are supposed to be far from thecentral pixel to have any attraction. In this study byusing sub-pixel attraction model, the spatial resolution of digitalelevation models (DEM) was increased (Sub-pixel mapping technology is apromising method of increasing the spatial resolution of the classificationresults derived from remote sensing imagery). The design of thealgorithm is accomplished by using digital elevation model (DEM) withspatial resolution of 30 m (Advanced Space borne Thermal Emissionand Reflection Radiometer (ASTER)) and 90 m (Shuttle Radar TopographyMission (SRTM)). This study was carried out in the EastMount Sahand, Iran is located at the longitude of N 37° 31َto 37° 30َand latitude of E 45° 55َto 45° 58َ. It is expected that usingattraction model increasesthe spatial resolution of DEM. The attraction model does not need any calibration and training similar to the machine learningalgorithms. So, to run the algorithm in the model, the computing time was reduced. In attraction model, scale factors of (2, 3 and 4) with two neighboring methods of touching andquadrant are applied to DEMs using Matlab software and thenusing RMSE (Root mean square error), determined the best model. The algorithm is evaluated using 2118 sample points that aremeasured by surveyors. As the result of Root mean squareerror (RMSE), it showed that the spatial attraction model withscale factor of (S=2 and T=2) for digitalelevation model (DEM) 30m and digital elevation model (DEM) 90mgives better results compared to scale factors that are greaterthan 2 and also touching neighborhood method proved to bemore accurate than quadrant. In fact, subtracting each pixel tomore than two sub-pixels caused to decrease the accuracyof resulted DEM which makes the value ofroot mean square error (RMSE)to increase and showed that attraction modelscould not be used for S which is greater than 2. So, according to the results, it is suggested that themodel to be used for increasing spatial resolution of DEM in the studiescatchment. Comparing the digital elevation model (DEM) as inputsin the attraction models determined that digital elevation model (DEM) 30 m (root mean square error < 5.54) has better spatialresolution than digital elevation model (DEM) 90 m (root meansquare error = 9.13) to find the best model for increasingspatial resolution. The results showed that by using the method, thespatial resolution of digital elevation model (DEM) with lower timeand cost could be increased. Digital elevation model (DEM) mapwith high resolution as a base can be used for findingmore information from the Earth surface. For different study such asamount of vegetation, temperature, rainfall and hydrological status the results of sub-pixel attractions on digital elevation model (DEM) can be used and more details of study area could be found. Therefore, it issuggested that the same researches should be done in other areas withdifferent topographic and geographical conditions in order to confirm theresults of this study.
1. Ardila, J.P., Tolpekin, V.A., Bijker, W., Stein, A. (2011). Markov-random-ﬁeld-based super-resolution mapping for identiﬁcation of urban trees in VHR images, ISPRS J. Photogram. Remote Sens. 66: 762–775.
2. Atkinson, P.M. (2005). Sub-pixel target mapping from soft-classiﬁed, remotely sensed imagery, Photogramm. Eng. Remote Sens. 71 (7): 839–846.
3. Boucher, P.C. Kyriakidis. (2006). Super-resolution land cover mapping with indicator geostatistics, Remote Sens. Environ. 104 (3): 264–282.
4. Boucher, P.C. Kyriakidis, C.C. Ratcliff. (2008). Geostatistical solutions for super resolution land cover mapping, IEEE Trans. Geoscience Remote Sensing. 46 (1): 272–283.
5. Grabs, T., Seibert, J., Bishop, K., and Laudon, H. (2009). Modeling spatial patterns of saturated areas: A comparison of the topographic wetness index and a dynamic distributed model. J. Hydrol. 373: 15-23.
6. Janzen, H.H., Ellert, B.H., and Anderson, D.W. (2002). Organic matter in the landscape. P 905-909, In: Lal, R. (Ed.), Encyclopedia of Soil Science. Marcel Dekker, Inc. New York.
7. Kasetkasem, T., Arora, M.K., Varshney, P.K. (2005). Super-resolution land cover mapping using a Markov random ﬁeld based approach, Remote Sens. Environ. 96 (3/4):302–314.
8. Mertens, K.C., Baets, B.D., Verbeke, L.P.C., Wulf, R.D. (2006). A sub-pixel mapping algorithm based on sub-pixel/pixel spatial attraction models, Int. J. Remote Sens. 27 (15): 3293–3310.
9. Mertens, K.C., Verbeke, L.P.C., Ducheyne, E.I., Wulf, R.D. (2014). Using genetic algorithms in sub-pixel mapping, Int. J. Remote Sens. 24 (21): 4241–4247.
10. Muad, A.M., Foody, G.M. (2012). Super-resolution mapping of lakes from imagery with a coarse spatial and ﬁne temporal resolution, Journal of Applied Earth Obs. Geoformation. 12: 79–91.
11. Sorensen, R., Zinko, U., and Seibert, J. (2005). On the calculation of the topographic wetness index: evaluation of different methods based on field observation. Hydrology and Earth System Sciences. 10: 101-112.
12. Starr, G.C., Lal, R., Malone, R., Hothem, D., Owens, L., and Kimble, J. (2000). Modeling soil carbon transported by water erosion processes. Land Degradation and Development. 11: 83-91.
13. Tatem, A, J., Lewis, H.G., Atkinson, P.M., Nixon, M.S. (2001). Super-resolution target identiﬁcation from remotely sensed images using a Hopﬁeld neural network, IEEE Trans. Geoscience Remote Sensing. 39 (4):781–796.
14. Tolpekin, V.A., Stein, A. (2009). Quantiﬁcation of the effects of land-cover-class spectral reparability on the accuracy of markov-random-ﬁeld-based super resolution mapping, IEEE Trans. Geosci. Remote Sens. 47 (9): 3283–3297.
15. Verhoeye, R.D. Wulf. (2002). Land cover mapping at sub-pixel scales using linear optimization techniques, Remote Sens. Environ. 79 (1): 96–104.
16. Wu, K., Zhang, L.P., Niu, R.Q., Du, B., Wang, Y. (2011). Super-resolution land-cover mapping based on the selective endmember spectral mixture model in hyperspectral imagery, Opt. Eng. 50 (12) 126201.
17. Wang, L.G., Wang, Q.M., Liu D.F. (2011). Sub-pixel mapping based on sub-pixel to subpixel spatial attraction model, in: Proceedings of the 2011 IEEE International Geoscience and Remote Sensing Symposium, IGARSS, 593–596.