عنوان مقاله
نقشه‌برداری با رزولوشن-برتر از آب‌گرفتگی‌های تالاب براساس ادغام شبکه‌های عصبی پس‌انتشار و الگوریتم ژنتیک
فهرست مطالب
مقدمه
روش ها
مطالعه موردی
نتایج و بحث
نتیجه گیری
بخشی از مقاله
روشSAM-SWMI
روش SAMبراساس مقدار کسری در پیکسل‌های مجاور متعامل در برابر پیکسل‌های فرعی درون پیکسل‌های اصلی می‌باشند. یک پیکسل فرعی تنها توسط پیکسل‌هایی در حول پیکسل‌های اصلی جذب می‌شود که به معنی این است که حداکثر هشت پیکسل همسایه برای جذب بررسی شده‌اند.
کلمات کلیدی:
Super-resolution mapping of wetland inundation from remote sensing imagery based on integration of back-propagation neural network and genetic algorithm Linyi Li a, ⁎, Yun Chen b , Tingbao Xu c , Rui Liu b,d , Kaifang Shi b,d , Chang Huang e a School of Remote Sensing and Information Engineering, Wuhan University, 129 Luoyu Road, Wuhan 430079, PR China b CSIRO Land and Water Flagship, Clunies Ross Street, Canberra 2601, Australia c Fenner School of Environment and Society, The Australian National University, Linnaeus Way, Canberra 2601, Australia d Key laboratory of Geographic Information Science (Ministry of Education), East China Normal University, 500 Dongchuan Road, Shanghai 200241, PR China e College of Urban and Environmental Sciences, Northwest University, 1 Xuefu Road, Xi'an 710127, PR China article info abstract Article history: Received 14 January 2015 Received in revised form 3 April 2015 Accepted 9 April 2015 Available online xxxx Keywords: Wetland inundation Super-resolution mapping Intelligent algorithm integration Remote sensing imagery Mapping the spatio-temporal characteristics of wetland inundation has an important significance to the study of wetland environment and associated flora and fauna. High temporal remote sensing imagery is widely used for this purpose with the limitations of relatively low spatial resolutions. In this study, a novel method based on integration of back-propagation neural network (BP) and genetic algorithm (GA), so-called IBPGA, is proposed for super-resolution mapping of wetland inundation (SMWI) from multispectral remote sensing imagery. The IBPGA-SMWI algorithm is developed, including the fitness function and integration search strategy. IBPGA-SMWI was evaluated using Landsat TM/ETM+ imagery from the Poyanghu wetland in China and the Macquarie Marshes in Australia. Compared with traditional SMWI methods, IBPGA-SMWI consistently achieved more accurate super-resolution mapping results in terms of visual and quantitative evaluations. In comparison with GA-SMWI, IBPGA-SMWI not only improved the accuracy of SMWI, but also accelerated the convergence speed of the algorithm. The sensitivity analysis of IBPGA-SMWI in relation to standard crossover rate, BP crossover rate and mutation rate was also carried out to discuss the algorithm performance. It is hoped that the results of this study will enhance the application of median-low resolution remote sensing imagery in wetland inundation mapping and monitoring, and ultimately support the studies of wetland environment. © 2015 Elsevier Inc. All rights reserved. 1. Introduction Wetlands are areas where water is the primary factor controlling the environment and associated plant and animal life (Ramsar, 2009). They are cradles of biological diversity, providing water and primary productivity upon which species of plants and animals depend for survival (Ramsar, 2009). Wetlands experience periodic flood inundation which exhibits changes in spatial distribution and temporal duration (Zhao, Stein, & Chen, 2011). Spatio-temporal characteristics of inundation have been studied using multi-spatial, multi-temporal and multispectral remote sensing imagery (Chen, Barrett, et al., 2014; Chen, Cuddy, Wang, & Merrin, 2011; Chen, Huang, Ticehurst, Merrin, & Thew, 2013; Chen, Wang, et al., 2014; Huang, Chen, & Wu, 2014b; Huang, Chen, Wu, Chen, et al., 2014; Huang, Chen, Wu, & Yu, 2012; Huang, Peng, Lang, Yeo, & McCarty, 2014; Marti-Cardona, Dolz-Ripolles, & Lopez-Martinez, 2013; Ticehurst, Chen, Karim, & Dushmanta, 2013). However, the current remote sensing systems generally do not have high temporal and spatial resolutions at the same time (Huang, Chen, & Wu, 2014a; Li, Chen, Yu, Liu, & Huang, 2015). It is worth mentioning that the new sensor systems, specifically constellation systems such as RapidEye, are beginning to shift this paradigm. The current high temporal remote sensing imagery usually has relatively low spatial resolution (Huang, Chen, & Wu, 2014a; Li et al., 2015). The spatial resolution range of medium-low resolution remote sensing imagery here is 10 m–1000 m. The accuracy of wetland inundation mapping from high temporal remote sensing imagery is severely compromised due to spatial resolution constraints. One of the most popular methods to tackle this issue is super-resolution mapping. Super-resolution mapping, also termed as sub-pixel mapping, is designed to obtain more sub-pixel spatial information within mixed pixels based on the spatial dependence assumption that observations close together are more alike than those that are further apart (Aplin & Atkinson, 2001; Atkinson, 1997, 2005). There are many methods developed for super-resolution mapping. Atkinson (1997) proposed a method to allocate land cover class proportions to sub-pixels based on a distance measure (proximate sub-pixels contributing more than distant ones). Verhoeye and De Wulf (2002) explored a method in Remote Sensing of Environment 164 (2015) 142–154 ⁎ Corresponding author. Tel.: +86 13545177585. E-mail address: lilinyi@whu.edu.cn (L. Li). http://dx.doi.org/10.1016/j.rse.2015.04.009 0034-4257/© 2015 Elsevier Inc. All rights reserved.
