Application of neural network swarm optimization for paddy-field classification from remote sensing data

Monitoring changes in paddy areas is important for economic and environmental research, since rice is a staple food in Asia and paddy agriculture is a major cropping system. Recently, remote sensing has been used to observe changes in the areas of paddy. However, monitoring paddy areas by remote sensing is difficult owing to the temporal changes in paddy, and the differences in the spatiotemporal characteristics of paddy agriculture between countries or regions. In our previous research using a multilayered perceptron and spatiotemporal satellite sensor data, the proposed classifier yielded a correct classification rate of 90.8%. In this article, we proposed a cooperative learning method using particle swarm optimization as the global search method and a multilayered perceptron as the local search method in order to improve the classification accuracy for practical use.     

Artificial neural networks paddy-field classifier using spatiotemporal remote sensing data

Monitoring changes in a paddy-field area is important since rice is a staple food and paddy agriculture is a major cropping system in Asia. For monitoring changes in land surface, various applications using different satellites have been researched in the field of remote sensing. However, monitoring a paddy-field area with remote sensing is difficult owing to the temporal changes in the land surface, and the differences in the spatiotemporal characteristics in countries and regions. In this article, we used an artificial neural network to classify paddy-field areas using moderate resolution sensor data that includes spatiotemporal information. Our aim is to automatically generate a paddy-field classifier in order to create localized classifiers for each country and region.     

基于特征波段的SPOT-5卫星影像水稻面积信息自动提取的方法研究

为了快速、准确地从遥感影像上提取水稻信息,满足国家农情遥感监测系统要求,以黑龙江省852农场水稻提取为例,利用SPOT-5卫星影像数据,分析了水稻和其它背景地物的光谱特征,发现利用原有波段难以提取复杂的水稻信息,因此利用植被特征波段：归一化植被指数（NDVI）作为新波段融入原始影像中,在增加有效信息量的同时运用简单决策树模型提取水稻信息,并参照地块现状矢量图进行精度评价。结果表明,该方法的总体提取效果较好,其提取精度与通常的监督分类方法相比有了较大的提高,只是在水稻和玉米交界处有误判现象。     

Evaluation of optical satellite remote sensing for rice paddy phenology in monsoon Asia using a continuous in situ dataset

In monsoon Asia, optical satellite remote sensing for rice paddy phenology suffers from atmospheric contaminations mainly due to frequent cloud cover. We evaluated the quality of satellite remote sensing of paddy phenology: (1) through continuous in situ observations of a paddy field in Japan for 1.5 years, we investigated phenological signals in the reflectance spectrum of the paddy field; (2) we tested daily satellite data taken by Terra/Aqua MODIS (MOD09 and L1B products) with regard to the agreement with the in situ data and the influence of cloud contamination. As a result, the in situ spectral characteristics evidently indicated some phenological changes in the rice paddy field, such as irrigation start, padding, heading, harvest and ploughing. The Enhanced Vegetation Index (EVI) was the best vegetation index in terms of agreement with the in situ data. More than 65% of MODIS observations were contaminated with clouds in this region. However, the combined use of Terra and Aqua decreased the rate of cloud contamination of the daily data to 43%. In conclusion, the most robust dataset for monitoring rice paddy phenology in monsoon Asia would be daily EVI derived from a combination of Terra/MODIS and Aqua/MODIS.     

Linking physical remote sensing models with crop growth simulation models,applied for sugar beet

Abstract

In recent years, remote sensing and crop growth simulation models have become increasingly recognized as potential tools for growth monitoring and yield estimation of agricultural crops. In this paper, a methodology is developed to link remote sensing data with a crop growth model for monitoring crop growth and development. The Cloud equations for radar backscattering and the optical canopy radiation model EXTRAD were linked to the crop growth simulation model SUCROS: SUCROS-Cloud-EXTRAD. This combined model was initialized and re-parameterized to fit simulated X-band radar backscattering and/or optical reflectance values, to measured values. The developed methodology was applied for sugar beet. The simulated canopy biomass after initialization and re-parameterization was compared with simulated canopy biomass with SUCROS using standard input, and with measured biomass in the field, for 11 fields in different years and different locations. The seasonal-average error in simulated canopy biomass was smaller with the initialized and re-parameterized model (225-475 kg ha^?1), than with SUCROS using standard input (390-700 kg ha^?1), with ‘end-of-season’ canopy biomass values between 5500 and 7000kgha^?1. X-band radar backscattering and optical reflectance data were very effective in the initialization of SUCROS. The radar backscattering data further adjusted SUCROS only during early crop growth (exponential growth), whereas optical data still adjusted SUCROS until late in the growing season (at high levels of leaf area index (LAI), 3-5).     

A semi‐empirical backscattering model for estimation of leaf area index (LAI) of rice in southern China

Most paddy rice in southern China grows in warm, humid and rainy areas where it is hard to acquire optical remote sensing data. In this study, a semi‐empirical backscattering model was proposed to estimate leaf area index (LAI) of rice in the area using ENVISAT Advanced Synthetic Aperture Radar (ASAR) alternating polarization data. Ground measurements of LAI, water content and height of rice in the test site were collected and the model fitted at the same time as the acquisition of ASAR data. LAI estimated from the model was compared with ground measurements to evaluate the accuracy of the model. The results showed that the model provides a promising alternative to optical remote sensing data for predicting LAI of rice in southern China.     

Data Assimilation Experiment ofPoor Quality of RemoteSensing Image in Critical Phenology

In the process of data assimilation，influenced by the water vapor and cloud cover in the study area，the qualities of remote sensing images are poor in the key crop phenological phases.This will cause not to get the perfect remote sensing images for a long time.So we try to solve this problem by using an improved EnKF method to assimilate the WOFOST crop growth model and the terrible quality of remote sensing images to forecast the maize’s yield in the Red Star Farm in Heilongjiang province.In order to improve the accuracy of simulated time series curve of the LAI and yield production results，the consideration on quality evaluation of the remote sensing images is introduced by using expansion coefficient and adjustable factor.The results shows based on the improved EnKF method，time series curve of the LAI keeps a normal tendency of LAI rather than negative fluctuations，and it also avoids the serrated fluctuation to a certain extent.In addition，compared with the original EnKF method，in the field level R2 can increased to 0.67 from 0.59，RMSE is reduced to 92.23 kg/hm2 from 240.57 kg/hm2 and in the farm level R2 can increased to 0.61 from 0.52，RMSE is reduced to 122.44 kg/hm2 from 310.94 kg/hm2 between simulated yield and measured yield.     

Towards an operational system for regional-scale rice yield estimation using a time-series of Radarsat ScanSAR images

This paper demonstrates that Radarsat ScanSAR data can be an important data source of radar remote sensing for monitoring crop systems and estimation of rice yield in large areas in tropical and sub-tropical regions. Experiments were carried out to show the effectiveness of Radarsat ScanSAR data for rice yield estimation in the whole province of Guangdong, South China. A methodology was developed to deal with a series of issues in extracting rice information from the ScanSAR data, such as topographic influences, levels of agro-management, irregular distribution of paddy fields and different rice cropping systems. A model was provided for rice yield estimation based on the relationship between the backscatter coefficient of multi-temporal SAR data and the biomass of rice. The study indicates that the whole procedure can become a low-cost and convenient operational system for large-scale rice yield estimation which is difficult for conventional methods.     

基于GF-1影像的耕地地块破碎区水稻遥感提取

耕地地块破碎区水稻遥感提取是作物监测研究的热点问题之一。以苏州市高新区为例，通过挖掘关键物候期水稻与下垫面水体光谱特征组合差异，基于分蘖期与齐穗期两景16 m分辨率的GF-1 WFV数据，构建归一化差值植被指数（NDVI）差值法、归一化水体指数和比值植被指数（NDWI-RVI）差值法提取水稻分布，并深入探究了水稻面积提取精度及空间重合度影响因素。结果显示：与非监督分类和监督分类方法相比，植被指数差值法水稻识别精度贡献率可提升30%以上，NDVI差值法提取水稻种植面积的精度、空间重合度、制图总体精度和Kappa系数分别为86.2%、66.1%、92.2%和0.72；NDWI-RVI差值法上述指标分别高达95.5%、78.4%、93.5%和0.846，实现了利用少量中高分辨率遥感影像精确提取耕地地块破碎区水稻分布的目的，可实际服务于太湖地区农业生产及相关决策支持。     

基于GF-1影像的耕地地块破碎区水稻遥感提取

耕地地块破碎区水稻遥感提取是作物监测研究的热点问题之一。以苏州市高新区为例,通过挖掘关键物候期水稻与下垫面水体光谱特征组合差异,基于分蘖期与齐穗期两景16 m分辨率的GF-1 WFV数据,构建归一化差值植被指数（NDVI）差值法、归一化水体指数和比值植被指数（NDWI-RVI）差值法提取水稻分布,并深入探究了水稻面积提取精度及空间重合度影响因素。结果显示：与非监督分类和监督分类方法相比,植被指数差值法水稻识别精度贡献率可提升30%以上,NDVI差值法提取水稻种植面积的精度、空间重合度、制图总体精度和Kappa系数分别为86.2%、66.1%、92.2%和0.72;NDWI-RVI差值法上述指标分别高达95.5%、78.4%、93.5%和0.846,实现了利用少量中高分辨率遥感影像精确提取耕地地块破碎区水稻分布的目的,可实际服务于太湖地区农业生产及相关决策支持。     

Multi-year monitoring of rice crop phenology through time series analysis of MODIS images

Precise phenological calendars, for each cultivated species and variety, are necessary both to highlight anomalous agronomic situations and to feed crop models. This study, conducted in the Italian rice area, focuses on the evaluation of the contribution of remote sensing satellite data to providing phenological information on rice cropping systems. A time series of 5 years (2001–2005) of the Moderate Resolution Imaging Spectrometer (MODIS) Normalized Difference Vegetation Index (NDVI) 16-day composites was analysed with the TIMESAT program in order to automatically retrieve key phenological information such as the start of season (emergence), peak (heading) and end of season (maturity). The procedure involved two steps: (1) interpolation and smoothing of MODIS NDVI temporal profile and (2) the analysis of a temporal signal for the extraction of the phenological metrics. The remote sensing estimates were evaluated using information regarding cultivated variety, sowing dates, management and production directly acquired from rice farmers. A good correlation (R ²?=?0.92, n?=?24) has been observed between estimates derived from satellites and estimates produced with the traditional Growing Degree Days (GDD) method based on thermal unit accumulation. Improved estimates of the maturity stage were obtained using a procedure that integrates satellite and GDD methods; however its application requires spatially distributed information on the cultivated varieties. Satellite derived maps of the retrieved phenological parameters showed an intra-seasonal pattern related to different cultivated varieties. Inter-seasonal analysis allowed the anomalous behaviour of the year 2003 to be highlighted, characterized by rapid growth at the beginning of the spring and an early senescence. The results confirm the potential of remotely sensed data for the monitoring of crop status and for the forcing of crop models in a spatially distributed way.     

基于多时相Landsat8 OLI影像的作物种植结构提取

针对基于多时相遥感影像、多种特征量提取多种作物种植结构在我国研究较少的现状,利用多时相Landsat8OLI影像数据,根据温宿县不同作物的农事历,通过分析主要地物的光谱特征和归一化植被指数的时间变化信息,构建不同作物种植结构提取的决策树模型,实现了对温宿县多种作物种植结构信息的提取。结果表明:1水稻的最佳识别依据是5月20日影像的近红外波段和7月23日影像的NDVI值;棉花和春玉米的最佳识别依据是5月20日~9月9日影像的NDVI变化值;冬小麦—夏玉米和林果的最佳识别依据是5月20日~7月23日影像的NDVI变化值;2与单时相监督分类相比,多时相决策树法对多种作物种植结构的提取效果更理想,总体精度提高了7.90%,Kappa系数提高了0.10;3Landsat8OLI影像数据分辨率高、成本低、获取方便,是农作物遥感的良好数据源。     

Scaling of coastal phytoplankton features by optical remote sensors: comparison with a regional ecosystem model

Different scales of hydrological and biological patterns of the Bay of Biscay are assessed using space‐borne and airborne optical remote sensing data, field measurements and a 3‐dimensional biophysical model. If field measurements provide accurate values on the vertical dimension, ocean colour data offer frequent observations of surface biological patterns at various scales of major importance for the validation of ecosystem modelling. Although the hydro‐biological model of the continental margin reproduces the main seasonal variability of surface biomass, the optical remote sensing data have helped to identify low grid resolution, input inaccuracies and neglect of swell‐induced erosion mechanism as model limitations in shallow waters. Airborne remote sensing is used to show that satellite data and field measurements are unsuitable for comparison in the extreme case of phytoplankton blooms in patches of a few hundred metres. Vertically, the satellite observation is consistent with near surface in situ measurements as the sub‐surface chlorophyll maximum usually encountered in summer is not detected by optical remote sensing. A mean error (δC/C) of 50.5% of the chlorophyll‐a estimate in turbid waters using the SeaWiFS‐OC5 algorithm allows the quantitative use of ocean colour data by the coastal oceanographic community.     

高性能遥感图像处理与空间信息网格建模

将空间信息网格技术（SIG）应用于遥感图像处理中，就是利用网格计算的特点，来解决遥感图像数据以及处理算法资源的共享、海量遥感图像数据的实时快速处理等问题。文章以此为主线，在分析高性能遥感图像处理问题的基础上，探讨了在网格环境下构造高性能遥感图像处理系统的可行性及相关关键技术的实现，提出了基于开放网格服务体系结构的系统模型，并实现了原型系统。实验表明，SIG用于遥感图像处理是可行的，并取得了一定的成果。     

基于HJ-1B卫星遥感数据的水稻识别技术研究

为快速、准确地在遥感图像上识别水稻作物的信息,满足县级尺度水稻遥感监测的需要,以野外实地调查资料、1∶5万地形图数据为辅助,通过光谱分析法,分析研究HJ-1B星CCD数据的水稻作物的光谱反射特性,建立水稻作物遥感信息识别模型。采用决策树分类方法提取水稻作物信息,并将该技术方法应用于广西宾阳县水稻作物信息提取研究。采用实测样地数据,利用混淆矩阵进行精度评价验证,总精度为94.9%,Kappa系数为0.8533。研究表明,该水稻作物的识别技术,可以为了解我国水稻种植情况,进行水稻长势监测和产量估测提供技术参考。     

运用NOAA AVHRR和Landsat TM数据估算多年水稻种植面积

介绍了综合运用NOAA AVHRR和Landsat TM数据进行多年水稻种植面积监测的一种方法,以湖北省为例,首先运用Landsat TM数据计算了该省1992年的水稻种植面积;接着运用1992年和1994年的NOAA AVHRR数据分别计算这两年的水稻像元数,以这两年水稻像元数的变化来反映水稻种植面积的变化;最后运用线性模型,估算1994年的水稻种植面积。所得的1994年水稻种植面积与湖北省农调队资料相比精度为84.5%。运用同样的方法估算1995年该省的水稻种植面积,精度达90%以上。     

基于网格环境的遥感图像并行处理

对于海量遥感数据处理而言,关键是如何处理大量的网络计算,借助网格整合异构计算资源的优势,设计开发适合遥感图像处理的网格计算环境和算法。本文结合实际研究,以Globus为网格中间件,CSF4为元调度器构建网格环境,依据MPICH-G2编程模式,以遥感图像增强中的平滑处理为例,在网格环境下实现遥感图像并行化处理,并处理效率进行了对比分析。     

基于卫星遥感预测作物成熟期的可行性分析

精准收获是精准农业的重要环节,首先分析了收获时间对作物产量与品质的影响,论述了作物成熟期监测的重要性,然后从气象统计模型、作物生长模型及遥感监测3个方面回顾了作物成熟期预测的研究进展。在此基础上,通过对目前主要作物成熟状态指示因子遥感监测研究进展的分析,认为在当前新型传感器不断涌现的条件下,利用卫星遥感预测大范围作物成熟期、制订收割顺序并指导农业生产的条件已经成熟。并指出研究面向遥感的作物成熟期指示因子及其变化规律,发展高精度的作物冠层叶绿素及水分含量的遥感估算方法,研究面向农田尺度动态监测的高时空分辨率数据集构建技术和多种模型的耦合将成为该领域未来的研究重点。     

利用中巴-2号卫星CCD估测土壤全氮的研究

遥感监测土壤对于及时快速掌握农田肥力状况,合理施肥意义重大。NDVI在监测地表植被覆盖中发挥着重要作用,水田植被覆盖种类单一,其他影响因素少,这使得通过NDVI监测水稻长势,间接监测土壤肥力状况变得可行。本文就是利用中巴-2号卫星的CCD的植被指数对南京溧水县水田的土壤质量进行监测,回归方程的决定系数R2=0.741,相对误差为0.1663。结论发现利用中巴卫星的植被指数对研究区的的水田氮素的监测是可行的,并在此基础上做了研究区的氮素等级分布图,实现了遥感监测溧水水田土壤全氮的估测。     

不同辐射校正水平下水稻植被指数监测对比分析

归一化植被指数(NDVI)是反映植被长势特征的重要参数之一。获取准确的植被指数对揭示植被长势变化等定量遥感分析至关重要。基于不同辐射校正水平(辐射定标与大气校正),分别利用Landsat ETM+影像的灰度值(DN)、表观(TOA)反射率与地表(Surface)反射率计算相应NDVI,并根据鄱阳湖区野外定点观测数据,从农田、景观尺度揭示不同辐射校正水平下水稻生育期内NDVI动态变化特征。结果表明,根据DN、TOA反射率与Surface反射率提取的NDVI基本上可以反映出年内水稻不同熟制种植信息变化特征,即移栽期NDVI处于谷值,孕穗抽穗期NDVI达到峰值。但相应NDVI逐渐增加,且波动范围逐渐增大。就不同熟制水稻生育期而言,根据DN值计算并构建的NDVI曲线差异较小,而根据TOA反射率与Surface反射率反演的NDVI曲线差异明显。在植被定量遥感研究中,通过大气校正反演地表反射率计算植被指数相对客观准确。