Mapping Rice Paddies in Complex Landscapes with Convolutional Neural Networks and Phenological Metrics     

Shuang Zhao  Chao Ding  Shuyuan Liu  Changshan Wu  Ling Wu 《地理信息系统科学与遥感》2020,57(1):37-48

ABSTRACT

Rice mapping with remote sensing imagery provides an alternative means for estimating crop-yield and performing land management due to the large geographical coverage and low cost of remotely sensed data. Rice mapping in Southern China, however, is very difficult as rice paddies are patchy and fragmented, reflecting the undulating and varied topography. In addition, abandoned lands widely exist in Southern China due to rapid urbanization. Abandoned lands are easily confused with paddy fields, thereby degrading the classification accuracy of rice paddies in such complex landscape regions. To address this problem, the present study proposes an innovative method for rice mapping through combining a convolutional neural network (CNN) model and a decision tree (DT) method with phenological metrics. First, a pre-trained LeNet-5 Model using the UC Merced Dataset was developed to classify the cropland class from other land cover types, i.e. built-up, rivers, forests. Then, paddy rice field was separated from abandoned land in the cropland class using a DT model with phenological metrics derived from the time-series data of the normalized difference vegetation index (NDVI). The accuracy of the proposed classification methods was compared with three other classification techniques, namely, back propagation neural network (BPNN), original CNN, pre-trained CNN applied to HJ-1 A/B charge-coupled device (CCD) images of Zhuzhou City, Hunan Province, China. Results suggest that the proposed method achieved an overall accuracy of 93.56%, much higher than those of other methods. This indicates that the proposed method can efficiently accommodate the challenges of rice mapping in regions with complex landscapes.  相似文献   

超大陆旋回与成矿作用   总被引：1，自引：0，他引：1  

李文渊 《西北地质》2012,(2):27-42

世界上绝大多数大矿床都是超级地质作用导致矿石异常集中的结果,一般多发生于一个地质作用结束的时期.地壳形成作用和板块构造、地幔柱模型已经成为如何认识矿床形成的关键所在,现代矿床成功的探索实践历史就是对大陆和海洋的深人认识进程.因此,矿床形成与全球构造和大陆的演化关系是研究成矿动力学的基础.本文以全球活动构造的视角,从超大陆聚散的历史纪录角度,对活动的地球演化环境下的成矿作用进行总结探索,以图激发对该领域更为深入的研究和思考.  相似文献   

Re–Os and U–Pb geochronology of porphyry Mo deposits in central Jilin Province: Mo ore-forming stages in northeast China     

《International Geology Review》2012,54(14):1763-1785

Central Jilin Province lies along the eastern edge of the Xing–Meng orogenic belt of northeast China. At least 10 Mo deposits have been discovered in this area, making it the second-richest concentration of Mo resources in China. To better understand the formation and distribution of porphyry Mo deposits in the area, we investigated the geological characteristics of the deposits and applied zircon U–Pb and molybdenite Re–Os isotope dating to constrain the age of mineralization. Our new geochronological data show the following: the Jidetun Mo deposit yields molybdenite Re–Os model ages of 164.6–167.1 Ma, an isochron age of 168 ± 2.5 Ma, and a weighted mean model age of 165.9 ± 1.2 Ma; the Houdaomu Mo deposit yields molybdenite Re–Os model ages of 167.4–167.7 Ma, an isochron age of 168 ± 13 Ma, and a weighted mean model age of 167.5 ± 1.2 Ma; and the Chang’anpu Mo deposit yields a zircon U–Pb age for granodiorite porphyry of 166.9 ± 1.5 Ma (N = 16). These new age data, combined with existing molybdenite Re–Os dates, show that intense porphyry Mo mineralization was coeval with magmatism during the Middle Jurassic (167.8 ± 0.4 Ma, r > 0.999). The geotectonic mechanisms responsible for Mo mineralization were probably related to subduction of the Palaeo-Pacific plate beneath the Eurasian continent. Combining published molybdenite Re–Os and zircon U–Pb ages for northeast China, the Mo deposits are shown to have been formed during multiple events coinciding with periods of magmatic activity. We identified three phases of mineralization, two of which had several stages: the Caledonian (485–480 Ma); the Indosinian comprising the Early–Middle Triassic (248–236 Ma) and Late Triassic (226–208 Ma) stages; and the Yanshanian phase comprising the Early–Middle Jurassic (202–165 Ma), Late Jurassic–early Early Cretaceous (154–129 Ma), and Early Cretaceous (114–111 Ma) stages. Although Mo deposits formed during each phase/stage, most of the mineralization occurred during the Early–Middle Jurassic.  相似文献   

Return period maps of dry spells for Catalonia (northeastern Spain) based on the Weibull distribution / Périodes de retour des périodes sèches en Catalogne (nord-est de l'Espagne) à partir de la distribution de Weibull     

《水文科学杂志》2013,58(1):48-64

  相似文献   

Two Mediterranean irrigation communities in front of water scarcity: A comparison using satellite image time series     

《Journal of Arid Environments》2013

Although a number of studies have analysed the methods to monitor crop water needs, the integration of phenological dynamics and irrigation using water stored in reservoirs has received little attention. This study is an example of such analysis applied to two irrigation communities (IC) located in a Spanish Mediterranean area, one with less water resources than the other, which it was carried out between 2002 and 2008. These years comprised periods of water surplus and water scarcity including the 2007–2008 drought considered by the Catalan public water authority the worst since 1944. The dynamics of maize, alfalfa, fruit trees and poplars were analysed using greenness and wetness extracted from remote sensing data. A statistical analysis was applied in order to find out the relationship between crop wetness and stored water. Results show that the IC used two methods for water-saving: crop substitution, by decreasing the area of maize, and timing the crop cycle, delaying or advancing sowing depending on water availability. In conclusion although one IC is drier in rainfall terms than the other, a similar crop wetness status was detected and no difference was observed between periods of water scarcity and water surplus.  相似文献   

新疆天山南部构造区不同时段小地震目录完整性分析   总被引：1，自引：0，他引：1  

王琼  曲延军  高歌  聂晓红 《地震研究》2011,34(1)

依据天山地区地质构造、运动强度和历史地震活动的差异性,将南部天山地区划分为南天山东段和南天山西段两个构造区.考虑到新疆台网建设的区域差异性和时间的阶段性,在对两个构造区不同震级下限地震频度时间特征分析的基础上,研究了各构造区不同时段的G-R关系和震级分布特征,并运用Z检验方法对各区不同时段小地震目录差异的显著性进行了分析,进而评估了两个构造区1970年以来不同时段的小地震目录完整性,为区域地震活动性研究中地震目录的合理运用提供参考依据.  相似文献   

基于高分六号宽幅影像的油菜种植分布区域提取方法     

姜楠  张雪红  汶建龙  葛州徽 《地球信息科学学报》2021,23(12):2275-2291

油菜作为我国主要的农业经济作物及食用油的主要来源,及时、准确地获取其种植分布信息,是全面掌握油菜种植状况、加强生产管理、优化作物种植空间格局的重要依据。高分六号（GF-6）的宽视场（Wide Field View,WFV）传感器在可见光-近红外波段基础上增设了2个红边波段、1个黄波段和1个紫波段,为油菜遥感识别提供了更加丰富的光谱信息,进而相较于蓝、绿、红、近红外4个“传统波段”的识别精度有所提升。本文以油菜开花期内两景不同时相GF-6 WFV影像拼接图像作为数据源,选择油菜生产优势区的河南省固始县为研究区,针对油菜同其他地物的“异物同谱”现象以及不同生长阶段油菜的“同物异谱”现象,利用油菜开花期独特的反射光谱特征,结合均值间标准化近距离提出了NDSI₂₈、S₃₄、NDSI₂₃和NDSI₄₆共4个光谱指数,并由此构建油菜种植区域提取的决策树模型。研究结果表明,基于4个指数组合构建的决策树模型对油菜种植分布信息的提取达到了较好的效果,总体精度为96.17%,与随机森林、支持向量机、最大似然法相比分别高出0.31%、0.88%和1.24%;制图精度方面,决策树法为98.15%,比随机森林、支持向量机、最大似然法分别高4.72%、4.21%和5.59%;对于用户精度,决策树法为86.89%,较随机森林、最大似然法分别低2.2%和1.63%,比支持向量机高0.11%。由此说明,GF-6 WFV数据中的新增波段极大地丰富了其光谱信息,使其在包括油菜在内的农作物种植分布信息提取中具有独特的优势和巨大潜力。  相似文献   

气象条件对浙江省泡桐物候期的影响研究   总被引：3，自引：1，他引：2  

郁珍艳  高大伟  吴利红  范广洲 《气象科学》2014,34(6):666-671

利用气候资料及泡桐物候期数据,分析了浙江省泡桐关键生育期与相关气象要素的关系,研究了泡桐开花展叶前相关气候特征,探讨了冷空气对泡桐秋季物候期的可能影响。从统计值的关系来看,其物候期主要取决于气温及≥10℃有效积温的高低,降水量的影响其次,日照时数的影响最小。泡桐物候期对当季的气温条件非常敏感,降水对其开花、展叶的影响有滞后作用。泡桐开花和展叶始期前3个月的≥10℃有效积温一般分别要达到200℃·d和300℃·d以上;由落叶末期至下一年开花和展叶始期的≥10℃有效积温也分别要基本达到200℃·d和300℃·d以上;开花展叶前3个月的降水量和日照时数一般在230 mm和在210 h以上。泡桐秋季物候期前7 d内的冷空气,特别是中等强度以上的冷空气对其影响很大,落叶前的冷空气促使泡桐开始落叶,落叶开始后,冷空气将加速落叶。开花末期前2周的冷空气,在一定程度上促进了泡桐花期的结束。  相似文献   

基于Sentinel-2时间序列数据及物候特征的棉花种植区提取          下载免费PDF全文

美合日阿依·莫一丁  买买提·沙吾提  李金朝 《干旱区地理》2022,45(6):1847-1859

棉花是中国重要的经济作物,在新疆大面积种植。及时、准确获取棉花种植面积,对农业政策制定与农业经济发展有重要意义。以渭干河—库车河三角洲绿洲棉花为主要研究对象,利用2018—2020年（1景/1月）36景哨兵2号（Sentinel-2）数据,构建归一化植被指数（Normalize difference vegetation index,NDVI）和红边归一化植被指数（Red edge normalize difference vegetation index,RENDVI₇₈₃）时序数据;采用Savitzky-Golay（S-G）滤波法对时序数据进行平滑、重构并提取11个物候特征;利用袋外误差法对11个物候特征进行特征优选;在此基础上利用重构后的时序数据（NDVI Fit）、（RENDVI₇₈₃ Fit）、物候特征（RENDVI₇₈₃ Ph）、物候特征优选组合构建6种不同的特征数据集,利用随机森林分类（RFC）方法分别进行分类和提取,并采用最大似然分类方法和支持向量机分类方法对分类效果进行验证。结果表明：（1） NDVI和RENDVI₇₈₃时序数据变化趋势较为一致,棉花在5月（苗期）到8月初（开花盛期）有明显的上升趋势,在8月末至9月（花铃期）达到峰值。相比NDVI,红边波段构成的RENDVI₇₈₃时序曲线峰值从0.7提高到0.9,棉花区分效果更佳。（2） 11个物候特征中拟合函数最大值、生长季长度、生长季振幅、生长季结束、生长季大积分和生长季小积分对分类的贡献性最大,重要性得分分别为1.43、1.40、1.23、1.16、1.02和1.01。（3） RFC方法对特征数据集（RENDVI₇₈₃ Fit+物候特征优选组合）分类精度最佳。总体精度和Kappa系数分别为92.20%和0.92。（4） 研究区内棉花分类精度达到了91.02%,种植面积约为3424 km²,占研究区总面积的24.67%。  相似文献   

STUDY ON MIXIS POTENTIAL OF ROTIFER RESTING EGGS（ BRACHIONUS PLICATILIS ） WITH DIFFERENT COLLECTION TIMES AND DIFFERENT PRESERVATION PERIODS   总被引：1，自引：0，他引：1  

周利郑严  相建海 《中国海洋湖沼学报》2001,19(3):222-227

The present study investigated the possible changes in the mixis potential of rotifer resting eggs produced by a single stock of Brachionus plicatilis and collected and preserved annually from 1985 - 1998. Several clones derived from each batch of resting eggs were cultured under the same conditions for 21 days. The percentage of clones appearing resting eggs and the average yield of resting eggs produced from each clone were recorded and statistically analyzed to find the differences between the mixis potential of those resting egg batches. Results showed that different batches of resting eggs had different mictic levels among their descendent clones; but no regular relationship was found between the mixis potential of resting eggs and their collection times/preservation periods. Several internal and external factors that might affect the mixis potential of resting eggs were discussed.  相似文献