基于SVM的灌区无人机影像渠系提取

灌区渠系制图配合现代节水灌溉技术,对合理配水、安全输水有着重大影响。但目前普遍使用的灌区遥感影像分辨率不高,给渠系提取与制图带来一定的困难。本文将无人机采集的高精度正射影像、高程、坡度数据相结合作为数据源,提取出具有显著描述能力的渠系特征来构建训练样本集,基于支持向量机的分类方法对目标渠系进行分割提取,再通过后处理对提取结果进行去噪、连接和优化,实现了无人机高分辨率多数据源的渠系提取。结果表明,该渠系提取方法可以识别提取灌区中的支渠、斗渠和部分农渠,渠系连续性良好,与手绘渠系对比,精度最高可达89.35%。其中提取误差主要由级别较低的渠系中渠床淤泥沉积导致影像、地形特征不明显造成。     

基于SRTM的地形因子提取方法研究

地形是影响土壤侵蚀的重要因子,在侵蚀估算模型中常用坡度和坡长（LS）来衡量,在大区域上常基于数字高程模型（DEM）提取。SRTM作为大区域尺度上质量高、易获取的高程数据,在全球土壤侵蚀评价中得到广泛应用。但现有地形因子提取算法要求高程和栅格单元的单位（通常为m）一致,需对SRTM进行坐标变换才能使用。针对大区域上SRTM坐标转换时间开销大的问题,提出了一种直接基于SRTM提取地形因子的算法（LSA-SRTM）。该算法利用地理坐标系下的经纬度信息计算栅格单元长度及单元坡长,结合最陡坡降策略获得坡度及流向,进而提取汇水面积,根据坡度设置坡度截断点,根据汇水面积阈值设置沟道截断点,经正、反遍历后获得累积坡长,采用CSLE的分段公式计算LS因子。以Himmelblau-Orlandini数学曲面、5个典型样区的1″SRTM作为数据源,将LSA-SRTM、投影坐标系下的LS算法（LSA-DEM）与手工测量的结果进行对比。LSA-SRTM方法与测量值在数学曲面和典型样区,坡长的R²分别为0.855 2、0.778 8、0.726 9、0.702 4、0.690 9、0.725...     

DEM超分辨率重构对梯田坡度提取的影响研究

坡度对地表水文、土壤侵蚀、土地利用规划有着重要的影响,区域尺度上的坡度通常基于数字高程模型(DEM)提取。区域尺度上,高分辨率坡度数据由于DEM获取途径、方式等原因,较难获得,通常通过超分辨率重构(又称降尺度变换)得到。以黄土高原地区水平梯田地形为研究对象,基于无人机摄影测量技术,生成不同分辨率的DEM数据并提取坡度,设计并给出了基于稀疏混合估计对DEM数据进行超分辨率重构的方法及流程,并与最近邻法、双线性插值法、三次卷积插值法比较,结果表明所提方法在空间分布和误差方面上均优于其他方法。     

基于无人机遥感图像提取农田微地形特征

近年来极端暴雨天气与自然灾害频发,导致农田损毁,影响耕作。该研究利用高精度农田数字地形模型（FDTM, Farmland Digital Terrain Model),基于地形因子综合属性提出一种识别农田微地形特征（凸起特征及洼地特征）的方法。首先,基于SfM（Structure from Motion）技术处理试验田的航拍图像,获取高精度农田FDTM。分析FDTM的高程方差随局部窗口尺度的变化趋势,确定分析窗口的尺度区间为31×31至51×51像素。其次,选择高程、地形起伏度和坡度综合评价在51×51像素窗口下提取的315个高程极值点,获取多窗口地形因子综合隶属度。最终,根据斯特吉斯公式确定阈值为0.627,提取16个农田凸起特征顶点,并结合等高线图识别凸起特征的外形轮廓。类似地,建立反转数字地形模型（RFDTM, Reverse-FDTM）,将FDTM中的洼地特征转变为RFDTM中的凸起特征,识别9个农田洼地特征。研究结果可为农田复垦及精准土地平整作业提供理论依据与方法支持。     

基于地形特征的无人机遥感梯田影像边缘提取方法

梯田具有蓄水固沙的作用,是旱作农业区重点建设的高产稳产农田设施,为粮食增产、农民增收提供了有力保障。因仅基于影像数据采用边缘提取方法进行梯田区域分割效果不理想,及时准确地掌握梯田信息较为困难。无人机遥感技术的不断发展为高精度梯田地形信息的获取提供了新方法。本研究以甘肃省榆中县为例,首先从数字高程模型DEM数据中提取坡度,将正射影像与坡度数据融合,并通过基于Canny算子的粗边缘提取方法和基于多尺度分割的精细边缘提取方法,对比分析坡度对无人机遥感梯田影像边缘提取的影响。试验结果表明,正射影像和坡度融合的提取效果均优于单一的正射影像数据提取效果,粗边缘提取方法中正射影像和坡度融合的数据源精度平均提高了23.97%,精细边缘提取方法中正射影像和坡度融合的数据源精度平均提高了17.84%。研究表明,在无人机遥感梯田影像边缘提取中加入一定的地形特征,可以取得更好的边缘提取效果。     

基于全卷积神经网络的灌区无人机正射影像渠系提取

为快速准确获取灌区渠系分布信息,科学调配区域农业水资源、提高水资源利用率,通过基于全卷积神经网络(Fully convolutional networks,FCN)的语义分割模型进行渠系轮廓提取。利用无人机采集正射影像并进行标注,以VGG-19网络为基础,通过多尺度特征融合的方式实现FCN-8s结构,使用Tensorflow深度学习框架构建FCN渠系提取模型;对数据集进行数据增强,分割后放入FCN模型中训练、测试。实验结果显示,针对不同复杂程度的测试区域,FCN模型的提取准确度、完整度、精度均高于支持向量机方法和改进霍夫变换方法,均值分别为95. 78%、92. 29%、89. 45%。结果表明,该方法能够实现灌区渠系轮廓的高精度提取,具有较好的泛化性和鲁棒性。     

GIS环境下基于DEM的三川河流域特征提取

以地理信息系统软件Arcgis10.0为平台,探讨了其水文分析模块在流域划分、地形分析、河网提取上所采用的算法和原理。并以三川河流域为例,基于数字高程模型（DEM）数据,详述了提取流域特征的方法和步骤,实现了流域边界、坡度、坡向、表面曲率及流域河网的自动提取。结果表明：上述成果均能较好的吻合流域的实际情况,该方法具有操作简单、灵活方便、成果精度高、提取速度快等特点。流域特征的自动提取为进一步研究水土保持、水文模型、流域规划及数字流域等提供了基础数据,具有重要的实际意义。     

基于GS的南方水库灌区塘堰蓄水

探讨了基于GIS技术建立塘堰蓄水能力与地形因子关系的方法。提出用稻田塘堰容积率指标来描述塘堰的相对供水能力。用GIS 的空间分析功能，从 DEM获得了研究区域的坡度、高程及河网密度等地形因子。基于实地调查数据，在分析稻田塘堰容积率与单个地形因子关系的基础上，建立了稻田塘堰容积率与地形因子的综合方程。为精确快速的估算区域塘堰供水能力提供了简易的途径，它也可为指导灌区水资源优化管理和灌区改造中新建塘堰容量及位置的确定提供决策依据。     

基于数字高程模型的淤地坝地区河网提取方法

淤地坝作为黄土高原水土流失严重地区重要而独特的治沟工程，具有滞洪、拦淤、蓄水等作用，但在该地区利用数字高程模型(DEM)进行流域水系分析时，提取的河网会受淤地坝的干扰出现河网偏移、错位问题。本文提出一种通过对淤地坝进行检测并修正DEM的自动化方法来实现河网的有效提取。该方法先利用区域生长和形态学算法提取河道中心线，以缩小淤地坝检测范围，再基于改进的直线段检测(LSD)算法以及角度滤波确定淤地坝候选轮廓直线，然后构建十字模型定位淤地坝，最后对淤地坝所在栅格高程进行修正以提取河网。该方法在黄土高原两个典型样区中检测淤地坝的F1值分别为86.67%和86.95%,相比于仅使用填充、裂开的两种河网提取方法，本文方法能有效解决河网偏移问题。结果表明，本文方法提取的河网结果更符合真实地形，可为黄土高原淤地坝地区的流域数字地形分析技术提供有益补充与支持。     

浅析DTM地形数据采集方法与比较

随着科技进步,数字地面模型（DTM）在公路设计中得到广泛应用,地形原始数据是建立数模的基础数据,它直接影响到建模的合理性以及高程内插结果是否可靠,原始数据点的分布及密度是否与地形变化、数据采集设备和方法、数模类型、建模目的、数模精度以及所研究地区的地形特征等诸多因素相适应,通过对DTM地形数据采集方法的分析比较,尽可能做到以不太多的地形点表示出具有一定精度的地形表面。     

节水产品的快速成形开发

随着我国节水灌溉事业的发展需求，开发适合于我国国情且具有自主知识产权的节水灌溉新产品迫在眉睫。快速成形技术在节水灌溉领域的应用，使得节水灌溉新产品的快速开发成为现实。介绍了激光快速成形技术原理，阐述了利用该技术进行喷微灌灌水器产品的开发制作过程及其特点，这些特点显示了利用激光快速成形技术进行节水产品开发的优越性。     

基于影像与坡度数据融合的梯田田块分割方法

梯田在很大程度上开发了坡耕地的农业生长潜力,具有蓄水、保土作用。由于梯田数量、面积等分布信息较难准确获得,使其定量研究难以深入展开。随着无人机技术的不断发展,高精度梯田地形信息的获取成为可能。本文基于无人机正射影像并结合坡度数据,通过Canny边缘检测算子对梯田的粗轮廓进行提取,结合梯田的结构特性,对梯田中的伪边缘进行剔除;再通过对梯田边缘强度叠加和边缘连接;最后利用区域生长算法对梯田进行分割。该方法有效解决了梯田形状不规则、田面堆积物干扰、图像光谱特征复杂等问题。与手工标注的梯田样区田块数据的对比结果表明,本文算法对梯田区的提取总精度可达84.9%,可为梯田区的快速制图提供解决方案。     

基于深层残差网络的山区DEM超分辨率重构

针对大区域高分辨率数字高程模型(DEM)数据较难获取、超分辨率重构(降尺度)较低分辨率的DEM精度不高、难以满足实际需要的问题,提出一种对起伏特征较明显的山区DEM超分辨率重构的方法。利用较深层的神经网络充分学习高低分辨率DEM之间的非线性映射关系;为了降低训练难度,结合残差学习的方法进行数据训练。将双立方插值法、稀疏混合估计法重构的DEM及提取的坡度结果分别同深层残差网络法的结果进行对比,结果表明,3种方法DEM结果的差值平均值分别为0.41、0.34、0.34 m,RMSE分别为0.5945、0.5715、0.4869 m;坡度结果的差值平均值分别为3.02°、2.04°、1.99°,RMSE分别为3.6498°、3.1360°、2.7387°;处理时间分别为0.052、663.39、2.16 s。研究表明,对于10、20、40 m的DEM,本文方法在空间分布和误差方面优于其他方法,在耗时效率上也优于稀疏混合估计法,适合应用于梯田等地形复杂的区域进行超分辨率重构。     

基于K-means聚类算法的草莓灌溉策略研究

为进一步提高日光温室封闭式栽培下草莓灌溉水肥利用率,研究了基质含水率和温度影响下的草莓灌溉策略优化方法。采用土壤水分传感器对草莓果期基质含水率进行实时监测,通过对基质含水率随时间变化的规律分析,并结合日平均温度进行K-means聚类分析,提出一种草莓优化灌溉策略。试验结果表明,灌溉第1阶段基质含水率快速上升,在灌溉结束时达到峰值,每次灌溉基质含水率平均提高21.5个百分点;第2阶段快速下降,在20 min内基质含水率平均下降3.5个百分点;第3阶段变化趋于平稳,在30 min内基质含水率平均下降1.2个百分点。在每个灌溉周期内,含水率呈线性下降趋势,在整个果期内,其斜率随日平均温度的升高逐渐增加,由0.0114增加至0.0365。研究结果表明,根据基质含水率变化和日平均温度区间进行定量灌溉,理论上果期每株草莓仅需要4.51 L水,可节水15.4%,该方法能有效提高水肥利用率,实现节水节肥。     

GIS user-interface based irrigation delivery performance assessment: a case study for Tanjung Karang rice irrigation scheme in Malaysia

Various indicators are used for evaluating the performance of different aspects of an irrigation system, and assessments also differ in terms of the types of performance indicators used. This paper describes a GIS-based assessment system which utilizes a new concept and evaluated the inadequacy of a widely used Relative Water Supply (RWS) concept to characterize the irrigation delivery performance for a rice irrigation system as the season advances. Development of this GIS-based assessment system resulted in the creation of new indicators, viz., the Rice Relative Water Supply (RRWS), Cumulative Rice Relative Water Supply (CRRWS) and Ponding Water Index (PWI). These indicators were determined from field tests and evaluated in a Malaysian Tanjung Karang Rice Irrigation Scheme (TAKRIS). The RWS concept was found to be inaccurate for characterizing the oversupply condition on irrigation deliveries for rice irrigation; and difficult to correctly quantify the oversupply condition for irrigation supplies. Besides, it was found that the RRWS indicator can distinctly characterize the oversupply condition for RRWS > 1.0 and undersupply condition for RRWS < 1.0 on irrigation delivery for any given period. A value of 1.0 for RRWS indicates an irrigation delivery that matches perfectly the actual field water demand. This study presents a cumulative RRWS plot that provides important information on irrigation supplies for any given time interval for management decisions. An increasing slope in the actual CRRWS curve with CRRWS = 1.0, means that irrigation supply can be slightly curtailed in the next period. On the other hand, if the slope is negative, supply has to be increased. If a computed CRRWS line follows the CRRWS = 1.0 line, it means that irrigation deliveries are perfectly matched with the field water demand. A graphical user-interface was developed for structuring the assessment tool within an ArcGIS platform. The system can instantly provide information on the uniformity of water distribution and the shortfall or excess, and provides vital information in terms of decisions that need to be made for the next period. The system helps to maintain continuous updating of input and output databases on real field conditions. Results are displayed on the computer screen together with color-coded maps, graphs and tables in a comprehensible form. The system is likely to be adopted for evaluating various water allocation scenarios and water management options. It can also be used as an analytical and operational tool for irrigation managers.     

马铃薯水肥一体化技术规程

针对目前我国马铃薯水肥一体化灌溉系统操作不规范问题,制定了马铃薯水肥一体化技术规程。马铃薯水肥一体化灌溉是目前世界上公认最好的灌溉方式,马铃薯水肥一体化灌溉不仅可以大大减少农业灌水的使用,还能显著减少化肥农药的使用,对保护环境、净化地下水起到了非常重要的作用。结合马铃薯种植农艺要求及水肥一体化操作要求,制定了智能马铃薯水肥一体化系统的使用操作技术规程,在设施要求、硬件系统操作及马铃薯农艺种植方面进行了规范,为广大马铃薯种植用户提供了合理的技术参考,为马铃薯高效滴灌施肥施药推广提供理论依据。      

Yield and olive oil characteristics of a low-density orchard (cv. Cordovil) subjected to different irrigation regimes

The impact of different irrigation scheduling regimes on the quantity and quality of olive oil from a low-density olive grove in southern Portugal was assessed during the irrigation seasons of 2006 and 2007. Olive trees were subjected to one of the following treatments: A—full irrigation; B—sustained deficit irrigation (SDI) with 60% of ETc water applied with irrigation; C—regulated deficit irrigation (RDI) with irrigation water applied at three critical phases: before flowering, at the beginning of pit hardening and before crop harvesting and D—rain-fed treatment. Olive oil yield was significantly higher than rain-fed conditions in 2006, an “on year” of significant rainfall during summer. No significant yield differences were observed in the following “off year”. Among the irrigated treatments, olive oil production of treatment B was 32.5% and 40.1% higher in 2006 and 2007, respectively than the fully irrigated treatment A, despite receiving 49% less irrigation water. Such strategy could allow for an efficient use of water in the region, of very limited available resources, and for modest but important oil yield increase. Nonetheless, on the “on year” of 2006 treatment C used 13.9% of the water applied to treatment B and produced only 23.9% less olive fruits which could also make it illegible as the next possible strategy to use for irrigating olive trees in the region, provided that water is secured latter in the summer, a period of vital importance for oil accumulation and very sensitive to water stress as the poor results of 2007 revealed. The different treatment water regimes did not impact on the chemical characteristics of olive oils that were within the set threshold limits. Similarly, the sensory characteristics of the olive oils as well as bitterness and pungency were negligible for all treatments allowing them to be assessed as of “superior quality”.Overall, irrigation treatments had no influence on the commercial value of produced oils, being all classified as “extra virgin”. Such funding may be of vital importance to farmers willing to further their irrigation area, save water and still retain the protected designation of origin (PDO) seal of quality for their oil.     

Optimal on-farm irrigation scheduling with a seasonal water limit using simulated annealing

As water resources are limited and the demand for agricultural products increases, it becomes increasingly important to use irrigation water optimally. At a farm scale, farmer's have a particularly strong incentive to optimize their irrigation water use when the volume of water available over a season is production limiting. In this situation, a farmer's goal is to maximize farm profit, by adjusting when and where irrigation water is used. However, making the very best decisions about when and where to irrigate is not easy, since these daily decisions require consideration of the entire remaining irrigation season. Future rainfall uncertainty further complicates decisions on when and which crops should be subjected to water stress. This paper presents an innovative on-farm irrigation scheduling decision support method called the Canterbury irrigation scheduler (CIS) that is suitable when seasonal water availability is limited. Previous optimal scheduling methods generally use stochastic dynamic programming, which requires over-simplistic plant models, limiting their practical usefulness. The CIS method improves on previous methods because it accommodates realistic plant models. Future farm profit (the objective function) is calculated using a time-series simulation model of the farm. Different irrigation management strategies are tested using the farm simulation model. The irrigation strategies are defined by a set of decision variables, and the decision variables are optimized using simulated annealing. The result of this optimization is an irrigation strategy that maximizes the expected future farm profit. This process is repeated several times during the irrigation season using the CIS method, and the optimal irrigation strategy is modified and improved using updated climate and soil moisture information. The ability of the CIS method to produce near optimal decisions was demonstrated by a comparison to previous stochastic dynamic programming schedulers. A second case study shows the CIS method can incorporate more realistic farm models than is possible when using stochastic dynamic programming. This case study used the FarmWi$e/APSIM model developed by CSIRO, Australia. Results show that when seasonal water limit is the primary constraint on water availability, the CIS could increase pasture yield revenue in Canterbury (New Zealand) in the order of 10%, compared with scheduling irrigation using current state of the art scheduling practice.     

灌区灌溉用水时空优化配置方法

将传统的灌溉水量在作物间的优化分配模型和建立的渠系工作制度多目标优化模型与地理信息系统相集成,提出了基于空间决策支持系统的灌区灌溉用水优化配置的新方法.综合考虑了灌区内作物、土壤、气象站点、渠系布置的空间差异、年季间气象以及作物不同生育阶段对应参数的时间差异.与传统优化方法相比,该方法可根据管理者对优化精度的要求,灵活选择优化尺度,同时,简化了求解时空优化配水问题的繁琐程度,结果表现形式更加丰富.在此基础上建立的空间决策支持系统界面友好,运行效率高,可移植性和通用性强.经实例验证,优化后的配水方案与原配水方案相比较,灌溉总用水量减少296％,产量增加243％,水分生产率提高05 kg／m3,灌溉净效益增加168％.优化后配水方案具有将有限的水资源向经济价值较高作物转移的趋势.该方法为灌区灌溉用水优化配置提供了新思路.     

Evaluating infiltration for border irrigation models

The infiltration characteristics of a soil are important to the design, evaluation and management of border irrigation systems. The use and verification of border irrigation models also rely heavily on infiltration. This paper presents a technique for determining infiltration when detailed information is available on the total infiltrated volume during the irrigation which can be obtained from measurements of inflow, outflow, and water depths on the border strip. The method uses a volume balance at progressive times and is an extension of earlier work. Data from this method were used as input to the zone-inertia border irrigation model and good agreement was found between measured and computed values of advance, recession, runoff rates and volumes, and surface water depths.