基于机器视觉的马铃薯晚疫病快速识别

晚疫病是马铃薯的一种严重病害,可造成减产甚至绝收。因此马铃薯晚疫病的识别与控制对提高其产量有非常重要的意义。该文基于机器视觉技术对马铃薯叶部晚疫病进行检测,根据马铃薯叶片上晚疫病斑的颜色、纹理和形状特征参数的不同,提取叶片表面的特征参数,并建立数学模型对病害程度做出评价。在RGB、HSV颜色空间中,根据马铃薯叶片在患病早期叶片颜色发生变化且与健康叶片不同,利用颜色特征,建立马铃薯晚疫病的无病和患病模型,该模型对马铃薯患病早期的识别率为67.5%。利用灰度共生矩阵,采用纹理统计参数进行病害等级评价,用熵值和能量值描述晚疫病的严重程度,纹理特征对患病程度的识别率比较稳定,对患病中期与后期的识别率分别为72.5%与80%。利用形状特征的相对特征,根据病斑面积比进行晚疫病诊断,该方法对马铃薯叶片晚疫病患病后期的诊断取得较好效果,识别率为90%,但由于叶片患病早期的病斑面积小且分散,识别难度大,识别率仅为50%。针对颜色、纹理及形状特征在识别马铃薯叶片晚疫病时的优势与局限性,提出颜色纹理形状特征结合的识别方法,对患病中期与后期的识别率分别为90%和92.5%。通常马铃薯晚疫病的理化值检测法耗时数天,但利用机器视觉识别马铃薯晚疫病患病情况非常快速,根据颜色特征进行病害识别的时间约为4 s,纹理特征识别的时间为7 s,形状特征特征识别的时间为3 s,综合颜色纹理形状特征的识别由于计算量较大,识别时间为9 s。该研究可为基于机器视觉的马铃薯晚疫病的快速检测提供理论依据。     

基于环境变量的中国土壤有机碳空间分布特征

研究中国土壤有机碳(Soil Organic Carbon,SOC)的空间分布特征对SOC储量估算以及农业生产管理具有重要意义。以全国第二次土壤普查2473个土壤典型剖面的表层(A层)SOC含量为研究对象,探寻地形、气候和植被等环境因素对SOC空间异质性分布的影响;以普通克里格法为对照,利用地理加权回归、地理加权回归克里格、多元线性回归和回归克里格模型建立SOC空间预测模型;并分别绘制了中国SOC的空间分布预测图。结果表明:(1)SOC含量与年均降水量、年均温、归一化植被指数、高程以及地形粗糙指数呈极显著相关关系;(2)平均绝对估计误差、均方根误差、平均相对误差和皮尔逊相关系数等模型验证指标表明地理加权回归的预测精度优于其他模型,可以更好地绘制SOC在大尺度上的空间分布特征;(3)较高SOC含量主要分布在研究区东北部、西南部以及东南部,而西北部SOC含量普遍偏低。本文以期从大尺度上探讨土壤属性与环境变量之间的相关关系,为全国土壤属性的空间制图提供一定的解决方案和思路。     

基于FCN的无人机玉米遥感图像垄中心线提取

为解决农业机器人在玉米田行间行走的全局路径规划问题,该研究提出一种基于全卷积神经网络(Fully ConvolutionalNetworks,FCN)的无人机玉米遥感图像垄中心线提取方法。基于无人机获取的高精度可见光遥感图像,设计了针对农田垄中心线提取的数据集标注方法,采用滑动窗口法进行图像分块,利用深度学习语义分割网络FCN对垄中心线附近7～17像素宽度范围的垄线区域进行提取,模型在测试田块上精确率达66.1%～83.4%,召回率达51.1%～73.9%,调和平均值为57.6%～78.4%;对拼接后的图像使用影像分割投影法提取中心线,探究了垄线区域宽度对垄中心线提取精度的影响,训练采用9像素的垄区域宽度,可得到垄中心线在77 mm左右偏差范围准确率为91.2%,在31.5 mm左右偏差范围内为61.5%。结果表明,基于FCN对无人机玉米遥感图像进行处理,可得到整片田地的垄中心线栅格地图,方便农业机器人进行全局路径规划。     

纳米氧化锌和纳米二氧化钛对尾草履虫急性毒性效应的比较研究

[目的][方法] 本研究以原生动物尾草履虫（Paramecium caudatum）为实验对象, 对目前广泛应用的两种纳米材料，纳米二氧化钛（TiO2）和纳米氧化锌（ZnO）的24小时急性毒性进行了比较研究，分析了草履虫在两种纳米材料胁迫下抗氧化酶活性的改变，并结合两种纳米材料的超微形态，探讨了纳米材料毒性效应的机制，为两种纳米材料的环境毒性监测与评价提供依据。[结果]结果显示：对尾草履虫的24小时半数效应浓度(24h-EC50)分别为51.580mg·L-1（TiO2）、0.444mg·L-1（ZnO）；最低有影响浓度(24h-LOEC)分别为1.712mg·L-1（TiO2）、0.352mg·L-1（ZnO）；以EC50浓度24小时胁迫尾草履虫，获得抗氧化酶活性改变为：超氧化物歧化酶(SOD)，过氧化氢酶(CAT)和过氧化物酶(POD)活性与对照组相比有不同程度的改变，经成对检验分析，两种纳米材料对尾草履虫的POD酶表现出较为明显的抑制作用。[结论]上述结果表明：纳米氧化锌对草履虫的24小时急性毒性大于纳米二氧化钛，这与纳米氧化锌对草履虫抗氧化酶的抑制作用更强，以及纳米氧化锌(30nm)的超微形态相较于纳米二氧化钛(25nm)更加纤细和尖锐有关；本研究获得的24小时急性毒性指标可用于监测和评价两种纳米材料的环境毒性，并为纳米材料的环境容量和生态安全阈值确定提供参考。     

三七素提取工艺及检测方法的优化

本研究以三七（Panax notoginseng）植株及其产品和近缘种为研究对象,采用L₉(3 ⁴)正交试验对三七素的提取工艺进行了优化,然后利用HPLC法检测三七素含量,并对各样品中的三七素含量进行了比较分析。结果表明：优化后的三七素提取条件为超纯水、料液比1∶20（g/mL）,提取1次（10 min）;三七素在三七花蕾中的含量最高,其次是根状茎和主根,但是三七茎叶中也有较高含量的三七素,说明三七花蕾和茎叶均可作为三七素的提取原料;在三七产品中三七花的三七素含量最高,达到3.33%,其次是剪口（根状茎）、筋条（侧根）和须根;但三七素的含量与三七个头大小（头数）并无明显的相关性;在三七近缘种姜状三七（P. zingiberensis）和野三七（P. vienamensis var. fuscidiscus）主根、珠子参（P. japonicus var. major）根状茎中的三七素含量显著低于三七主根三七素含量,也显著低于三七其他部位的三七素含量,说明三七是最佳的三七素来源植物。三七素的含量主要受到物种和部位的影响,与产品规格无关。本文所建立的三七素提取方法和HPLC检测方法,能够快速简便、稳定可靠测定样品中三七素含量;并可为三七素相关标准的建立奠定基础,为三七的综合利用开发提供依据。     

Carbon stocks and carbon fluxes from a 10‐year prescribed burning chronosequence on a UK blanket peat

Prescribed burning is a common land management technique in many areas of the UK uplands. However, concern has been expressed at the impact of this management practice on carbon stocks and fluxes found in the carbon‐rich peat soils that underlie many of these areas. This study measured both carbon stocks and carbon fluxes from a chronosequence of prescribed burn sites in northern England. A range of carbon parameters were measured including above ground biomass and carbon stocks; net ecosystem exchange (NEE), net ecosystem respiration (R_eco) and photosynthesis (P_g) from closed chamber methods; and particulate organic carbon (POC). Analysis of the CO₂ data showed that burning was a significant factor in measured CO₂ readings but that other factors such as month of sampling explained a greater proportion of the variation in the data. Carbon budget results showed that whereas all the plots were net sources of carbon, the most recent burn scars were smaller sources of carbon compared with the older burn scars, suggesting that burning of Calluna‐dominated landscapes leads to an ‘avoided loss’ of carbon. However, this management intervention did not lead to a transition to a carbon sink and that for carbon purposes, active peat‐forming conditions are desirable.     

Greater humification of belowground than aboveground biomass carbon into particulate soil organic matter in no-till corn and soybean crops

Quantifying the amount of carbon (C) incorporated from decomposing residues into soil organic carbon (C_S) requires knowing the rate of C stabilization (humification rate) into different soil organic matter pools. However, the differential humification rates of C derived from belowground and aboveground biomass into C_S pools has been poorly quantified. We estimated the contribution of aboveground and belowground biomass to the formation of C_S in four agricultural treatments by measuring changes in δ¹³C natural abundance in particulate organic matter (C_POM) associated with manipulations of C₃ and C₄ biomass. The treatments were (1) continuous corn cropping (C₄ plant), (2) continuous soybean cropping (C₃), and two stubble exchange treatments (3 and 4) where the aboveground biomass left after the grain harvest was exchanged between corn and soybean plots, allowing the separation of aboveground and belowground C inputs to C_S based on the different δ¹³C signatures. After two growing seasons, C_POM was primarily derived from belowground C inputs, even though they represented only ∼10% of the total plant C inputs as residues. Belowground biomass contributed from 60% to almost 80% of the total new C present in the C_POM in the top 10 cm of soil. The humification rate of belowground C inputs into C_POM was 24% and 10%, while that of aboveground C inputs was only 0.5% and 1.0% for soybean and corn, respectively. Our results indicate that roots can play a disproportionately important role in the C_POM budget in soils. Keywords Particulate organic matter; root carbon inputs; carbon isotopes; humification rate; corn; soybean.     

Indicators and trade-offs of ecosystem services in agricultural soils along a landscape heterogeneity gradient

Soil functions can be classified as supporting (nutrient cycling) and provisioning (crop production) ecosystem services (ES). These services consist of multiple and dynamic functions and are typically assessed using indicators, e.g. microbial biomass as an indicator of supporting services. Agricultural intensification negatively affects indicators of soil functions and is therefore considered to deplete soil ES. It has been suggested that incorporating leys into crop rotations can enhance soil ES. We examined this by comparing indicators of supporting soil services – organic carbon, nitrogen, water holding capacity and available phosphorous (carbon storage and nutrient retention); net nitrogen mineralisation rate and microbial biomass (nutrient cycling and retention) – in barley fields, leys and permanent pastures along a landscape heterogeneity gradient (100, 500 and 1000 m radii). In addition, barley yields (provisioning service) were analysed against these indicators to identify trade-offs among soil services. Levels of most indicators did not differ between barley and ley fields and were consistently lower than in permanent pastures. Leys supported greater microbial biomass than barley fields. Landscape heterogeneity had no effect on the indicators or microbial community composition. However, landscape heterogeneity correlated negatively with yield and soil pH, suggesting that soils in heterogeneous landscapes are less fertile and therefore have lower yields. No trade-offs were found between increasing barley yield and the soil indicators. The results suggest that soil ES are determined at the field level, with little influence from the surrounding landscape, and that greater crop yields do not necessarily come at the expense of supporting soil services.