基于机器视觉的嫁接用苗外观特征自动检测

提出了基于数学模型的幼苗外观特征自动检测方法,检测项目包括生长状态、子叶参数和胚轴参数。首先经过图像预处理提取幼苗二值图,利用行像素统计图确定特征参数基准点位置。然后以标定胚轴最小矩形倾斜度和宽度判定弯曲状态;子叶跨度通过两子叶端点距离确定,子叶展开角通过两子叶底端平展位置拟合线夹角判定;胚轴弯曲度通过胚轴中心线上曲率最大的位置为分界点分别判断两段斜度而求得,胚轴长、轴径结合斜度补偿求得。与手工测量数据对比,轴长、轴径和子叶跨度的相关系数分别为0.935 1、0.899 9和0.903 4,相对误差分别小于7%、5%和7%,绝对误差分别小于4 mm、0.2 mm和6 mm。     

给定两连杆运动面的Watt-I六杆机构空间解域综合方法

针对运动生成Watt-I型六杆机构综合问题,给出了一种基于空间解域的综合方法。该型机构中作为末端执行构件的连杆和一个四杆机构的连杆运动面相连,因此需要给出这个连杆运动面的位置。该方法首先综合四杆机构,然后将二杆组添加到四杆机构上。对于四位置问题,四杆机构综合可得到无穷多机构解,而对于任意四杆机构又有无穷多种添加二杆组的方法。本文以四杆机构的平面解域为基础,将综合所得无穷多六杆机构在一个有限的空间区域内表示出来,还给出一种能有效判断Watt-I型六杆机构回路和分支缺陷的方法。通过对综合所得无穷多机构解进行判断,建立六杆机构去除回路分支缺陷的空间可行域。空间解域的应用使设计者能够从整体上把握六杆机构的可行解的全部解空间,从而使获得附加更多条件的最优机构成为可能。     

基于热重法的植物油氧化动力学-甘油三酯相关性研究

通过热重(TG)及微商热重(DTG)法研究了6种常见植物油的热氧化特性,分析了不同温控程序对质量变化曲线的影响,通过微商热重曲线确定了不同升温速率下的氧化起始温度(T_(on))和氧化峰值温度(T_p),结合OzawaFlynn-Wall等转化率法对油脂的氧化动力学特性进行推导,并对油脂热氧化特性与其甘油三酯组成与分布的相关性进行了分析。研究结果表明:随着升温速率的增加,T_(on)和T_p均随之升高且T_p变化更为敏感;植物油的热力学特性与其甘三酯组成间具有不同程度的相关性,其中,T_(on)及T_(on)处活化能与高饱和度甘三酯呈正相关而与高不饱和度甘三酯呈负相关,T_p及T_p处活化能与各组分间具有相似的关系但相关性略有减弱,具体相关性与甘油骨架上脂肪酸链的种类有关。热重分析法所需样品微量、操作快捷、灵敏度高,在油脂加工及质量检测领域具有广阔的应用前景。     

Google Maps在农业WebGIS中的应用探索

针对目前国内农业WebGIS存在的信息单一、功能单调等问题,提出了将标准的地理信息系统开发平台和Google Maps同时应用于农业WebGIS的方法。利用Google Maps提供的开放地图接口自定义GoogleMap类可以为农业WebGIS系统提供详细的地理位置信息的显示、标注以及路线查询等功能。最后通过SuperMap.IS.NET开发平台与Google Maps实现对小汤山农业基地的信息管理,证明该方法实用可行。     

基于时空信息比较的温室环境传感器故障识别

为了提高温室环境测控系统中传感器数据的准确性,针对温室环境参数变化的时间相关性和空间相似性特点,提出了一种基于主成分分析(Principal component analysis,PCA)的故障检测与基于时空信息比较的温室环境监测系统的传感器故障识别方法。首先利用基于PCA的传感器故障检测方法,通过监控统计量T2和SPE的变化实现传感器系统故障检测;再针对检测出故障的传感器节点,对该时刻传感器节点采用基于时空特性的节点信息比较实现不同传感器的故障识别。分别对比基于时间尺度、空间尺度、时空尺度的节点信息比较方法对传感器故障识别的影响进行了分析与试验验证,验证结果表明:基于PCA的传感器故障检测方法能够有效地实现对传感器系统故障的初步检测,提出的基于时空信息比较的传感器故障识别方法,融合考虑时间尺度和空间尺度的节点信息,能够有效地实现传感器具体故障定位;所建立的传感器故障识别方法检测正确率CDR为98.37%、平均虚警率FAR为1.72%,较传统的传感器故障识别方法检测正确率CDR提高了22.067个百分点,而平均虚警率FAR则降低了15.762个百分点,能够有效地保证故障诊断效率、提高故障诊断精度、降低虚警率,具有可靠性和准确性。     

稻麦轮作制中秸秆钾与化肥钾利用的研究

采用盆栽试验,研究秸秆钾和化肥钾两种不同来源的钾素在作物和土壤中的吸收利用情况.结果表明:施用秸秆钾与化肥钾,稻麦轮作制中两季作物总产量基本相当;作物对秸秆钾与化肥钾的两季利用情况有较大差异,但两季总吸收量与利用率相差不大;作物施用秸秆钾当季应加强氮素的配合施用,特别是在小麦上;作物所吸收的秸秆钾与化肥钾有约80%以上贮存于秸秆中,只有约10%左右通过籽粒带走,另有约5%存在根茬.     

氮肥用量及施用时间对土体中硝态氮移动的影响

土连续两年小麦—玉米轮作条件下 ,播前一次施氮量 130～ 5 2 0kghm-2 a-1时 ,氮肥用量对硝态氮在土体中的移动深度没有影响 ,但土壤剖面中残留的硝态氮量随施氮量增加显著增加。播前一次施用氮肥 ,差减法计算的肥料氮表观回收率 (作物携出量和土壤硝态氮的残留量 )为 6 2 %～ 82 7% ;就作物而言小麦的携出率高于玉米 ,在玉米生长季节有更多的硝态氮可能被淋移至土壤剖面的下层。小麦—玉米轮作一年 ,不同的施氮时间对肥料氮的表观回收率以及硝态氮在土壤剖面中的分布、累计没有明显影响。土区合适的氮肥用量是控制硝态氮向深层移动的主要措施     

陕西省农田土壤硫含量空间变异特征及亏缺评价

陕西省 13种耕地土壤 30 5个表层 (0～ 2 0cm)土壤样点分析结果表明 :该省土壤全硫和用0 0 1molL-1Ca(H2 PO4) 2 浸提的有效硫含量分别为 33～ 76 9mgkg-1(均值 36 0mgkg-1)和 4 6～ 15 7 3mgkg-1(均值 2 9 3mgkg-1) ,其中黑垆土 (普通干润均腐土 )、绵砂土 (灌淤湿润砂质新成土 )的有效硫平均含量最低。土壤全硫具有中等空间相关性 ,有效硫空间相关性较差。土壤全硫和有效硫分布具有渐变性 ,其最大相关距离分别为 5 31km和 34km。全硫以较大区域变异为主 ;有效硫以小区域变异为特征。陕西省耕地土壤有效硫含量低于硫亏缺临界值 (18 5mgkg-1)的占总面积的 13 9% ,相当于 5 3 7万hm2 农田缺硫。陕西省中部地区 ,尤其是延安西部地区土壤硫亏缺风险较大     

Effects of rake angle of chisel plough on soil cutting factors and power requirements: A computer simulation

A computer simulation was conducted to predict the effects of rake angle of a chisel plough and soil bulk density on angle of soil failure plane, rupture distance, width of side crescent, frictional, overburden, cohesion and adhesion soil cutting factors, draft forces and drawbar power requirements. The experimental work was carried out in two locations. Soil of the first location was sandy clay with the soil bulk densities of 1.75 and 1.70 g/cm³ for firm and loose soil conditions, respectively, with an angle of internal friction of 30° and a surface friction angle of 20°, cohesion of 2.5 kN/m² and adhesion of 1.2 kN/m². Soil of the second location was clay loam with the soil bulk densities of 1.65 and 1.50 g/cm³ for firm and loose soil conditions, respectively, with an angle of internal friction of 34° and a surface friction angle of 23°, cohesion of 2.4 kN/m² and adhesion of 1.14 kN/m². The prediction showed that the angle of failure plane found to decrease with the rake angle. The rupture distance decreased with the rake angle from 15° to 55° and then increased as the rake angle increased over 55°. The width of the side crescent increased as the rake angle increased and the maximum value and the minimum value were recorded at 75° and at 15°. Values of frictional and overburden factors decreased as rake angle increased. The maximum and minimum values were recorded at 15° and 75°, respectively. The values of cohesion factor increased as rake angle increased. The maximum value was recorded at rake angle of 75° and the minimum value was recorded at rake angle of 15°. Adhesion factor was found to change inversely with the rake angle from 15° to 55° and then to change directly with the rake angle over 55°. The draft force decreased with the rake angle and reached its minimum value at 45° rake angle. Over 45°, the draft force increased and reached its maximum value at 75° rake angle. The draft increased with soil bulk density. The power required for moving the plough recorded the maximum value at rake angle of 15°, while the minimum value was recorded at 55° rake angle. The values of power increased with decrease of soil bulk density. The predicted values demonstrated some deviations from the experimental values of the draft force and the drawbar power.     

Fractal features of soil particle size distribution and the implication for indicating desertification

Some of the effects of land desertification on soil properties are manifested by the coarsening of the soil particlesize distribution (PSD) and the losses in organic C and nutrients. The changes and characteristics of PSD and selected chemical properties in soils at the 0-15 cm plough layer from different degrees of desertified croplands are analyzed in the semiarid Horqin Sandy Land, northern China. The fractal dimension of the PSD is emanated to characterize the patterns of PSD. The relationships between the fractal dimension of the PSD and selected soil properties are discussed. The results show that: (1) in the transformation from potential desertified cropland to extremely desertified cropland, the sand content at the 0-15 cm soil increased from 69% to 93%, organic C and total N contents decreased by 65% and 69%, respectively; (2) the fractal dimension of PSD ranged from 2.179 to 2.611, the more the contents of sand, the lower the fractal dimension and the higher the desertified degree of farmland. In the desertification process within the studied area, the mean fractal dimension decreased from 2.555 for the potential desertified soils to 2.298 for extremely desertified soils; (3) there existed considerable linear relationships between fractal dimension and soil properties. It was shown that fractal dimensions of PSD are useful parameters able to monitor soil degradation and to estimate the degree of soil desertification.