水稻不同叶位层物理结构与冠层反射光谱的定量研究

 研究了不同土壤水分、施氮条件下水稻冠层反射光谱与冠层生长特征（叶面积指数、叶片干质量和鲜质量）的量化关系。结果表明，不同层次叶面积指数、叶片鲜质量和叶片干质量与冠层光谱的相关性大小均为：L12345>L1234>L123>L12>L1，而顶部5张叶片的面积和生物量（干、鲜质量）对冠层特征反射光谱的贡献大小顺序为：L2>L1>L3>L4>L5，顶部2张叶片对冠层光谱作用最大。发现比值指数R(760,710) 与水稻叶面积指数呈极显著线性相关；而比值指数R(1650,1100) 与水稻地上部鲜质量和干质量均呈极显著幂函数相关关系。因此，比值指数R(760,710) 和R(1650,1100) 可分别用来定量反演水稻叶面积信息和地上部生物量信息。     

基于分数阶微分光谱指数的冬小麦根域土壤含水率估算模型

为探讨分数阶微分（fractional-order differentiation,FOD）技术联合光谱指数改善高光谱反演冬小麦根域土壤含水率（soil moisture content,SMC）的效果,该研究以冬小麦为研究对象,测取高光谱反射率和土壤含水率数据,将高光谱反射率经Savitzky-Golay(SG)平滑处理后计算典型光谱指数以此构建偏最小二乘回归（partial least squares regression,PLSR）、随机森林（random forest,RF）和BP神经网络（back propagation neural network,BPNN）3种土壤含水率反演模型;将高光谱反射率进行0～2.0阶（步长为0.2）的分数阶微分处理后计算比值指数（ratio index,RI）和归一化指数（normalized difference index,NDI）,分析不同阶的RI、NDI与SMC之间的二维相关性,筛选得出敏感光谱指数并分组,以此构建3种反演模型（PLSR、RF和BPNN）。结果表明：不同典型光谱指数与土壤含水率的相关性存在很大差异,相关系数波动范围在0....     

基于RGB图像光谱重建的鱼糜掺假定量检测研究

实现基于RGB图像的光谱重建对降低光谱的硬件要求、扩大其实际应用具有重大意义。该研究以鱼糜掺假检测为例,比较多元多项式最小二乘回归算法（polynomial multivariate least-squares regression,PMLR）与深度学习HRNet网络对光谱重建的性能,建立基于重建光谱多种掺假鱼糜检测模型并验证其实际应用的有效性。结果表明,2种方法的重建光谱误差较小,HRNet网络、PMLR算法重建光谱的均方根误差（root mean square error,RMSE）分别为0.010 4和0.012 6,大多数掺假检测模型有较高的预测准确性,其预测相关系数大于0.91,预测均方根误差小于9%。在基于重建光谱建立的掺假检测模型中,效果最佳的是基于PMLR算法重建光谱使用标准正态变量变换（standard normal variate,SNV）预处理的极限学习机回归模型,其预测均方根误差为3.954 4%、预测相关系数为0.983 0。因此,PMLR算法和HRNet网络均能较好的实现基于RGB图像的光谱重建,且重建光谱均能实现对鱼糜掺假样本的较好检测结果,为基于重建光谱的食品和农产品品质与安全检测提供了新思路。     

LANDSAT IMAGES REVEAL TRENDS IN THE AEOLIAN DESERTIFICATION IN A SOURCE AREA FOR SAND AND DUST STORMS IN CHINA'S ALASHAN PLATEAU (1975–2007)

The development and reversal of aeolian desertification have been widely recognised as the dominant causes of changes in the source areas of dust storms. The Alashan Plateau, which is near China's Hexi Corridor, is one of the major dust storm source regions in northern China. To investigate the relationship between climatic and anthropogenic factors and the development of aeolian desertification, we reconstructed the temporal and spatial evolution of aeolian desertification and the underlying processes from 1975 to 2007 using Landsat images and analysed the driving forces. The results show that aeolian desertified land in the study area covered 32 847·79 km² in 2007, amounting to 11·3 per cent of the total area, with an initial period of desertification expansion from 1975 to 2000 and a reversal of desertification from 2000 to 2007. On the basis of the analysis of effects of climate changes and human activities in the region, aeolian desertification was principally driven by human activities in this area; climatic variations had little effect on the area of severe desertification. Copyright © 2011 John Wiley & Sons, Ltd.     

不同土壤水分条件下玉米叶片/冠层气孔导度的光谱监测模型

通过玉米水分控制试验,测定不同水分条件下各生育期叶片气孔导度、叶面积指数和冠层光谱反射率等,以分析玉米叶片气孔导度的变化规律及其与光谱植被指数的相关性,从而建立基于光谱植被指数和土壤湿度的叶片气孔导度模型。结果表明：玉米在可见光区和近红外中、长波区的反射率随着土壤水分的降低而上升,但叶片气孔导度（Gs）、叶面积指数（LAI）、比值植被指数（RVI）和归一化植被指数（NDVI）随着土壤水分的下降而降低;玉米NDVI和RVI与单叶片和冠层气孔导度均呈极显著指数函数关系（P〈0.01）,且对单叶片气孔导度的拟合效果优于对冠层导度的拟合效果,而经土壤湿度订正的RVI监测模型优于NDVI监测模型。表明通过测定冠层反射光谱率可实时、迅速地定量监测玉米叶片的气孔导度,为大面积作物气孔导度估算奠定基础。     

基于改进湍流模型和偏振成像技术的水下退化图像复原方法

为了满足工厂化水产养殖中过程控制的信息化要求,该文提出一种基于改进湍流模型结合偏振成像技术,具有较强鲁棒性的水下退化图像复原方法。考虑内外尺度对波结构函数影响,结合折射率谱,改进水下湍流退化模型以提高复原算法先验知识的完备性;基于改进的退化模型和水下前向散射光的偏振特性,利用偏振成像技术提取退化图像中的噪声特征;基于退化图像噪声特征,采用约束最小二乘滤波法进行退化图像复原;最后,对复原效果进行相应比较,结果表明在强湍流条件下本算法具有更为理想的复原效果。该研究可为复杂水流条件下水下退化图像复原方法研究提供参考。     

基于土壤光谱特征的宁夏银北地区盐碱地盐分预测研究

对宁夏回族自治区银北地区盐碱地野外土壤表层光谱反射率和土壤全盐及盐分进行定量分析,筛选出各土壤盐分指标的敏感波段,然后采用全回归和逐步回归的方法建立各盐分的预测模型。结果表明：表层土壤高光谱反射率r,及其平滑去嗓处理后的值R,lg(R)与全盐含量呈极显著正相关关系,1/R,lg(1/R)与全盐呈极显著负相关关系,(R)'和〔lg(1/R)〕'在特定单波段处表现较佳;土壤表层光谱反射率与CO₃^2-的相关性最强,其次是SO₄^2-;土壤光谱反射率与Na⁺的相关性在各种变换方法下均较强,其次为Mg²⁺,与Ca²⁺的相关性最弱。基于R的逐步回归方程为全盐含量预测的最佳模型;基于土壤光谱反射率拟合土壤CO₃^2-的准确度略高于对土壤HCO₃^-;敏感波段估测土壤SO₄^2-含量的决定系数明显高于其他阴离子;采用〔lg(1/R)〕'逐步回归得到的方程拟合土壤Na⁺,K⁺和Mg²⁺含量相对于其他变换方式效果更理想。预测模型中对土壤全盐和Na+的模型精度较高,预测能力强;光谱对土壤SO₄^2-和Mg²⁺的预测能力也较强;对土壤Cl^-和Ca²⁺的预测稳定性、预测能力和精度都较差。     

Abstracts of nippon Dojyohiryogaku Zasshi

Color change of city refuse during composting process was investigated according to the methods of measurement for color of materials based on the CIE 1931 Standard Colorimetric System. Stimulus value Y (the degree of lightness) and chromaticity coordinates (x, y) were determined with Color Analyzer by measuring relative spectral reflectance. Stimulus value Y of city refuse decreased during composting process, but chromaticity coordinates (x, y) scarcely changed.

Color of various composts, which were produced from city refuse, straw, hog fecal wastes, tree bark, and tree bark mixed with activated sludge, were also investigated by measuring relative spectral reflectance. The shapes of the reflection spectra of city refuse were different from those of the other composts. Colors of the various composts were similar to each other when specified according to their three attributes: value, hue, and chroma (Munsell renotation).

While city refuse was rotting and maturing, stimulus value Yand C/N ratio equally decreased. A positive correlation was found between stimulus value Y and C/N ratio. It was concluded that stimulus value Y can be used as a criterion for determining the degree of maturity of city refuse compost.

The correlation between stimulus value Y and C/N ratio of various composts was also investigated. According to the position on the two coordinates having stimulus value Y and C/N ratio as axe s, various composts were classified into three groups: (i) city refuse compost group, (ii) straw compost group, and (iii) tree bark compost group.     

Systolic time intervals assessed by 2-D echocardiography and spectral Doppler in the horse

In the present study, the ‘acoustic windows’ for the measurement of the left ventricular systolic time intervals is studied by means of 2‐D echocardiography and cardiac Doppler ultrasonography, and the normal values in the horse (n = 112) are determined. The left ventricular isovolumetric contraction time, the pre‐ejection period (PEP), as well as the left ventricular ejection time (LVET) have been measured, and the values of the left ventricular total electromechanical systole (LVTES) and the PEP‐to‐LVET ratio have been calculated. It has been established that the most suitable window for the measurement of the aforementioned indices in 2‐D echocardiography is the right parasternal window in a view in short axis at the level of the cardiac base to measure the aortogram. In Doppler ultrasonography, the preferred window is the left parasternal using the five chambers apical view. The following values have been acquired: PEP = 0.071 ± 0.01 s; LVET = 0.532 ± 0.097 s; LVTES = 0.6 ± 0.1 s and PEP‐to‐LVET = 0.138 ± 0.025 measured by 2‐D echocardiography and PEP = 0.068 ± 0.009 s; LVET = 0.527 ± 0.076 s; LVTES = 0.598 ± 0.098 s and PEP‐to‐LVET = 0.131 ± 0.01 measured by Doppler ultrasonography.     

杂交稻与常规稻的光谱特性差异初步分析

通过杂交晚稻和常规晚稻的反射光谱试验,发现杂交稻由于其生长发育特性及株型的差异,在整个生长期内其反射光谱特性与常规晚稻有着明显的差异。