“一分钟游程”构园法则的适用性研究

“一分钟游程”构园法则将人的视觉感受与园林组景共同考虑,以25~30 m作为组景转换的理想距离,对游路结构进行反复烘托与不断强化,从而形成柔和、丰富的“构景曲线”。选择南北方各具代表性的苏州沧浪亭和北京恭王府花园,分析其园林的景观布局,对比园林空间序列的“起”“承”“转”“合”间距。结果表明,“一分钟游程”构园法则在南方私家园林及北方皇家园林中同样适用,为今后的游线组织与组景布设提供了设计参考。     

面向养殖水体分布差异的COD光谱法检测研究

面向养殖水体，传统光谱法对化学需氧量（Chemical Oxygen Demand，COD）检测模型构建的基础：源域（现有样本库）与目标域（检测地水体）间光谱数据独立同分布。但是当源域与目标域分布间存在差异时，由源域得到的低误差模型常在目标域上表现下滑。针对该问题，提出面向UV Vis光谱的域对抗训练网络（DAUVwpNet），将分布不同的源域和目标域数据映射至相同分布的特征空间中，使其在该空间的分布距离尽可能接近，从而在特征空间中对源域训练的目标函数也可以迁移至目标域上，以降低模型在目标域的误差。试验表明：面向同一批测试数据，DAUVwpNet的预测误差为0.78，要低于传统模型的预测误差（0.85）；DAUVwpNet预测值与实测值间相关系数为0.95，要高于传统模型的相关系数（0.89）。表明了该网络能够较好对齐两域特征空间数据分布，降低因分布差异带来的COD检测误差。     

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。     

农网配电线路损耗分析

配电网起着输送和分配电能的作用,线路损耗给供电企业造成很大的成本投入,降低能源的利用率。分析配电网线路损耗的原因,提出降低线路损耗的具体措施,以期为配电网节能降耗、安全运行提供参考。     

柞蚕雄蛾浓缩液对肝细胞的保护和对高脂血症的治疗效果

为探讨柞蚕雄蛾浓缩液的保健和医用价值,分别就柞蚕雄蛾浓缩液对小鼠的保肝降酶作用和对半胱氨酸所致大鼠高脂血症的影响进行了实验。结果表明：该浓缩液可明显降低四氯化碳中毒小鼠血清中的谷丙转氨酶水平,因而能减少肝细胞变性、坏死的程度和范围,对受损伤的肝组织具有明显保护作用;可明显减轻肝组织脂肪样变性及降低由半胱氨酸所致高脂血症大鼠血清中的胆固醇、甘油三酯、低密度脂蛋白浓度,提高高密度脂蛋白浓度,因而对于高脂血症有良好的治疗效果。     

柞蚕雄蛾油对大鼠辅助降血脂功能的影响研究

为研究柞蚕雄蛾油对小鼠血脂代谢的影响,进行了雄性Wistar大鼠高脂饲料与不同剂量柞蚕雄蛾油的喂养试验,每4天称体重一次,每4天测定小鼠肝系数及血清总胆固醇（TC）、甘油三酯（TG）、高密度脂蛋白胆固醇（HDL‐C）和低密度脂蛋白胆固醇（LDL‐C）一次。结果表明：3个剂量组TC、TG、LDL‐C含量与高脂模型组比较显著低于高脂模型组,而HDL‐C显著高于模型组,具有显著性差异（P＜0．05）。     

Dietary protein requirement for juvenile mandarin fish,Siniperca scherzeri

An 8‐week feeding trial was conducted to investigate the dietary protein requirement of juvenile mandarin fish, Siniperca scherzeri, a demersal freshwater piscivore with high commercial value in East Asia. Five isolipidic (100 g/kg) and isoenergetic (20 MJ/kg) practical diets were formulated to contain graded levels of 450, 500, 550, 600, and 650 g/kg crude protein, using mackerel fishmeal as the sole source of protein. Each of the five test diets was fed to visual satiety to triplicate groups of fish (initial mean body weight ± SE; 20.1 ± 0.2 g) reared in a recirculating freshwater system over the experimental period. Fish growth performance in terms of weight gain tended to improve with increasing dietary protein level, and the highest values were observed in those fish fed the 600–650 g/kg diets. Feed intake significantly declined as the protein content of the diet increased. Feed efficiency and protein retention tended to increase with increasing dietary protein levels, from 450 to 600 g/kg, and then declined when dietary protein content further increased to 650 g/kg. A similar trend was also found for the protein efficiency ratio, although the observed changes did not reach statistical significance. Whole‐body composition and plasma biochemical parameters were not significantly affected by the dietary protein content. Based on these findings and a broken‐line model of growth, 614.4 g/kg seems to be the proper amount of protein in a practical diet for >20 g mandarin fish.     

A genetic tool for evaluating male‐mediated stock introgression in Atlantic salmon

1. The Atlantic salmon shows strong population structure with reduced gene flow, owing to homing behaviour. Supportive breeding with allochtonous parr or fish‐farm escapees could affect native population genetic structure by the introgression of foreign genes. Mature male parr are potentially powerful vehicles for promoting introgression: they can fertilize eggs in competition with anadromous males. As a consequence, foreign males contribute more to introgression than foreign females. This effect must be taken into account for the correct development of conservation programmes. Foreign males constitute an inadvertent way to disrupt the genetic structure, as the allochtonous parr or fish‐farm escapees can survive in the river during the parr stage.
2. A short polymorphic fragment of the sex‐determining gene (sdY) was used as a marker to assess introgression via males into south European salmon populations. In order to find haplotype variants for the intronic sequence of sdY, samples were investigated from 16 different rivers across the distribution range of the salmon, together with historical samples (1950–1960) from three Spanish rivers.
3. Two novel haplotypes, in addition to the three already described in Atlantic salmon, were found for this locus. Most samples, including historical ones, displayed the common haplotype (D) previously found in salmon all over Europe; however, north European haplotype variants were also detected in salmon inhabiting the southern European rivers that were systematically supplemented with foreign stocks between 1981 and 1994.
4. Analysis of haplotype variation in the sdY intron constitutes a powerful and inexpensive tool, not only to specifically assess male‐mediated introgression into natural populations of Atlantic salmon, but also for monitoring salmon escapees.