高分辨率遥感影像的边缘损失增强地物分割

目的 针对高分辨率遥感影像语义分割中普遍存在的分割精度不高、目标边界模糊等问题,提出一种综合利用边界信息和网络多尺度特征的边缘损失增强语义分割方法。方法 对单幅高分辨率遥感影像,首先通过对VGG-16（visual geometry group 16-layer net）网络引入侧边输出结构,提取到图像丰富的特征细节;然后使用深度监督的短连接结构将从深层到浅层的侧边输出组合起来,实现多层次和多尺度特征融合;最后添加边缘损失增强结构,用以获得较为清晰的目标边界,提高分割结果的准确性和完整性。结果 为了验证所提方法的有效性,选取中国北方种植大棚遥感影像和Google Earth上的光伏板组件遥感影像进行人工标注,并制作实验数据集。在这两个数据集上,将所提方法与几种常用的语义分割方法进行对比实验。实验结果表明,所提方法的精度在召回率为00.9之间时均在0.8以上,在2个数据集上的平均绝对误差分别为0.079 1和0.036 2。同时,通过消融实验分析了各个功能模块对最终结果的贡献。结论 与当前先进方法相比,本文提出的边缘损失增强地物分割方法能够更加精确地从遥感影像的复杂背景中提取目标区域,使分割时提取到的目标拥有更加清晰的边缘。     

Comparing interpolation techniques for monthly rainfall mapping using multiple evaluation criteria and auxiliary data sources: A case study of Sri Lanka

Interpolating climatic variables such as rainfall is challenging due to the highly variable nature of meteorological processes, the effects of terrain and geography, and the difficulty in establishing a representative network of stations. While interpolation models are being adapted to include these effects, often the rainfall data contain significant gaps in coverage. In this paper, we evaluated rainfall data from an agro-ecological monitoring network for producing maps of total monthly rainfall in Sri Lanka. We compared four spatial interpolation techniques: inverse distance weighting, thin-plate splines, ordinary kriging, and Bayesian kriging. Error metrics were used to validate interpolations against independent data. Satellite data were used to assess the spatial pattern of rainfall. Results indicated that Bayesian kriging and splines performed best in low and high rainfall, respectively. Rainfall maps generated from the agro-ecological network were found to have accuracies consistent with previous studies in Sri Lanka.     

基于邻居决策的协同过滤推荐算法

协同过滤技术应用于个性化推荐系统中,稀疏性问题和可扩展性问题成为亟需解决的问题。针对传统方法的不足,提出一种凭借邻居数做决策的方法,比较各个待测位置的用户邻居数和项目邻居数,由数量多的一方作预测,同时对预测值判定给出一种合理而有效的度量方法。实验结果表明,该方法能够提高推荐质量。     

用户间多相似度协同过滤推荐算法

传统的User-based协同过滤推荐算法仅采用了单一的评分相似度来度量用户之间对任何项目喜好的相似程度。然而根据日常经验,人们对不同类型事物的喜好程度往往是不同的,单一的评分相似度显然无法准确描述这种不同。针对上述问题,提出了一种基于用户间多相似度的协同过滤推荐算法,即基于用户间对不同项目类型的多个评分相似度来计算用户对未评分项目的预测评分。实验结果表明,该算法可以有效地提高预测评分的准确性及推荐质量。     

Influences of copolymers (Copovidone,Eudragit RL PO and Kollicoat MAE 30 DP) on stability and bioactivity of spray-dried and freeze-dried lysozyme

Protein stability is the most crucial factor in protein pharmaceutical preparations. Various techniques were applied for producing stable protein formulations such as spray-drying and freeze-drying. However, heating and freezing stresses are disadvantages for proteins using these methods, respectively. Accordingly, excipients have been used to preserve therapeutic effects of proteins during processing and for long period of time. Therefore, influences of Copovidone, Eudragit^® RL-PO and Kollicoat^® MAE-30 DP (as excipients) on stability and integrity of lysozyme (as a model protein) in spray-dried and freeze-dried forms were investigated. Protein formulations in both dried forms were prepared without and with the addition of mentioned excipients at different concentrations. Protein formulations were characterized for yield determination, morphology using scanning electron microscopic (SEM), thermal analysis by Differential Scanning Calorimetry (DSC), secondary structure stability using Fourier transform infrared (FT-IR) spectroscopy and biological activity. All protein formulations were subjected to a stability study as solid protein formulations for 3 weeks at 24?°C/76% relative humidity and aqueous protein samples were stored at 50?°C for 30?min in a water bath. Results showed that Copovidone successfully preserved integrity and biological activity of lysozyme before and after storage in both spray-dried and freeze-dried forms with more advantage for using higher concentration of the same excipient. Smooth spheres of spray-dried lysozyme formulations with Copovidone were smaller than spray-dried lysozyme without and with Kollicoat^® MAE-30 DP, which affected %yield produced. Copovidone has demonstrated valuable protection ability for lysozyme.     

A rapid method for simultaneous determination of triterpenoid saponins in Pulsatilla turczaninovii using microwave-assisted extraction and high performance liquid chromatography–tandem mass spectrometry

In the present study, a microwave-assisted extraction (MAE) and high performance liquid chromatography–tandem mass spectrometry (HPLC–MS/MS) method was developed for the simultaneous determination of seven triterpenoid saponins in Pulsatilla turczaninovii. Operational conditions of MAE were optimized using a central composite design. Multiple-reaction monitoring was employed for quantification while switching the electrospray ion source polarity between positive and negative modes. Full validation of the method was carried out (linearity, precision, accuracy, limit of detection and limit of quantification). The results indicated that the method was rapid, specific and reliable. The developed method was successfully applied to determine the investigated compounds in nine batches of P. turczaninovii. These samples were collected in different seasons from the Liaoning province and varied greatly. The total triterpenoid saponins content was the highest in April, the seedling stage, and decreased from May to August. The contents of total triterpenoid saponins decreased significantly after the blooming period.     

微波辅助提取墨旱莲总黄酮的工艺优化

采用微波辅助提取墨旱莲中的总黄酮。以总黄酮得率为指标,考察微波辐射时间、液固比、溶剂pH、提取级数等因素对得率的影响,确定了总黄酮的最佳提取工艺条件:以去离子水为溶剂,液固比(V/m)=25 mL/g,微波提取210 s,提取2次,此时墨旱莲总黄酮得率为1.42%,而常规水提取的时间为60 min,得率1.24%,故微波辅助提取法高效,省时。     

酵母细胞中谷胱甘肽的微波辅助提取

采用微波辅助方法提取了酵母细胞中的谷胱甘肽。研究表明 ,在 65 0W功率下对酵母泥直接微波处理 1 2s ,再以 2 0mL/g的物料比提取 1 0min ,或者物料比为 1 5mL/g的酵母悬浮液在 65 0W下处理 60s,都能有效地提取谷胱甘肽 ,提取率分别达到 2 0 89mg/g和 2 1 1 7mg/g。和其他提取方法比较 ,微波辅助提取是一种节省能源、快速、简便、有效的方法。     

Determination of pesticide residues in food matrices using the QuEChERS methodology

The determination of pesticide residues in food matrices is a formidable challenge mainly because of the small quantities of analytes and large amounts of interfering substances which can be co-extracted with analytes and, in most cases, adversely affect the results of an analysis. However, safety concerns require that pesticides of the wide range of chemical properties (including acidic, basic and neutral) should be monitored. Because of the wide variety of food matrices, the sample must initially be cleaned up before final analysis. That is why the analytical chemist is faced with the need to devise new methodologies for determining such residues to be determined in a single analytical run. To accomplish the goal, QuEChERS methodology has been developed. It is a streamlined and effective extraction and cleanup approach for the analysis of diverse analyte residues in food matrices. So far, there have been achieved promising results by liquid or gas chromatography analysis, including pesticides, but also acrylamide, pharmaceuticals and veterinary drugs.     

微波辅助法提取紫玉米植株花青素工艺条件优化研究

以紫玉米植株为材料,研究微波辅助法提取紫玉米花色苷的工艺条件,确定微波条件下花色苷最佳提取条件。结果表明,微波功率是影响花色苷提取的主因素,各因子对提取率的影响依次为:微波功率,微波辐射时间,提取温度,料液比;最佳条件为:微波辐射功率700w,辐射时间6min,提取温度60℃,料液比1∶20(g:mL)。与传统的浸提法相比,时间由24h缩短为7min,色素产率由12.24%增加到16.87%。微波辅助提取色素具有提取时间短、减少环境污染、节能省耗、提取率高等优点。