眼球的自动定位

提出了一种由粗到细自动定位人脸的一个重要特征即眼球圆心的方法，该方法综合运用局部二值化，霍夫变换，像纱聚类，边缘提取等技术，对均匀或非均匀光照下拍摄或戴眼镜的照片都能获得很高的定位精度，适用于处理证件照。     

利用马铃薯淀粉下脚液生产饮料工艺的研究

本实验尝试从马铃薯中提取蛋白质等成份，并适当添加姜汁，制成一种新型饮料，对饮料生产的工艺条件作了详细探讨，成功地解决了马铃薯褐变及沉淀等问题，确定了最佳配方及工艺。     

指纹细节特征提取与剪枝

指纹图像经预处理后，由于各种原因，会产生数以百计的特征。为了图像匹配的精确和快速，提出一套在细化后的指纹点线图上提取细节特征并对其剪枝的方法，并根据细节特征的拓扑特性，设计了验证其可靠性的算法。通过实验，证明了该后处理过程在去除指纹图像伪特征方法面具有良好的效果。     

Separation of di- and tripeptides with solvent extraction and an emulsion liquid membrane

The extraction equilibria of various di- and tripeptides with di-2-ethylhexylphosphoric acid (D2EHPA) were studied at low pH values. The complex extracted to organic phase consisted of one molecule of peptide and two molecules of D2EHPA dimer. The extraction constants of the peptides correlated well with the distribution coefficients of peptides between 1-octanol and water, which is a measure of hydrophobicity. The permeation rates of peptides through an emulsion liquid membrane were examined by using D2EHPA as a carrier, Span 80 as an emulsifier and kerosene as a diluent. The rates varied considerably with peptide type, depending upon the hydrophobicity.     

利用锁模光纤激光器做时钟提取器的OTDM通信系统

简短评述了大容量和超大容量光通信系统的ＷＤＭ和ＯＴＤＭ技术，提出了一个在ＯＴＤＭ通信系统中用锁模光纤激光器做时钟提取器的新方法，给出了一个利用设置波分预标时钟脉冲方案构成的新的全光型ＯＴＤＭ通信实验系统结构。     

填料萃取塔液泛速度的计算

介绍了三种有代表性的填料萃取塔的液泛速度计算公式，并通过它们用于三种典型的液-液实验体系和两种不同填料的计算结果的比较，对这些公式的适用范围进行了讨论。     

聚合铝生产工艺研究

以铝矿为原料，在常压下，采用两段溶出法，对聚合铝的合成工艺进行了系统研究。     

Extraction of Anthocyanin Pigments from Purple Sunflower Hulls

The anthocyanin pigments in hulls of purple sunflower seeds were extracted using three solvent systems, 50:1:49 ethanol-acetic acid-water (EAW), 0.01M acetic acid (AAc), or water containing SO₂. Solvent, extraction time, size of ground hulls, pH of extracting solvent, hulls/solvent ratio, and concentration of SO₂ in water, were investigated. All were significant factors affecting yield of extracted anthocyanins. EAW and SO₂ were more effective extractants than AAc. Optimum conditions from response surface optimization for the extraction with sulphurous water were: SO₂ at 200 mg/L, hull size 20 mesh, extraction time 5 min at 22°C, and solvent/hulls ratio = 20.     

Post-Supervised Template Induction for Information Extraction from Lists and Tables in Dynamic Web Sources

Dynamic web sites commonly return information in the form of lists and tables. Although hand crafting an extraction program for a specific template is time-consuming but straightforward, it is desirable to automatically generate template extraction programs from examples of lists and tables in html documents. Supervised approaches have been shown to achieve high accuracy, but they require manual labelling of training examples, which is also time consuming. Fully unsupervised approaches, which extract rows and columns by detecting regularities in the data, cannot provide sufficient accuracy for practical domains. We describe a novel technique, Post-supervised Learning, which exploits unsupervised learning to avoid the need for training examples, while minimally involving the user to achieve high accuracy. We have developed unsupervised algorithms to extract the number of rows and adopted a dynamic programming algorithm for extracting columns. Our method achieves high performance with minimal user input compared to fully supervised techniques.