A strategy for allowing meaningful and comparable scores in approximate matching

Approximate data matching aims at assessing whether two distinct instances of data represent the same real-world object. The comparison between data values is usually done by applying a similarity function which returns a similarity score. If this score surpasses a given threshold, both data instances are considered as representing the same real-world object. These score values depend on the algorithm that implements the function and have no meaning to the user. In addition, score values generated by different functions are not comparable. This will potentially lead to problems when the scores returned by different similarity functions need to be combined for computing the similarity between records. In this article, we propose that thresholds should be defined in terms of the precision that is expected from the matching process rather than in terms of the raw scores returned by the similarity function. Precision is a widely known similarity metric and has a clear interpretation from the user's point of view. Our approach defines mappings from score values to precision values, which we call adjusted scores. In order to obtain such mappings, our approach requires training over a small dataset. Experiments show that training can be reused for different datasets on the same domain. Our results also demonstrate that existing methods for combining scores for computing the similarity between records may be enhanced if adjusted scores are used.     

The role of spectral resolution and classifier complexity in the analysis of hyperspectral images of forest areas

Remote sensing hyperspectral sensors are important and powerful instruments for addressing classification problems in complex forest scenarios, as they allow one a detailed characterization of the spectral behavior of the considered information classes. However, the processing of hyperspectral data is particularly complex both from a theoretical viewpoint [e.g. problems related to the Hughes phenomenon (Hughes, 1968) and from a computational perspective. Despite many previous investigations that have been presented in the literature on feature reduction and feature extraction in hyperspectral data, only a few studies have analyzed the role of spectral resolution on the classification accuracy in different application domains. In this paper, we present an empirical study aimed at understanding the relationship among spectral resolution, classifier complexity, and classification accuracy obtained with hyperspectral sensors for the classification of forest areas. We considered two different test sets characterized by images acquired by an AISA Eagle sensor over 126 bands with a spectral resolution of 4.6 nm, and we subsequently degraded its spectral resolution to 9.2, 13.8, 18.4, 23, 27.6, 32.2 and 36.8 nm. A series of classification experiments were carried out with bands at each of the degraded spectral resolutions, and bands selected with a feature selection algorithm at the highest spectral resolution (4.6 nm). The classification experiments were carried out with three different classifiers: Support Vector Machine, Gaussian Maximum Likelihood with Leave-One-Out-Covariance estimator, and Linear Discriminant Analysis. From the experimental results, important conclusions can be made about the choice of the spectral resolution of hyperspectral sensors as applied to forest areas, also in relation to the complexity of the adopted classification methodology. The outcome of these experiments are also applicable in terms of directing the user towards a more efficient use of the current instruments (e.g. programming of the spectral channels to be acquired) and classification techniques in forest applications, as well as in the design of future hyperspectral sensors.     

基于HMRF先验模型的超分辨率重建

针对基于最大后验概率(MAP)的超分辨率重建算法在重建图像过程中存在的问题,提出一种基于Huber-马尔可夫随机场(HMRF)先验模型的超分辨率重建方法,采用HMRF作为图像先验模型,对图像进行分段超分辨率重建。仿真实验结果表明,与传统的MAP算法相比,该方法能更好地保存重建图像的边缘细节,有效提高重建图像的质量。     

面向自然语言文本的否定性与不确定性信息抽取

目前,信息抽取研究主要面向肯定性信息,而自然语言文本中包含了大量否定性和不确定性信息,为了将此类信息与肯定性信息区分开,有必要针对否定性与不确定性信息抽取进行深入研究.针对这一任务,首次构建了一个16 841句的汉语语料资源,利用序列标注模型与卷积树核模型,系统地探索了各种序列化依存特征和结构化句法树特征的有效性,并提出了元决策树模型,对二者进行融合.实验结果显示,该方法在否定性和不确定性信息抽取任务上的精确率分别达到69.84%和58.57%,为相关研究打下了坚实的基础.     

A small‐area and low‐power data driver IC using two‐stage DAC with a capacitor array for active matrix flat‐panel displays

A small‐area and low‐power data driver integrated circuit (IC) using a two‐stage digital‐to‐analog converter (DAC) with a capacitor array is proposed for active matrix flat‐panel displays. The proposed data driver IC employs a capacitor array in the two‐stage DAC so as to reduce the DAC area and eliminate the need for a resistor string, which has high‐power consumption. To verify the proposed two‐stage DAC, a 20‐channel data driver IC with the proposed 10‐bit two‐stage DAC was fabricated using a 0.18‐μm complementary metal–oxide–semiconductor process with 1.8 and 6 V complementary metal–oxide–semiconductor devices. The proposed 10‐bit two‐stage DAC occupies only 43.8% of the area of a conventional 10‐bit two‐stage DAC. The measurement results show that the differential nonlinearity and integral nonlinearity are +0.58/?0.52 least significant bit and +0.62/?0.59 least significant bit, respectively. The measured interchannel deviation of the voltage outputs is 8.8 mV, and the measured power consumption of the 20‐channel data driver IC is reduced to 7.1 mW, which is less than half of the power consumed by the conventional one.     

Development of active matrix LCD for use in high‐resolution adaptive headlights

Conventional adaptive driving beam headlamps are limited in achieving still higher quantities of switchable pixels by the number of LEDs and movable elements needed. In this paper, it is shown that by integrating an active matrix liquid crystal display module, it is possible to realize fully adaptive high‐resolution headlights without mechanical elements and a finite number of LED with 30 k switchable pixels.     

一种基于XML的非结构化数据转换方法

XML作为半结构化的语言,因其能预先定义标记等优势被普遍应用于非结构化到结构化信息的转换中。利用POI技术把网络上繁杂的非结构化数据转化为XML半结构化数据,把半结构化数据转化为结构化数据,使用户能够简便地查询所需信息。通过实验对SAX,DOM的解析效率进行了对比,实验表明解析相同大小的XML文件,SAX比DOM效率更高,而且此种差距会随着XML文件的增大而逐渐增大。     

小型高分辨率中阶梯光栅光谱仪光学设计

为了克服传统光栅光谱仪小体积与高分辨率无法共存的缺点,提出一种以中阶梯光栅为核心色散元件、反射棱镜为辅助色散元件的光谱仪光学设计方法.具体分析了中阶梯光栅的基本原理,从光路设计和元件参数设计两方面进行仪器光学设计,最后设计了一个焦距为150 mm、光谱范围为200 nm～600 nm、理论光谱分辨能力为0.1 nm的中阶梯光栅光谱仪,仿真和实验表明,设计的光谱仪在Hg灯253.652 nm处光谱分辨能力为0.023 9 nm,在Hg灯546.074 nm处光谱分辨能力为0.061 nm,满足设计指标要求.     

高精度智能质子磁力仪的最新发展

本文从质子磁力仪核心技术的角度对高精度智能质子磁力仪的最新发展进行了科学的论述,总结了我国及当今该仪器领域的最新成果,为我国广大地磁勘探工作者选购高质量的质子磁力仪从事物探工作提供科学指南。     

Tool path smoothing of a redundant machine: Application to Automated Fiber Placement

Automated Tape Laying and Fiber Placement of composite materials are the two principal automated processes used for fabrication of large composite structures in aeronautical industry. The aluminum parts produced by High Speed Machining tend to be replaced by carbon fiber composite parts realized with these processes. However, structural parts present reinforcement zones which disturb the tool path follow-up and generate an increase of the manufacturing time. Thus, this paper deals with the optimization of tool paths of a 7-axis machine tool of Fiber Placement with the objective of reducing the manufacturing time while ensuring the requested quality of the final part. In this paper, two complementary methods are detailed. The first method takes advantage of the degree of redundancy of the machine tool to decrease the kinematic loads of the control joints. The second method aims to smooth the orientation of the machine head along the tool path while ensuring quality constraints. These two methods are then applied on a test tool path and bring to a significant decrease of the manufacturing time (32.9%).