Subspace distance analysis with application to adaptive Bayesian algorithm for face recognition

We propose subspace distance measures to analyze the similarity between intrapersonal face subspaces, which characterize the variations between face images of the same individual. We call the conventional intrapersonal subspace average intrapersonal subspace (AIS) because the image differences often come from a large number of persons. An intrapersonal subspace is referred to as specific intrapersonal subspace (SIS) if the image differences are from just one person. We demonstrate that SIS varies significantly from person to person, and most SISs are not similar to AIS. Based on these observations, we introduce the maximum a posteriori (MAP) adaptation to the problem of SIS estimation, and apply it to the Bayesian face recognition algorithm. Experimental results show that the adaptive Bayesian algorithm outperforms the non-adaptive Bayesian algorithm as well as Eigenface and Fisherface methods if a small number of adaptation images are available.     

D-side: A facility and workforce planning group multi-criteria decision support system for Johnson Space Center

“To understand and protect our home planet, to explore the universe and search for life, and to inspire the next generation of explorers” is NASA's mission. The Systems Management Office at Johnson Space Center (JSC) is searching for methods to effectively manage the Center's resources to meet NASA's mission. D-Side is a group multi-criteria decision support system (GMDSS) developed to support facility decisions at JSC. D-Side uses a series of sequential and structured processes to plot facilities in a three-dimensional (3-D) graph on the basis of each facility's alignment with NASA's mission and goals, the extent to which other facilities are dependent on the facility, and the dollar value of capital investments that have been postponed at the facility relative to the facility's replacement value. A similarity factor rank orders facilities based on their Euclidean distance from Ideal and Nadir points. These similarity factors are then used to allocate capital improvement resources across facilities. We also present a parallel model that can be used to support decisions concerning allocation of human resources investments across workforce units. Finally, we present results from a pilot study where 12 experienced facility managers from NASA used D-Side and the organization's current approach to rank order and allocate funds for capital improvement across 20 facilities. Users evaluated D-Side favorably in terms of ease of use, the quality of the decision-making process, decision quality, and overall value-added. Their evaluations of D-Side were significantly more favorable than their evaluations of the current approach.     

Automatic processing of psychological distance: Evidence from a Stroop task.

A picture-word version of the Stroop task was used to test the automatic activation of psychological distance by words carrying various senses of psychological distance: temporal (tomorrow, in a year), social (friend, enemy), and hypotheticality (sure, maybe). The pictures implied depth, with the words appearing relatively close to or distant from the observer. The participants classified the spatial distance of words faster when the word's implicit psychological distance matched its spatial distance (e.g., a geographically close word was classified faster when it was "friend" than when it was "enemy"). The findings are consistent with the idea that psychological distance is accessed automatically, even when it is not directly related to people's current goals, and suggest that psychological distance is an important dimension of meaning, common to spatial distance, temporal distance, social distance, and hypotheticality. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Optical pattern recognition using two-channel joint transform correlation

A new joint transform correlation (JTC) technique, named two-channel JTC (TJTC), is proposed in this paper for optical pattern recognition applications. The TJTC technique independently evaluates the autocorrelation and crosscorrelation values of the reference and the target images and employs a modified decision algorithm. In addition, optical threshold operation and fringe-adjusted filter are incorporated in the proposed technique to enhance the correlation output and to improve the discrimination performance. The proposed technique shows better recognition performance compared to existing JTC techniques. Computer simulation are presented to investigate the salient features of the proposed TJTC technique with noise-free as well as noisy input scenes. The text was submitted by the authors in English.     

布尔函数的对偶性和线性点

本文主要讨论了部分自对偶和部分反自对偶函数、部分线性函数以及具有线性点的函数的性质。     

On principal component analysis, cosine and Euclidean measures in information retrieval

Clustering groups document objects represented as vectors. An extensive vector space may cause obstacles to applying these methods. Therefore, the vector space was reduced with principal component analysis (PCA). The conventional cosine measure is not the only choice with PCA, which involves the mean-correction of data. Since mean-correction changes the location of the origin, the angles between the document vectors also change. To avoid this, we used a connection between the cosine measure and the Euclidean distance in association with PCA, and grounded searching on the latter. We applied the single and complete linkage and Ward clustering to Finnish documents utilizing their relevance assessment as a new feature. After the normalization of the data PCA was run and relevant documents were clustered.     

High‐error approximate dictionary search using estimate hash comparisons

A method for finding all matches in a pre‐processed dictionary for a query string q and with at most k differences is presented. A very fast constant‐time estimate using hashes is presented. A tree structure is used to minimize the number of estimates made. Practical tests are performed, showing that the estimate can filter out 99% of the full comparisons for 40% error rates and dictionaries of up to four million words. The tree is found to be efficient up to a 50% error rate. Copyright © 2006 John Wiley & Sons, Ltd.     

A novel document similarity measure based on earth mover’s distance

In this paper we propose a novel measure based on the earth mover’s distance (EMD) to evaluate document similarity by allowing many-to-many matching between subtopics. First, each document is decomposed into a set of subtopics, and then the EMD is employed to evaluate the similarity between two sets of subtopics for two documents by solving the transportation problem. The proposed measure is an improvement of the previous OM-based measure, which allows only one-to-one matching between subtopics. Experiments have been performed on the TDT3 dataset to evaluate existing similarity measures and the results show that the EMD-based measure outperforms the optimal matching (OM) based measure and all other measures. In addition to the TextTiling algorithm, the sentence clustering algorithm is adopted for document decomposition, and the experimental results show that the proposed EMD-based measure does not rely on the document decomposition algorithm and thus it is more robust than the OM-based measure.     

单片机在超声测距系统中的一种应用

为提高人们在测量距离时的方便性和准确性,现介绍一种基于单片机控制的超声测距系统的构成、工作原理和硬软件设计,利用该系统及其设计方法可以作为机器人的辅助视觉系统,也可用于其他的检测系统中.     

采矿爆破地震安全距离与设计参数的应用研究

依据爆破安全规程给出了实际检测资料的处理方法。根据爆破设计和检测数据拟合结果的差异性，讨论了爆破地震安全距离S随设计经验参数K、α值的变化规律和范围，得出K取最大值，α取最小值能够最大限度保证现场建筑物安全的结论。