Locally linear discriminant embedding: An efficient method for face recognition

In this paper an efficient feature extraction method named as locally linear discriminant embedding (LLDE) is proposed for face recognition. It is well known that a point can be linearly reconstructed by its neighbors and the reconstruction weights are under the sum-to-one constraint in the classical locally linear embedding (LLE). So the constrained weights obey an important symmetry: for any particular data point, they are invariant to rotations, rescalings and translations. The latter two are introduced to the proposed method to strengthen the classification ability of the original LLE. The data with different class labels are translated by the corresponding vectors and those belonging to the same class are translated by the same vector. In order to cluster the data with the same label closer, they are also rescaled to some extent. So after translation and rescaling, the discriminability of the data will be improved significantly. The proposed method is compared with some related feature extraction methods such as maximum margin criterion (MMC), as well as other supervised manifold learning-based approaches, for example ensemble unified LLE and linear discriminant analysis (En-ULLELDA), locally linear discriminant analysis (LLDA). Experimental results on Yale and CMU PIE face databases convince us that the proposed method provides a better representation of the class information and obtains much higher recognition accuracies.     

Test frame updating in CPM testing of Prolog programs

Category Partition Method (CPM) is a general approach to specification-based program testing, where test frame reduction and refinement are two important issues. Test frame reduction is necessary since too many test frames may be produced, and test frame refinement is important since during CPM testing new information about test frame generation may be achieved and considered incrementally. Besides the information provided by testers or users, implementation related knowledge offers alternative information for reducing and refining CPM test frames. This paper explores the idea by proposing a call patterns semantics based test frame updating method for Prolog programs, in which a call patterns analysis is used to collect information about the way in which procedures are used in a program. The updated test frames will be represented as constraints. The effect of our test frame updating is two-fold. On one hand, it removes “uncared” data from the original set of test frames; on the other hand, it refines the test frames to which we should pay more attention. The first effect makes the input domain on which a procedure must be tested a subset of the procedure’s input domain, and the latter makes testers stand more chance to find out the faults that are more likely to show their presence in the use of the program under consideration. Our test frame updating method preserves the effectiveness of CPM testing with respect to the detection of faults we care. The test case generation from the updated set of test frames is also discussed. In order to show the applicability of our method an approximation call patterns semantics is proposed, and the test frame updating on the semantics is illustrated by an example.

遗传算法与文本特征选择方法相结合的研究

对文本特征选择的传统方法进行了深入分析并对各自的优缺点进行了详细的对比;其次,通过利用遗传算法自身的寻优特点,把它与传统特征选择方法相结合,并得出了算法。最后,在对各自算法进行了分析,表明了在性能上有了明显的提高。     

一种基于浓缩树结构的属性约简算法

属性约简是粗糙集理论的重要研究内容之一,其中基于区分矩阵的约简算法是一种高效的约简算法,但算法具有很高的空间复杂度.为了减少区分矩阵的空间开销,利用浓缩树结构,结合区分矩阵单个属性一定为核属性的特征,提出改进的生成浓缩树算法,压缩存储区分矩阵中的非空数据项,且不丢失原区分矩阵的所有信息;利用生成的浓缩树结构结合启发式策略,给出属性约简算法.实验结果表明,算法正确有效并且空间复杂度有明显降低.     

不同算法下生成的数字高程模型对比分析研究

目前数字高程模型的生成在大多数商用地理信息系统软件或遥感图像处理软件中都能实现,但提供的算法有所不同.结合目前常用专业软件提供的不同算法,对同一数据源的高程数据分别生成数字高程模型(DEM),进行对比分析后发现:使用现有专业软件生成的DEM均存在着差异,这些差异主要来源于不同软件采用的算法差别,并且DEM的差异具有空间分布的规律性和负高差多于正高差的特点.由此得出以下结论,TIN算法具有生成的DEM精度高等优点,LRS和Non-LRS算法具有生成的DEM结构简单等优点.若要进行三维显示、土方量计算等,则选择TIN算法生成;若要检查高程数据错误,则选择LRS或Non-LRS算法生成.     

基于队列的动态电源管理策略

动态电源管理技术是一项系统级电源优化策略,它通过有选择性的关闭(或者降低性能)空闲的可控设备来降低功耗.在回顾以往系统级动态电源管理策略的基础上,提出了一种基于队列的新型系统级动态电源管理策略,利用队列不但可以提高对于可控模块何时被唤醒时间的预测命中率,并且能够根据不断变化的任务负载情况进行自身的快速调节,以适应不同任务的要求,通过搭建的SkyEye环境进行仿真,结果验证了在性能和节能方面,该方法优于传统方法.     

一种基于粗糙集理论的快速并行属性约简算法

将并行计算的思想融入基于粗糙集理论的快速属性约简中,提出了一种基于粗糙集理论的快速并行属性约简算法.该算法在保证约简结果是Pawlak约简的情况下,将属性约简任务划分到多个处理器中同时处理,从而大大提高了属性约简的效率.仿真实验结果说明了该算法的高效性.     

一种基于相对粒度的决策表约简算法

在知识粒度的基础上,针对决策表提出了相对粒度和属性相对重要性的概念,证明了知识的相对粒度随着知识粒度的增大而单调增加的变化规律,在此基础上提出了一种基于相对粒度的启发式约简算法,以弥补基于正区域的约简方法处理不一致决策表时存在的不足.通过理论分析和实例验证表明,该算法是有效的,且其时间复杂度相对较低.     

时间自动机的LTL性质模型检测研究

为了增强模型检测工具的检测能力,拓宽模型检测技术的应用范围,对基于时间自动机的LTL性质模型检测进行了研究,对自动机的状态空间的存储方式和状态空间的展开过程进行了分析,讨论了LTL性质模型检测工具的检测流程和检测算法的实现策略对工具检测性能的影响,针对制约模型工具的检测能力和检测效率的因素,采取了一些相应的优化改进策略.采用了BDD(二叉决策图)共享存储技术和位编码压缩存储,较有效地减小了空间消耗,缓解了模型检测中状态爆炸引起的内存空间不足问题.与DTSpin等著名的模型检测工具进行了实验比较,取得了较好的实验结果.     

A closed-loop cross-correlation method for detecting model mismatch in MIMO model-based controllers

Correlation between a dithering signal and the prediction error has been used for detecting model mismatch in univariate model based control systems. This paper extends that approach to MIMO control systems. A closed-loop cross-correlation method is presented to detect which specific input-output pairings of a model-based controller are mismatched. This method may be used in screening the complete set of models and in selecting candidate models for re-identification. The method first finds the rows and columns of the transfer function matrix that contain mismatch, and then the candidates are found by the intersection of the said rows and columns. Placing the system under partial control, whereby one or more of the manipulated variables are held constant, can be used to further reduce the set of candidate models.