基于特征索引的图相似查询过滤算法

分析图相似查询候选集的产生过程以及特征图之间的关系对候选图集的影响,提出一种基于特征索引的图相似查询过滤算法,使用GIndex算法建立特征图索引结构,通过特征图之间的选择性关系给出一个有序的特征集,并借助特征-图矩阵对数据库进行筛选得到候选图集。实验结果证明,该方法能准确地产生候选图集,从而提高图查询的效率。     

基于自适应神经模糊推理系统的视频烟雾检测

提出一种基于自适应神经模糊推理系统的视频烟雾检测算法。从视频图像中提取烟雾特征,采用减法聚类确定模糊规则数,建立初始模糊系统。通过神经网络的自学习机制调整前提参数和结论参数,确定模糊推理规则。实验结果表明,与传统BP神经网络算法及支持向量机算法相比,该算法具有较优的ROC曲线特性。     

一种改进的Harris特征点检测算法

现有Harris特征点检测算法采用高斯滤波进行平滑,图像存在角点信息丢失与偏移的现象。为解决该问题,提出基于变分B样条滤波与快速局部窗口搜索相结合的Harris特征点检测算法,选择具有低通特性的B样条函数作为平滑函数构造滤波器,引入形态学滤波中的极大值滤波思想,利用快速局部窗口搜索算法进行特征点局部极值的提取,从而提高特征点提取的精度和速度。实验结果表明,改进算法具有特征点提取快速均匀、检测定位准确、抑噪性好的特点。     

基于特征相似度的贝叶斯网络入侵检测方法

传统贝叶斯入侵检测方法未考虑属性和属性权值对检测结果的影响。为此,提出基于特征相似度的贝叶斯网络入侵检测方法。利用相似度对网络连接数据的属性特征进行选择,抽取其关键特征,并降低属性的冗余度,以优化朴素贝叶斯的分类性能。实验结果表明,该方法能降低分类数据的维数,提高分类的准确率。     

基于成对约束和稀疏保留的数据降维算法

结合以成对约束形式给出的监督信息和无监督信息,提出一种基于成对约束和稀疏保留的数据降维算法。通过成对约束信息进行鉴别分析,利用稀疏表示方法保留数据集在变换空间中的全局稀疏结构。实验结果表明,与传统特征抽取算法相比,该算法的识别效果更好,需要调节的参数更少,且鲁棒性较高。     

一种基于循环谱特征识别的频谱感知算法

针对传统循环谱计算复杂度高的问题,提出一种基于循环谱特征识别的频谱感知算法,该算法采用自相关矩阵替代频域平滑方法计算信号的谱特征,在降低算法复杂度的同时,能够准确区分噪声与有用信号,利用软件无线电构建测试平台,用以验证算法的有效性.实验结果表明,该算法能够准确检测出频谱空穴,准确度大于90％,并能降低时间复杂度,可以满...     

基于文本相似度的网页消重策略

针对在网页检索结果中经常出现内容相同或相似的问题,提出了一种通过计算网页相似度的方法进行网页消重。该算法通过提取网页特征串,特征串的提取在参考以往特征码提取的基础上,加入了文本结构特征的提取,通过比较特征串之间差异性的基础上得到网页的相似度。经与相似方法比较,结果表明,该方法减少了时间复杂度,具有较高的查全率和查准率,适于大规模网页消重。     

HDR Image Generation based on Intensity Clustering and Local Feature Analysis

This paper describes a cluster-based method for combining differently exposed images in order to increase their dynamic range. Initially an image is decomposed into a set of arbitrary shaped regions. For each region we compute a utility function which is based on the amount of presented information and an entropy. This function is used to select the most appropriate exposure for each region. After the exposures are selected, a bilateral filtering is applied in order to make the interregional transitions smooth. As a result we obtain weighting coefficients for each exposure and pixel. An output image is combined from clusters of input images using weights. Each pixel of the output image is calculated as a weighted sum of exposures. The proposed method allows recovering details from overexposed and underexposed parts of image without producing additional noise. Our experiments show effectiveness of the algorithm for the high dynamic range scenes. It requires no information about shutter speed or camera parameters. This method shows robust results even if the exposure difference between input images is 2-stops or higher.     

A knowledge base model for complex forging die machining

Recent evolutions on forging process induce more complex shape on forging die. These evolutions, combined with High Speed Machining (HSM) process of forging die lead to important increase in time for machining preparation. In this context, an original approach for generating machining process based on machining knowledge is proposed in this paper. The core of this approach is to decompose a CAD model of complex forging die in geometrical features. Technological data and topological relations are aggregated to a geometrical feature in order to create machining features. Technological data, such as material, surface roughness and form tolerance are defined during forging process and dies design. These data are used to choose cutting tools and machining strategies. Topological relations define relative positions between the surfaces of the die CAD model. After machining features identification cutting tools and machining strategies currently used in HSM of forging die, are associated to them in order to generate machining sequences. A machining process model is proposed to formalize the links between information imbedded in the machining features and the parameters of cutting tools and machining strategies. At last machining sequences are grouped and ordered to generate the complete die machining process. In this paper the identification of geometrical features is detailed. Geometrical features identification is based on machining knowledge formalization which is translated in the generation of maps from STL models. A map based on the contact area between cutting tools and die shape gives basic geometrical features which are connected or not according to the continuity maps. The proposed approach is illustrated by an application on an industrial study case which was accomplished as part of collaboration.     

Multilevel augmentation algorithms based on fast collocation methods for solving ill-posed integral equations

This paper continues the theme of the recent work Chen et al. (2008) [18], in which fast collocation methods are introduced for solving ill-posed Fredholm integral equations of the first kind. We develop in this paper multilevel augmentation algorithms, which lead to fast solutions of the discrete equations resulting from fast collocation methods. Regularization parameter choice strategies are given for proposed methods. The theoretical analysis and numerical experiments illustrate the accuracy and efficiency of the algorithm.