基于分类器相关性的Adaboost人脸检测算法

为了提高传统Adaboost算法的集成性能,提出一种基于分类器相关性的Adaboost算法。该方法在弱分类器的训练过程中加入分类器的相关性判定,使每一个弱分类器的生成不仅与当前分类器有关,而且与前面若干个分类器相关,并将由此生成的弱分类器组合成新的强分类器。在CMU正面人脸检测集上的仿真结果表明,较传统的Adaboost算法,基于分类器相关性的Adaboost人脸检测算法具有更好的检测效率,同时降低了误检率。

基于相关性的AdaBoost人脸检测算法

为提高传统AdaBoost算法的集成性能,降低算法复杂度,提出2种基于分类器相关性的AdaBoost算法。在弱分类器的训练过程中,加入Q统计量进行判定。每个弱分类器的权重更新不仅与当前分类器有关,而且需要考虑到前面的若干分类器,以有效降低弱分类器间的相似性,剔除相似特征。仿真结果表明,该算法具有更好的检测率,同时可降低误检率,改进分类器的整体性能。

基于Multi-Agent的分类器融合

针对决策层输出的分类器融合问题 ,该文提出了一种基于Multi Agent思想的融合算法 .该算法将分类器融合问题建模为人类发源地问题 ,通过引入决策共现矩阵 ,并在智能体之间进行信息交互 ,从而利用了分类器之间的决策相关信息 .算法根据在融合训练集上得到的统计参量 ,指导各个智能体向不同类别溯源 ,并通过智能体之间的信息交换改变溯源概率 ,最终达到群体决策 ,得到决策类别 .本文在标准数据集上对该算法进行了实验研究 ,通过与其它一些融合方法的比较 ,得出在用于融合的分类器较少时 ,该算法得到比其它方法更低的分类错误率 ,其空间复杂度相对BKS方法较小 .实验证实 ,该算法是收敛的 .

面向白内障识别的临床特征校准注意力网络

近年来,卷积神经网络(convolutional neural network,CNN)模型已经被广泛用于年龄相关性白内障自动分类任务,然而,鲜有研究工作将年龄相关性白内障的临床先验知识注入卷积神经网络架构设计中,以此来提高年龄相关性白内障的分类效果和改善模型决策过程的可解释性。提出了一种临床特征校准注意力网络(clinical feature recalibration attention network,CFANet)模型用于自动识别年龄相关性白内障严重级别。在CFANet中,设计了一个简单且有效的临床特征校准注意力模块(clinical feature recalibration attention block,CFA),其不仅能对不同临床特征类型进行自适应地加权融合,还通过门控操作符来突出重要通道和抑制不重要通道。在一个核性白内障的眼前节光学相干断层成像影像(anterior segment optical coherence tomography,AS-OCT)数据集和一个公开眼科影像数据集上进行了充分实验,实验结果表明,相较于squeeze-and-excitation network(SENet)、efficient channel network(ECANet)、style-based recalibration module(SRM),CFANet在AS-OCT数据集上的分类准确率至少提升了3.54个百分点,同时在公开的眼科影像数据集上的分类结果比先进的神经网络模型和已发表的研究工作提升了1个百分点以上。此外,还通过可视化方法分析临床特征的权重分布和通道的注意力权重分布来提高该文模型决策过程的可解释性。

基于位置社交网络的上下文感知的兴趣点推荐

随着基于位置社交网络(Location-Based Social Networks,LBSN)的快速发展,兴趣点(Point-of-Interest,POI)推荐为基于位置的服务提供了前所未有的机会.兴趣点推荐是一种基于上下文信息的位置感知的个性化推荐.然而用户-兴趣点矩阵的极端稀疏给兴趣点推荐的研究带来严峻挑战.为处理数据稀疏问题,文中利用兴趣点的地理、文本、社会、分类与流行度信息,并将这些因素进行有效地融合,提出一种上下文感知的概率矩阵分解兴趣点推荐算法,称为TGSC-PMF.首先利用潜在狄利克雷分配(Latent Dirichlet Allocation,LDA)模型挖掘兴趣点相关的文本信息学习用户的兴趣话题生成兴趣相关分数;其次提出一种自适应带宽核评估方法构建地理相关性生成地理相关分数;然后通过用户社会关系的幂律分布构建社会相关性生成社会相关分数;另外结合用户的分类偏好与兴趣点的流行度构建分类相关性生成分类相关分数,最后利用概率矩阵分解模型(Probabilistic Matrix Factorization,PMF),将兴趣、地理、社会、分类的相关分数进行有效地融合,从而生成推荐列表推荐给用户感兴趣的兴趣点.该文在一个真实LBSN签到数据集上进行实验,结果表明该算法相比其他先进的兴趣点推荐算法具有更好的推荐效果.

Analysis on the Main Botanical Characters of 41 Cucurbita moschata Varieties

In order to scientifically evaluate the values of Cucurbita moschata cultivars, main botanical characters including the initial flowering date, the first fruiting node, fruit length, fruit stem length, stem diameter, internode length, the transverse and longitudinal diameters of the largest leaf, single fruit weight, flesh thickness and soluble solid content of 41 cultivars were measured for conducting diversity, correlation and cluster analysis. The results revealed that the pumpkin cultivars showed large variations in fruit stem length, single fruit weight, fruit length and flesh thickness, but small variations in initial flowering date. Significant, even highly significant correlations were found among the tested traits. Cluster analysis demonstrated that the 41 old Cucurbita moschata cultivars were divided into three groups, of which multiple traits of Group 1 were better than those in the other two groups. High similarities existed in three groups and the cultivars in each group. This research provided basis for selecting excellent traits and parents for the breeding of hybrids.

基于概率密度增强的高效多角度物体检测

许多物体检测工作都涉及目标物体的面内旋转问题。为了提高多角度物体检测的精度,在Gualdi等提出的以物体位置和尺度的概率密度采样为核心的粒子窗口方法基础上,提出充分利用不同角度物体分类器的相关性,通过用一个角度的分类器来增强另一个角度分类器所对应的采样概率密度,使得概率密度更快的收敛到待检测物体的位置和尺度上。在人手检测上的实验表明,所提方法比传统基于遍历滑动窗口的方法和基于现有基于粒子窗口的方法有更高的检测效率和精度。

MEASURING STRUCTURE COMPLEXITY OF UML CLASS DIAGRAMS

To provide system designer a valid measure to evaluate the structure complexityof class diagrams objectively, this letter first proposes a method to transform a class diagramsinto a weighted class dependence graph, then presents a structure complexity measure for classdiagrams based on entropy distance.

Digital Soil Mapping Using Artificial Neural Networks and Terrain-Related Attributes

Detailed soil surveys involve costly and time-consuming work and require expert knowledge. Since soil surveys provide information to meet a wide range of needs, new methods are necessary to map soils quickly and accurately. In this study, multilayer perceptron artificial neural networks(ANNs) were developed to map soil units using digital elevation model(DEM) attributes. Several optimal ANNs were produced based on a number of input data and hidden units. The approach used test and validation areas to calculate the accuracy of interpolated and extrapolated data. The results showed that the system and level of soil classification employed had a direct effect on the accuracy of the results. At the lowest level, smaller errors were observed with the World Reference Base(WRB)classification criteria than the Soil Taxonomy(ST) system, but more soil classes could be predicted when using ST(7 soils in the case of ST vs. 5 with WRB). Training errors were below 11% for all the ANN models applied, while the test error(interpolation error) and validation error(extrapolation error) were as high as 50% and 70%, respectively. As expected, soil prediction using a higher level of classification presented a better overall level of accuracy. To obtain better predictions, in addition to DEM attributes, data related to landforms and/or lithology as soil-forming factors, should be used as ANN input data.