微表情峰值帧定位引导的分类算法

目的 微表情是人在外界信息和刺激下做出的无意识面部动作,是判断受试人情绪和行为的重要佐证,在社会安全、商业谈判和心理辅导等领域都有着广泛的应用。微表情不同于一般的表情,分类与定位较为困难。针对这种情况,提出了一种基于光流窗口的双分支微表情定位网络（dual-branch optical flow spotting network,DFSN）和一种利用峰值帧光流信息的微表情分类网络,以识别视频中的微表情。方法 在定位任务中,首先提取面部图像,选择光流窗口大小和位置,计算面部光流并进行预处理;接下来输入双分支网络中进行两次分类,分别针对有无微表情和在有微表情前提下微表情所处阶段分类,并结合两个损失函数抑制过拟合;最后绘制出微表情强度曲线,曲线峰值所处位置即为所求微表情峰值帧。在分类任务中,选取视频起始帧和定位网络取得的峰值帧作为光流窗口,并利用欧拉运动放大算法（Eulerian motion magnification,EMM）放大微表情,最后采用峰值帧光流信息分类微表情视频。结果 微表情定位网络分别在CASME II （Chinese Academy of Sciences Micro-expression Database II）数据集和CASME数据集上按照使用留一被试交叉验证法进行了实验,与目前最好的定位方法比较,此网络在CASME II上获得了最低的NMAE（normalized mean absolute error）值0.101 7,比Optical flow+UPC方法提高了9%。在CASME上获得的NMAE值为0.137 8,在此数据集上为次优定位方法。在定位网络得到的峰值基础上,分类网络在CASME II上取得了89.79%的准确率,在CASME上取得了66.06%的准确率。若采用数据集标注的峰值,分类网络在CASME II上取得了91.83%的准确率,在CASME上取得了76.96%的准确率。结论 提出的微表情定位网络可以有效定位视频中微表情峰值帧的位置,帮助后续网络进行分类,微表情分类网络可以有效区分不同种类的微表情视频。     

面向维吾尔语关键词检索的等宽切词算法

该文提出了面向维吾尔语关键词检索的两种切词算法,并给出MATLAB实现的算法代码及详细说明;在同等条件下对两种算法的切词效果和关键词识别效率进行对比分析;提出两种算法的优化方法和构想。     

生物芯片点样针性能指标实验研究

对生物芯片点样针的重要性能指标之一－－样品点一致性的检测方法进行了研究，提出了一种新的检测与评价指标－－样品点像素总数变异系数；并讨论了具体的实验检测方法与步骤。     

Thermoelastic instability in a thin layer sliding between two half-planes: transient behaviour

The susceptibility of brakes and clutches to the known phenomenon of frictionally-excited thermoelastic instability is estimated studying the interface temperature and pressure evolution with time. A simple model has been considered where a layer with half-thickness a slides with speed V between two rigid and non-conducting half-planes. The advantage of this fairly simple model is that it permits us to deduce analytically the critical conditions for the onset of instability, that is the relation between the critical speed V_cr and the growth rate b of the interface temperature and pressure. It has then been verified that as the thickness a reduces the system becomes more and more prone to instability, and that a symmetrical pressure/temperature distribution at the layer interfaces can be more unstable than an antisymmetrical one. Moreover, the analysis of the evolution of the system beyond the critical conditions has shown that even if low frequency perturbations are associated with small critical speed, they might be less critical than high frequency perturbations if the working sliding speed is much larger than the actual critical speed of the system.     

数字微阵列点样喷头试验研究

利用压电陶瓷驱动器产生的低频微振动，实现了连续微流体的均匀离散和喷射。在试验所用的低频范围内(1～10Hz)，点样系统能够对每个驱动脉冲作出响应，具有数字可控性。通过疏水化处理技术消除微喷嘴的“流挂”现象，实现了微喷射量可控。微振动可以有效地消除微小颗粒可能对微喷嘴造成的堵塞，提高了点样喷头工作的可靠性。驱动频率对液滴直径有显著的影响。可以通过改变频率高低和微喷嘴内径大小实现皮升(pL)级点样喷射。     

基于过渡帧概念训练的微表情检测深度网络

为了更准确地从视频中检测面部微表情,针对微表情数据库样本规模较小的特点,采用迁移学习方法将深度卷积神经网络应用于微表情检测问题. 选取预训练过的深度卷积神经网络模型,保留卷积层及预训练参数,添加全连接层和分类器,构造一个二分类的微表情检测深度网络（MesNet）. 为了去除微表情数据库中影响网络训练的噪声标签,提出过渡帧的概念和自适应识别过渡帧算法. MesNet在CASME II、SMIC-E-HS与 CAS(ME)²数据库上的曲线下面积（AUC）分别达到0.955 6、0.933 8与0.785 3,其中在CASME II短视频数据库和 CAS(ME)²长视频数据库上均取得最优结果,表明MesNet具有高精度和广适用范围的特点;过渡帧对比实验结果表明,构造训练集时从原始视频中去除过渡帧能够有效提高MesNet微表情检测性能.     

Rejection of non-meaningful activities for HMM-based activity recognition system

This paper presents a new test to distinguish between meaningful and non-meaningful HMM-modeled activity patterns in human activity recognition systems. Operating as a hypothesis test, alternative models are generated from available classes and the decision is based on a likelihood ratio test (LRT). The proposed test differs from traditional LRTs in two aspects. Firstly, the likelihood ratio, which is called pairwise likelihood ratio (PLR), is based on each pair of HMMs. Models for non-meaningful patterns are not required. Secondly, the distribution of the likelihood ratios, rather than a fixed threshold, is used as the measurement. Multiple measurements from multiple PLR tests are combined to improve the rejection accuracy. The advantage of the proposed test is that the establishment of such a test relies only on the meaningful samples.     

某款机盖外板拉深模不等料厚加工方案

通过对帝豪某款机盖外板拉深模具机加工数据非等料厚间隙数据修改。按照料厚变化后的数据加工拉深模具。不用钳工大量的时间进行间隙延配就能达到合理的研合率要求及冲压质量要求。从而缩短模具的开发周期和降低钳工作业时间。     

电泳漆置换过程中的缺陷处理

介绍了电泳漆置换前期试验,并针对置换过程中的缩孔和花斑问题提出处理方法。     

Stop word location and identification for adaptive text recognition

Abstract. We propose a new adaptive strategy for text recognition that attempts to derive knowledge about the dominant font on a given page. The strategy uses a linguistic observation that over half of all words in a typical English passage are contained in a small set of less than 150 stop words. A small dictionary of such words is compiled from the Brown corpus. An arbitrary text page first goes through layout analysis that produces word segmentation. A fast procedure is then applied to locate the most likely candidates for those words, using only widths of the word images. The identity of each word is determined using a word shape classifier. Using the word images together with their identities, character prototypes can be extracted using a previously proposed method. We describe experiments using simulated and real images. In an experiment using 400 real page images, we show that on average, eight distinct characters can be learned from each page, and the method is successful on 90% of all the pages. These can serve as useful seeds to bootstrap font learning. Received October 8, 1999 / Revised March 29, 2000