基于图像识别的虚拟翻书系统

设计了一种基于图像识别的虚拟翻书系统,采用普通摄像头作为传感器捕获人的手势动作,并利用图像识别技术和计算机视觉技术实现虚拟翻书效果.该系统具有虚拟触摸功能,可以点击播放页面中的视频、语音等.     

掌纹识别算法综述

掌纹识别作为一种新兴的生物识别技术, 近年来得到了广泛的关注与研究. 与其他生物特征相比, 掌纹有许多独特的优势,包括识别率高、采集设备价格低廉、用户可接受性好等. 这些优势使得掌纹识别成为一种有着广泛应用前景的生物识别方法. 本文首先介绍了掌纹的特点、掌纹的采集设备和预处理方法, 之后详细介绍了近几年来提出的各种掌纹识别方法. 根据特征提取以及匹配方法的不同, 本文将掌纹识别方法分为基于结构的、基于子空间的、基于编码的和基于统计的四类方法. 在回顾和比较了各种算法的特点之后, 对未来的掌纹识别方法的发展方向作了展望.     

语料资源缺乏的连续语音识别方法的研究

由于少数民族语言有其本身的特点, 不能简单地套用现有的连续语音识别的方法. 本文以蒙古语为例, 研讨了声学和语言模型的建立, 并在日本国际电气通信基础技术研究所的连续语音识别器上实现了蒙古语的语音识别系统. 本文侧重于语言模型的建立, 基于蒙古语黏着性语言特点, 提出用相似词聚类方法建立多类N-gram模型. 实验结果显示, 应用我们提出的语言模型, 识别精度比用传统的词的N-gram识别法提高了5.5%.     

基于人造物体直线段结构特征的不变性识别

传统的用假设验证法进行三维物体识别的方法需要通过一组非线性方程组求解从模型到场景的坐标系变换,具有非常高的复杂度.文中提出了一种基于能够表明物体几何构造的直线段特征的人造物体识别方法,将假设验证法中对于全局坐标系变换的求解分散在各个平面单应性变换的求解中,降低了求解的复杂度.该方法首先利用几何不变量预匹配特征点,进而假设并求出场景和模型平面之间的单应矩阵,随后通过模型与场景之间直线段特征匹配的结果进行验证.实验证明,该方法能够快速准确地识别含有较多共面直线段特征的人造物体.     

基于SoPC的实时说话人识别控制器

运用软硬件协同设计,在DE2-70开发板上实现了一个基于SoPC的实时说话人识别控制器,控制器有很好的实时性和良好的识别性能。控制器的语音特征参数采用线性预测倒谱系数(LPCC),匹配算法采用动态时间规整算法(DTW)。     

基于VQ-MAP与LS-SVM融合的说话人识别系统

传统的最小二乘支持向量机(LS-SVM)使用特征向量作为训练样本,在说话人识别系统中应用时区分性不够明显。对此,提出VQ-MAP与LS-SVM融合的方法,使用通用背景模型(UBM)经过VQ-MAP过程得到说话人自适应参数集,把此参数集作为最小二乘支持向量机的训练样本应用于说话人识别系统中。用Matlab进行仿真实验,结果表明,该识别系统SVM训练时间短,且具有较高的识别率。     

An EDBoost algorithm towards robust face recognition in JPEG compressed domain

In this paper, we describe a novel multiclass boosting algorithm, EDBoost, to achieve robust face recognition directly in JPEG compressed domain. In comparison with existing boosting algorithms, the proposed EDBoost exploits Euclidean distance (ED) to eliminate non-effective weak classifiers in each iteration of the boosted learning, and hence improves both feature selection and classifier learning by using fewer weak classifiers and producing lower error rates. When applied to face recognition, the EDBoost algorithm is capable of selecting the most discriminative DCT features directly in JPEG compressed domain to achieve high recognition performances. In addition, a new DC replacement scheme is also proposed to reduce the effect of illumination changes. In comparison with the existing techniques, the proposed scheme achieves robust face recognition without losing the important information carried by all DC coefficients. Extensive experiments support the conclusion that the proposed algorithm outperforms all representative existing techniques in terms of boosted learning, multiclass classification, lighting effect reduction and face recognition rates.     

Method of classifier selection using the genetic approach

Abstract: The paper presents a novel machine learning algorithm used for training a compound classifier system that consists of a set of area classifiers. Area classifiers recognize objects derived from the respective competence area. Splitting feature space into areas and selecting area classifiers are two key processes of the algorithm; both take place simultaneously in the course of an optimization process aimed at maximizing the system performance. An evolutionary algorithm is used to find the optimal solution. A number of experiments have been carried out to evaluate system performance. The results prove that the proposed method outperforms each elementary classifier as well as simple voting.     

Iris recognition: Analysis of the error rates regarding the accuracy of the segmentation stage

Iris recognition has been widely used in several scenarios with very satisfactory results. As it is one of the earliest stages, the image segmentation is in the basis of the process and plays a crucial role in the success of the recognition task. In this paper we analyze the relationship between the accuracy of the iris segmentation process and the error rates of three typical iris recognition methods. We selected 5000 images of the UBIRIS, CASIA and ICE databases that the used segmentation algorithm can accurately segment and artificially simulated four types of segmentation inaccuracies. The obtained results allowed us to conclude about a strong relationship between translational segmentation inaccuracies – that lead to errors in phase – and the recognition error rates.