AFIS试验与日常质量控制

介绍了USTER AFIS PRO在使用过程中对提高产品质量，特别是对降低棉结，降低成本起到的重要作用。     

AFIS与aQura测试参数稳定性的对比研究

 对AFIS与 aQura 2种测试仪的测试稳定性进行了对比研究,将2次不同实验（每次实验包括12个方案）条件下所生产的生条分别用AFIS与aQura仪器进行检测,用统计学的方法对检测结果进行分析,得出了2种仪器共有测试参数的稳定测试次数的范围和均值。认为AFIS与aQura 2种仪器在检测带籽屑棉结含量和尺寸及棉结总数方面的稳定性差异相当明显,aQura的测试稳定性要远远高于AFIS;在检测棉结平均尺寸、5%长度、质量短绒率和根数短绒率指标方面,2种仪器的测试稳定性差异不大,测试稳定性均较高。     

基于生物特征识别的自动勤务系统设计

提出一种基于生物指纹特征提取的小型企事业单位勤务系统。介绍了该系统各模块引入的相关算法。将这些算法的原理应用到系统设计中,设计出系统各模块的电路结构,并在FPGA芯片上进行了仿真,验证了该系统的可行性。最后提出了该系统的改进方案。     

一种基于新特征的有效指纹图像分割算法

指纹图像分割在自动指纹识别系统中发挥了非常重要的作用,有效的分割不但可以减少后续处理的时间,而且可以大大增强特征提取的可靠性,提高系统识别的准确性.主要做了两个方面的工作:提出了一种称之为有效点聚集度的新的指纹图像分割特征;依据有效点聚集度及文献[1]中提出的块聚集度特征,提出了一种有效的指纹图像分割方法,该方法首先采用有效点聚集度对指纹图像做粗分割.然后对粗分割结果采用基于迭代的方法进行后处理,接着运用块聚集度在第一次后处理结果的基础上做细分割,最后采用形态学方法对细分割后的结果做第二次后处理.大量实验证明:相对于已有常用的指纹图像分割特征,有效点聚集度具有鉴别能力强、鲁棒性好、分割出的前景、背景区域较为集中的特点;基于有效点聚集度及块聚集度提出的指纹图像分割算法具有较高的准确性及较强的适应性.     

基于连续分布方向图和改进的Poincaré Index的指纹分类

本文提出了指纹连续分布方向图(场)的概念及其算法,连续分布方向图具有很好的连续性、渐变性、抗噪性和较高的精确度;对经典的Poincaré Index计算公式和指纹奇异点检测算法进行了改进,改进后的Poincaré Index不仅能精确表示向量场的旋转角度,而且还能精确表示向量场的旋转方向,能够在像素级水平精确定位指纹奇异点(core 点和delta点).在此基础上,提出了一种新的基于连续分布方向图和改进的Poincaré Index的5类自动指纹分类算法,在江苏科技大学指纹库(含4000幅指纹)上的分类结果表明,本算法能抗任意角度的指纹图像旋转,成功地解决了指纹分类中的图像平移、旋转和形变不变性问题,分类正确率达到97.05%,具有较好的健壮性,满足实用要求.     

Fingerprint verification based on minutiae features: a review

Fingerprints have been an invaluable tool for law enforcement and forensics for over a century, motivating research into automated fingerprint-based identification in the early 1960s. More recently, fingerprints have found an application in biometric systems. Biometrics is the automatic identification of an individual based on physiological or behavioural characteristics. Due to its security-related applications and the current world political climate, biometrics is presently the subject of intense research by private and academic institutions. Fingerprints are emerging as the most common and trusted biometric for personal identification. The main objective of this paper is to review the extensive research that has been done on automated fingerprint matching over the last four decades. In particular, the focus is on minutiae-based algorithms. Minutiae features contain most of a fingerprints individuality, and are consequently the most important fingerprint feature for verification systems. Minutiae extraction, matching algorithms, and verification performance are discussed in detail, with open problems and future directions identified.

利用神经网络与AFIS纤维测试系统预测纱线质量

评述了人工神经网络以及AFIS纤维测试系统的特点，提出利用人工神经网络和AFIS纤维测试系统进行纺纱质量预测的工作原理和网络构建方法，并提供了实例，说明神经网络可以有效地解决纱线质量预测问题，指出人工神经网络技术在纺纱质量预测和控制中的广泛应用前景。     

Application of an Automatic Yarn Dismantler to Track Changes in Cotton Fiber Properties During Processing on a Miniature Spinning Line

This paper reports on the application of a newly developed automatic yarn dismantler for dismantling short staple ring-spun yarns, to track changes in cotton fiber properties from lint to yarn, during processing on a miniature spinning line. The results obtained on different Upland cottons have clearly demonstrated the practical value of the yarn dismantler in enabling yarns to be automatically dismantled into their constituent fibers. Based upon Advanced Fiber Information System (AFIS) fiber tests, miniature opening and cleaning, on a Shirley Analyzer, increased fiber neps and short fiber content and decreased fiber length, whereas the miniature card did not have a consistent effect on mean fiber length, but decreased neps substantially. The first miniature drawframe passage resulted in a significant increase in mean fiber length, fiber linear density, and maturity ratio and a decrease in short fiber content, immature fiber content, and neps. The mean length of fibers from the dismantled yarn tended to be slightly longer than that of the fibers just prior to the miniature spinning frame, while the immature fiber content tended to be slightly lower and the maturity ratio slightly higher.     

基于矢量三角法的指纹特征匹配算法的研究

指纹匹配是自动指纹识别系统(Automatic Fingerprint Identification System,AHS)中最重要的问题之一。文中主要针对如何确定两幅指纹图(模板和输入图像)的参照点问题,提出了一种基于指纹脊线结构和矢量三角法相结合的算法,将脊线信息引入匹配过程中,并在极坐标下进行细节点匹配。仿真实验表明,该方法不依赖指纹图的中心区域,具有旋转平移不变性,不仅能很好地区分来自不同指纹的图像,而且有较好的匹配结果。