生物特征识别手背静脉特征图像分割特征提取LBP硕士论文
基于手背静脉的身份识别
信号与信息处理， 2011， 硕士
【摘要】 随着对社会安全的要求不断提高,以及鉴别技术的不断发展,生物特征识别在计算机视觉和模式识别领域得到越来越多的认识和应用。手背静脉识别作为一种新的具有不易伪造、不易复制、唯一性高、长期稳定等特点的生物特征识别技术,也得到了相当的发展。虽然已经有应用该技术的一些产品问世,但仍然有一些技术难点需要解决:设备
设计
领导形象设计圆作业设计ao工艺污水处理厂设计附属工程施工组织设计清扫机器人结构设计
的优化;针对性的图像分割
方法
快递客服问题件处理详细方法山木方法pdf计算方法pdf华与华方法下载八字理论方法下载
;特征提取和识别方法的改进等。本文在对系统各部分研究的基础上,设计并实现了基于手背血管特征的身份鉴别系统。首先,设计并实现了自主静脉特征采集设备。从红外静脉图像采集的原理入手,经过了原型、实验、设计、试制、改进、定型等步骤,在较为低廉的成本之下,尽量提高静脉成像质量。最后采用红外阵列反射式
方案
气瓶 现场处置方案 .pdf气瓶 现场处置方案 .doc见习基地管理方案.doc关于群访事件的化解方案建筑工地扬尘治理专项方案下载
研制了低成本的静脉图像采集设备。其次,设计了针对性的图像预处理算法。提出了自己的图像剪切校正及ROI提取算法。然后对一些预处理方法,像滤波及图像增强进行了研究。再次,将基于梯度增强的图像分割算法与常用的Niblack,阈值图像等算法进行了比较,为随后的结构特征提取提供了条件。实验证明,对于红外静脉图像分割,我们采用的算法优于常用的算法。最后,提出了新的特征提取和识别算法。... 更多还原
【Abstract】 As the increasing demand of social security and
improvement of identification technology, biometric
recognition is becoming more and more important and usable in computer vision as well as in pattern recognition. Hand-dorsa vein pattern with many excellent characteristics, such as uniqueness, stability and hard to copy or forgery, is developed considerably. Although this technique had been put forward to product, there is a lot of work to do:optimizing device design,
customizing image segmentati... 更多还原
【关键词】 生物特征识别; 手背静脉特征; 图像分割; 特征提取;
LBP;
【Key words】 biometric recognition; hand-dorsa vein pattern;
image segmentation; feature extraction; Local Binary Patterns
(LBP);
Abstract 4-5
摘要 6-7
Outline 7-9
1 Introduction 9-17
1.1 Background 9
1.2 Significance 9-12
1.3 Research status of biometrics 12-13
1.4 Research status of vein pattern 13-14
1.5 Works 14-15
1.6 Arrangement 15-17
2 Design of Acquisition Device 17-27
2.1 Overview 17
2.2 Design of device 17-24
2.2.1 Illumination 17-21
2.2.2 Imaging 21-24
2.3 Implement 24-25
2.4 Database 25-27
3 Pre-processing of Vein Images 27-36
3.1 Region of Interest 27-30
3.1.1 Overview 27
3.1.2 Correction 27-29
3.1.3 ROI extraction 29-30
3.2 Normalization 30-31
3.2.1 Size normalization 30-31
3.2.2 Gray normalization 31
3.3 Filtering and enhancement 31-35
3.3.1 Denoise 31-34
3.3.2 Enhancement 34-35
3.4 Conclusion 35-36
4 Segmentation of Vein Images 36-45
4.1 Overview 36
4.2 Segmentation algorithms 36-42
4.2.1 Threshold methods 36-39
4.2.2 Boundary methods 39-40
4.2.3 Gradient based image segmentation 40-42
4.3 Post-processing 42-44
4.3.1 Morphological filtering 43
4.3.2 Thinning 43-44
4.4 Conclusion 44-45
5 Feature Extraction and Pattern Recognition 45-58
5.1 Overview 45-46
5.2 Feature extraction algorithms 46-51
5.2.1 Structure features 46-48
5.2.2 Texture features 48-50
5.2.3 Partition Local Binary Patterns 50-51
5.3 Classifiers 51-54
5.4 Experiments and results 54-56
5.4.1 Feature experiments 54-56
5.4.2 Classifier experiments 56
5.5 Conclusion 56-58
6 Implement of Identification System 58-61
6.1 Workflow 58
6.2 Hardware 58-59
6.3 Software 59-61
7 Conclusion 61-63 References 63-65
Published Papers 65-66 Acknowledgements 66
浙公网安备 33021202002483