結合紋理特徵和幾何特徵的指紋影像品質評估;Fingerprint Image Quality Evaluation Combining Texture and Geometric features

NCU Institutional Repository > 資訊電機學院 > 資訊工程學系碩士在職專班 > 博碩士論文 > Item 987654321/72018

jsp.display-item.identifier=請使用永久網址來引用或連結此文件: http://ir.lib.ncu.edu.tw/handle/987654321/72018

题名:	結合紋理特徵和幾何特徵的指紋影像品質評估;Fingerprint Image Quality Evaluation Combining Texture and Geometric features
作者:	安晨碩;AN,CHEN-SHUO
贡献者:	資訊工程學系在職專班
关键词:	指紋影像;紋理特徵;品質評估;機率神經網路
日期:	2016-06-23
上传时间:	2016-10-13 14:21:55 (UTC+8)
出版者:	國立中央大學
摘要:	指紋影像評估主要應用於指紋感測器的影像品質評估。一個設計良好的指紋影像評估系統同時有助於提升指紋註冊的正確性，並有效提升指紋辨識率。本研究提出了一個結合紋理特徵分析及機率神經網路(Probability Neural Network，PNN)的指紋影像評估系統，以便改良現有的指紋影像品質評估方法。本研究的指紋影像評估流程可分為紋理特徵分析，PNN神經網路分類，整合PNN分類結果與傳統NFIQ評估，得到最終品質分數。就實驗結果顯示，結合紋理特徵分析的指紋影像評估可以提升評估指紋品質時的精確度。 ;Fingerprint image assessment is primarily applied to assess the quality of images stored in fingerprint readers. A satisfactorily designed fingerprint image assessment system effectively improves the accuracy of fingerprint registration and the rate of fingerprint recognition. This study incorporated a new fingerprint image assessment system that combines fingerprint texture analysis with a probability neural network (PNN) to improve the existing fingerprint image quality assessment method. The fingerprint image assessment procedure involved texture analysis and PNN neural network classification. Subsequently, the PNN classification results were organized and compared to the conventional National Institute of Standards and Technology fingerprint image quality assessment results to obtain the final quality scores. According to the result of the experiment, the fingerprint image assessment system that incorporated texture analysis improved the accuracy of fingerprint image quality assessment.
显示于类别:	[資訊工程學系碩士在職專班 ] 博碩士論文

文件中的档案:

档案	描述	大小	格式	浏览次数
index.html		0Kb	HTML	506	检视/开启

在NCUIR中所有的数据项都受到原著作权保护.

社群 sharing

数据加载中.....