Computer Engineering and Applications ›› 2008, Vol. 44 ›› Issue (2): 241-244.

• 工程与应用 • Previous Articles     Next Articles

THz image recognition based on wavelet transform

HAN Yu1,GE Qing-ping1,ZHANG Cun-lin1,ZHANG Liang-liang2   

  1. 1.Capital Normal University,Beijing 100037,China
    2.Beijing Institute of Technology,Beijing 100081,China
  • Received:1900-01-01 Revised:1900-01-01 Online:2008-01-11 Published:2008-01-11
  • Contact: HAN Yu

基于小波变换的太赫兹图像识别

韩 煜1,葛庆平1,张存林1,张亮亮2   

  1. 1.首都师范大学,北京 100037
    2.北京理工大学,北京 100081
  • 通讯作者: 韩 煜

Abstract: Terahertz(THz) has a functional imaging for a lot of material.It can detect the chemical nature of the material for identification.The imaging technology displays a tremendous prospect in the fight against terrorism,medical imaging,nondestructive testing,etc.In the paper,the author explores and puts forward a method of testing material which uses the technology of wavelet transform and histogram matching.The results show that this method can accurately class sunflower kernel,sunflower peel and other substances,realizing automatic identification of material by the computer.This will be conducive to further research on terahertz technology in the application of identification of material.

Key words: THz, wavelet transform, histogram, euclidean distance, image recognition

摘要: 太赫兹(THz)波对很多物质都有成像功能,可以侦测物质的化学性质,进行物质的识别。其成像技术在反恐、医学成像、无损探伤等方面显示出了巨大的应用前景。采用小波变换和图像的灰度直方图匹配技术对葵花籽的太赫兹信号数据进行分析实验,提出了一种识别被测物质的方法,在此方面进行了探索尝试,结果表明此方法可以对葵花籽仁、葵花籽皮和其他区域的太赫兹信号数据进行准确地分类,实现利用计算机自动识别物质的功能,这将有利于进一步进行太赫兹技术在物质鉴别方面的应用研究。

关键词: 太赫兹, 小波变换, 直方图, 欧式距离, 图像识别