跳到主要导航 跳到搜索 跳到主要内容

Sound recovery via intensity variations of speckle pattern pixels selected with variance-based method

  • Ge Zhu
  • , Xu Ri Yao*
  • , Peng Qiu
  • , Waqas Mahmood
  • , Wen Kai Yu
  • , Zhi Bin Sun
  • , Guang Jie Zhai
  • , Qing Zhao
  • *此作品的通讯作者
  • Beijing Institute of Technology
  • CAS - National Space Science Center
  • Chinese Academy of Sciences

科研成果: 期刊稿件 › 文章 › 同行评审

摘要

In general, the sound waves can cause the vibration of the objects that are encountered in the traveling path. If we make a laser beam illuminate the rough surface of an object, it will be scattered into a speckle pattern that vibrates with these sound waves. Here, an efficient variance-based method is proposed to recover the sound information from speckle patterns captured by a high-speed camera. This method allows us to select the proper pixels that have large variances of the gray-value variations over time, from a small region of the speckle patterns. The gray-value variations of these pixels are summed together according to a simple model to recover the sound with a high signal-to-noise ratio. Meanwhile, our method will significantly simplify the computation compared with the traditional digital-image-correlation technique. The effectiveness of the proposed method has been verified by applying a variety of objects. The experimental results illustrate that the proposed method is robust to the quality of the speckle patterns and costs more than one-order less time to perform the same number of the speckle patterns. In our experiment, a sound signal of time duration 1.876 s is recovered from various objects with time consumption of 5.38 s only.

源语言英语
期刊论文编号026117
期刊Optical Engineering
卷57
期2
DOI
出版状态已出版 - 1 2月 2018

学术指纹

探究 'Sound recovery via intensity variations of speckle pattern pixels selected with variance-based method' 的科研主题。它们共同构成独一无二的学术指纹。

引用此