多频谱声学成像机理与实验研究

批准号:
11504282
项目类别:
青年科学基金项目
资助金额:
21.0 万元
负责人:
郭新华
依托单位:
学科分类:
A2301.线性与非线性声学
结题年份:
2018
批准年份:
2015
项目状态:
已结题
项目参与者:
余先涛、韩雪、李彰、贺舒杰、郭承远、陈雅辉
国基评审专家1V1指导 中标率高出同行96.8%
结合最新热点,提供专业选题建议
深度指导申报书撰写,确保创新可行
指导项目中标800+,快速提高中标率
微信扫码咨询
中文摘要
声学成像具有重要的应用价值,成为学者们不断创新研究的热点问题。然而,大多数学者在常规声学成像的研究中都是基于单一频率或者有限的频率数,不能充分地描述和解释物体特性,如物体的表面轮廓、内部结构和材料属性等。这些物体特性表现出频率依存性,宽频域声波下的频率响应可体现出单一频率或者有限频率下无法呈现的物体特性。本项目在声学领域模拟多频谱光学成像,以宽频域声波与物体特性相互作用为基础,对物体在宽频域声波照射下的成像(简称多频谱声学成像)进行基础科学研究。具体研究内容包括:建立宽频域声波与物体特性相互作用的数学模型;理论计算物体特性在宽频域声波激励下的特征频率,并进行特征频率实验测量;研究多频谱声学数据信息提取与处理技术,实现物体频率特征的可视化。本研究旨在通过多频谱成像确立物体特性与多频谱声学数据的定量关系,以便更精确地描述和解释物体特性,为识别和分类物体提供一种新的理论与方法。
英文摘要
Acoustic imaging has significant application value, and becomes a hot research topic in the academic field. However, conventional acoustic imaging has been performed by using a single frequency or a limited number of frequencies, which is not sufficient to clearly describe or explain the characteristics of the object, such as surface profiles, structures hidden under surfaces or material properties of the object. These characteristics of the object exhibit frequency dependence. The frequency response under wide frequency range is unique, which cannot be shown under a single frequency or a limited number of frequencies. In this study, as an analogy to the multispectral imaging in optics, we do research based on the interaction between the characteristics of the object and sound waves with wide frequency range, and investigate the imaging under acoustic excitation conditions with wide frequency range, which is called multispectral acoustic imaging (MSAI). The concrete contents include establishing the mathematical model on the interaction between the characteristics of the object and sound waves with wide frequency range, making theoretical calculation of the characteristic frequency of the object under conditions of acoustic excitation with wide frequency range, doing experimental measurement of the characteristic frequency, and doing research on the extraction and processing of multispectral acoustic data. The purpose of this study is to conform the relationship of the object characteristics and the frequency by MSAI, so as to describe and explain the characteristics of the object and provide a novel theory and method for identifying and classifying objects.
声学成像具有重要的应用价值,成为学者们不断创新研究的热点问题。然而,大多数学者在常规声学成像的研究中都是基于单一频率或者有限的频率数,不能充分地描述和解释物体特性,如物体的表面轮廓、内部结构和材料属性等。这些物体特性表现出频率依存性,宽频域声波下的频率响应可体现出单一频率或者有限频率下无法呈现的物体特性。在基金资助下,针对多频谱声学成像开展了相关研究。通过对典型物体结构与宽频域声波的相互作用,建立物体的频率响应数学模型,并进行了仿真分析和理论计算;搭建了实验平台,在宽频域声波激励下,以单一传声器扫描方式获得了所需的声学数据;在获取实验数据后,采用无监督学习算法实现了物体频率特性显示和3D图像重构,物体图像的频率特性显示结果与理论计算结果基本一致;为了获得更好的宽频域声波,开展了新型传感器阵列设计工作。在应用前景方面进行了人体成像与识别的试探性研究。本项目对多频谱声学成像研究,可以实现更精确地描述和解释物体特性,为物体立体成像、识别和分类提供一种新的理论与方法。
期刊论文列表
专著列表
科研奖励列表
会议论文列表
专利列表
Frequency Representation: Visualization and Clustering of Acoustic Data Using Self-Organizing Maps
频率表示:使用自组织图对声学数据进行可视化和聚类
DOI:10.1177/0161734617706476
发表时间:2017-05
期刊:Ultrasonic Imaging
影响因子:2.3
作者:Guo Xinhua;Yu Xiantao;Sun Song;Wang Pan;Nakamura Kentaro;Guo XH
通讯作者:Guo XH
国内基金
海外基金
