| 题名 | 快速声成像算法研究 |
| 作者 | 谌颖 |
| 学位类别 | 博士 |
| 答辩日期 | 2007-05-31 |
| 授予单位 | 中国科学院声学研究所 |
| 授予地点 | 声学研究所 |
| 关键词 | 成像声纳 波束形成 频域实现 分级聚焦 随机共振 |
| 其他题名 | Research on Fast Acoustic Imaging Algorithm |
| 学位专业 | 信号与信息处理 |
| 中文摘要 | 在成像声纳系统中，波束形成器是整个系统的关键组成部分，其精度和设计的优劣将直接影响到最终生成的图像质量。在体积有限的便携式声纳系统中，受设备尺寸和功耗的严格限制，常规的波束形成算法已经难以满足要求。本文主要研究可减少运算量的多种快速波束形成算法及其应用。论文的主要工作如下： 开展了频域波束形成的研究。针对高频窄带信号，提出了基于正交解调的频域波束形成方法，有效地解决了高运算量问题。利用频域波束形成解决了线阵阵元缺失时的波束形成问题。分析了不等间距傅立叶变换（NUFFT），并用其实现不等间距阵的频域波束形成。 提出了分级聚焦波束形成方法。将基阵划分成各级子阵分级进行处理，对较短的子阵进行远场处理，然后再对各个子阵的波束输出进行近场聚焦补偿，逐级合并子阵，并逐级进行波束插值。跟常规波束形成相比，该方法大大减小了波束形成所需的计算量和存储量，尤其适合多焦点聚焦场合。 研究了基于随机共振(SR)的快速波束形成。分析了阵列随机共振系统(ASR)，并将阵列随机共振的思想应用于分级聚焦波束形成。通过采用二值量化器阵列处理实现了寻址代替计算的快速波束形成，进而快速成像。该方法还可以应用于非高斯噪声中的声成像问题，在背景噪声概率密度函数非确知的情况下可提高信噪比。 |
| 英文摘要 | A beamformer is the key part of the system in a imaging sonar, whose precision and quality of design will directly affect the quality of the resulting images. In a portable sonar system of limited dimension, conventional beamforming can’t meet the demand with the strict restriction of equipment size and power. This paper mainly deal with kinds of fast beamforming algorithms and their applications, which can reduce the operation requirements. The main work of this paper is as follows. Frequency-domain beamforming is studied. As for high frequency narrow band signal, a frequency-domain beamforming algorithm based on quadrature demodulation is put forward, which settled the high computation problem effectively. Frequency-domain beamforming can also be used in the case of element inactivation. Nonuniform Fast Fourier Transform (NUFFT) is also discussed. And it can be used in the beamforming of unequally spaced array. Multistage focused beamforming algorithm is promoted. The array is divided into subarrays. Far field processing is taken on shorter subarrays at ealier stages. Focused beamforming is implemented on the beam outputs of subarrays of previous stage. Subarrays are combined with beam interpolation stage by stage. Multistage focused beamforming shows great savings on computational burden and data storage compared with conventional beamforming. This method is especially fit for the situation of multifocus focusing. Fast beamforming based on Stochastic Resonance(SR) is researched. Array Stochastic Resonance(ASR) is studied. And it is applied in multistage focused beamforming. Fast beamfoming instead of computation of location is implemented by binary-state quantizer array processing, which reduces the computation burden greatly. It can also be used in the imaging problem in non-Gaussian noise, which can improve the SNR with the noise PDF unknown. |
| 语种 | 中文 |
| 公开日期 | 2011-05-07 |
| 页码 | 87 |
| 内容类型 | 学位论文 |
| 源URL | [http://159.226.59.140/handle/311008/240]  |
| 专题 | 声学研究所_声学所博硕士学位论文_1981-2009博硕士学位论文 |
| 推荐引用方式 GB/T 7714 | 谌颖. 快速声成像算法研究[D]. 声学研究所. 中国科学院声学研究所. 2007. |
| 个性服务 |
| 查看访问统计 |
| 相关权益政策 |
| 暂无数据 |
| 收藏/分享 |
除非特别说明，本系统中所有内容都受版权保护，并保留所有权利。
修改评论