波束赋形
计算机科学
架空(工程)
平面波
传输(电信)
计算复杂性理论
帧(网络)
帧速率
计算机视觉
平面(几何)
干扰(通信)
图像质量
算法
窗口(计算)
领域(数学)
人工智能
还原(数学)
视野
图像(数学)
复配
选择(遗传算法)
质量(理念)
反射(计算机编程)
选择算法
信号处理
对比度(视觉)
声学
图像处理
光学
自适应波束形成器
作者
Soham Nivargi,Satish Mulleti
出处
期刊:
[Institute of Electrical and Electronics Engineers]
日期:2025-09-15
卷期号:: 1-4
标识
DOI:10.1109/ius62464.2025.11201334
摘要
Plane wave (PW) ultrasound imaging achieves ultrafast frame rates by insonifying the entire field of view with wide-angle plane waves and coherently compounding echoes across multiple transmission angles. However, acquiring and processing a large number of angles imposes significant computational and storage burdens, especially in real-time systems. To address this, we propose a two-fold framework that jointly optimizes beamforming and angle selection. First, we introduce a Weighted Window-based Beamforming (WBF) algorithm that adaptively emphasizes regions of interest using smooth window functions, thereby improving local contrast and mitigating artifacts from angle undersampling. Second, we propose a Queue-based Angle Selection Algorithm (QASA) that sequentially identifies the most informative transmission angles while maintaining balanced angular coverage. Experiments on PICMUS datasets demonstrate that WBF combined with QASA achieves image quality comparable to conventional delay-and-sum compounding with all 75 angles, while using only 15 angles—a reduction of 80%. This enables higher frame rates, reduced data acquisition, and lower computational overhead without compromising image fidelity.
科研通智能强力驱动
Strongly Powered by AbleSci AI