覆盖
软件
激光器
超声波传感器
计算机科学
领域(数学分析)
材料科学
合成孔径雷达
计算机硬件
光学
嵌入式系统
声学
人工智能
物理
操作系统
数学分析
数学
作者
Junfu Tan,Chenyin Ni,Chanjuan Wu,Kai-Ning Ying,Lu-Nan Dai,Zhonghua Shen
标识
DOI:10.3389/fmats.2023.1283151
摘要
The application of laser ultrasonic technology to defect detection is being widely studied. The Synthetic Aperture Focusing Techniques (SAFTs) is one of the imaging algorithms for detecting and imaging internal defects. SAFT based on Delay and Sum (DAS-SAFT) is known for high-quality imaging results. However, it has the disadvantages of the large calculation amount, which severely restricts its application. This work proposes a Co-Overlay Software and Hardware method (SAFT-COSH) based on Zynq, which uses parallelism and pipeline technology on FPGA and can efficiently accelerate DAS-SAFT operations. It can be found that the result obtained by the SAFT-COSH method is similar to the original method. Moreover, the time consumption of the SAFT-COSH system is 1/28 of that of the I9-13900KF CPU used for DAS-SAFT calculation. Finally, it was found by analyzing the time consumption of the DAS-SAFT procedure that the time consumption of SAFT-COSH system is short enough to support real-time 3D defect imaging. Characteristics of high efficiency and low cost in the SAFT-COSH method make it applicable in fields that require fast imaging using SAFT-based methods.
科研通智能强力驱动
Strongly Powered by AbleSci AI