基于最大似然估计算法实现空气温度分布的可视化
针对空气中温度差值难以捕捉的问题,以空气中温度分布的可视化作为研究对象,采用基于最大事后概率的最大似然估计算法,研究空气中温度分布图像化问题.可视化测量系统中,在被测区域设置32个收发分离的超声波换能器,基于一发多收模式实现渡越超声信号数据采集,通过实验获取16×16=256个渡越时间参数TOF(Time of Flight).实验系统采用测量角度插补与渡越时间参数平行插补两种方法进一步补充成像所需渡越时间参数,确保重建图像可读性.对实验数据进行了基于最大似然估计算法的超声波CT图像重建,重建图像结果能成功分辨空气场温度值差异.实验结果表明,基于最大似然估计算法实现空气中温度差异可视化的有效性...
Saved in:
Published in | 实验室研究与探索 Vol. 33; no. 5; pp. 12 - 16 |
---|---|
Main Author | |
Format | Journal Article |
Language | Chinese |
Published |
江苏理工学院云计算与智能信息处理常州市重点实验室,江苏常州213001%江苏理工学院计算机工程学院,江苏常州,213001%江苏理工学院电气信息工程学院,江苏常州,213001%日本山形大学大学院理工学研究科,日本山形县米泽市992-8510
2014
江苏理工学院计算机工程学院,江苏常州213001 |
Subjects | |
Online Access | Get full text |
ISSN | 1006-7167 |
Cover
Summary: | 针对空气中温度差值难以捕捉的问题,以空气中温度分布的可视化作为研究对象,采用基于最大事后概率的最大似然估计算法,研究空气中温度分布图像化问题.可视化测量系统中,在被测区域设置32个收发分离的超声波换能器,基于一发多收模式实现渡越超声信号数据采集,通过实验获取16×16=256个渡越时间参数TOF(Time of Flight).实验系统采用测量角度插补与渡越时间参数平行插补两种方法进一步补充成像所需渡越时间参数,确保重建图像可读性.对实验数据进行了基于最大似然估计算法的超声波CT图像重建,重建图像结果能成功分辨空气场温度值差异.实验结果表明,基于最大似然估计算法实现空气中温度差异可视化的有效性. |
---|---|
Bibliography: | FAN Hong-hui,ZHU Hong-jin,LIU Xiao-jie,TAMURA Yasutaka( 1 a. College of Computer Engineering; 1 b. Key Laboratory of Cloud Computing & Intelligent Information Processing of Changzhou; 1 c. Department of Electrical Engineering, Jiangsu University of Technology, Changzhou 213001, China; 2. Graduate School of Science and Engineering, Yamagata University, Yonezawai Yamagata 992-8510, Japan) maximum likelihood;maximum a posteriori;air temperature visualization ;time of flight interpolation 31-1707/T This research aims to at evaluate evaluating the temperature distribution in the air using an ultrasonic computed tomography imaging technique.The Maximum likelihood algorithm was is applied to ultrasonic time of flight (TOF) computed tomography (CT) for temperature distribution in the air based on maximum a posteriori.32 ultrasonic transducers (receive and transmission function separation) are set in measure area,ultrasonic signal which transits measure area with temperature difference is collected based on TOF,so we get |
ISSN: | 1006-7167 |