核反应堆燃料组件水下热湍流图像复原  

Image Restoration of Nuclear Reactor Fuel Components with Thermal Turbulence

作　　者：张子耀 赵晓艳[2] 吴雨杰 张震[1] Zhang Ziyao, Zhao Xiaoyan, Wu Yujie, Zhang Zhen

机构地区：[1]上海大学机电工程与自动化学院,上海200072 [2]江苏省水文水资源勘测局扬州分局

出　　处：《计量与测试技术》2018年第3期64-66,共3页Metrology & Measurement Technique

摘　　要：核电站在停堆复查期限,需采用视觉手段对燃料组件进行核心验证,但是燃料组件的余热会导致水下热湍流,严重影像图像质量。为获取热湍流下核燃料组件的清晰图像,提出了一种去除热湍流干扰的燃料组件图像复原算法,算法主要包括时域高通滤波和空间解卷积滤波。首先构建信噪比时域响应高通滤波器取得以稳定但模糊的视频,接着利用基于经验模型的点扩散函数,对水下核燃料组件模型图像进行空间解卷积滤波,并计算复原图像的质量参数值,确定合适的经验模型。实验结果显示,利用离焦模糊模型解卷积复原的燃料组件图像质量最好,表明该算法能够去除湍流对图像产生的干扰影响。After fuel reloading in the nuclear reactor, visual inspection is necessary for core identification of nuclear fuel assembly. The residual heat can induce thermal turbulence which strongly decreases the image quality. In order to obtain the clear image with turbulence, an image restoration algorithm removing turbulence that mainly consists of the time domain high pass filter and space deconvolution filter is proposed in this paper. First of all, a temporal infinite impulse response filter is used to get a stabilized but blurry video. After that, a second spatial deconvolution fil- ter based on point spread function of empirical models is used to estimate the images without turbulence. And then find the most appropriate empirical model according to the quality of restored images. The experiment result showed, the restoration for nuclear fuel components image with turbulence based on defocused model is better and the algorithm can remove the noise caused by turbulence in the image.

关 键 词：热湍流 图像复原 燃料组件 

分 类 号：TP391.4[自动化与计算机技术—计算机应用技术]