一种在线快速图像抖动消除方法及其在疲劳试验中的应用  

An online fast image dithering elimination methodand its application in fatigue test

作　　者：张文东[1,2] 陈莉 陈先民[1,2] 苏少普 ZHANG Wendong;CHEN Li;CHEN Xianmin;SU Shaopu(Aircraft Strength Research Institute of China,710065 Xi'an,China;National Key Laboratory of Strength and Structural Integrity,710065 Xi'an,China)

机构地区：[1]中国飞机强度研究所,西安710065 [2]强度与结构完整性全国重点实验室,西安710065

出　　处：《应用力学学报》2024年第6期1274-1281,共8页Chinese Journal of Applied Mechanics

基　　金：国家自然科学基金资助项目(No.51601175);航空基金资助项目(No.2020Z058023006)。

摘　　要：提出一种基于先验信息的直方图匹配算法,以实现疲劳试验中试件观测图像抖动的在线快速消除。首先,根据灰度阈值对基准帧图像和目标帧图像计数统计形成直方图;其次,采用傅里叶变换将直方图变换为频域信息,并根据先验信息确定偏移阈值,进行互相关分析从而获得互相关数组;最后,选取互相关数组中的最大值,并根据最大值位置确定目标帧图像的抖动量,进行图像抖动的消除。基于先验信息的互相关频域分析极大地降低了计算量,从而使得本研究方法可以实现高分辨率图像中抖动的在线快速消除。试验结果表明,对于分辨率为4096×3000的图像,在保持抖动消除效果不变的情况下,本研究算法与传统方法相比,计算效率提高了3倍,能满足疲劳试验中试件观测的试验应用需求。This paper proposes a histogram matching algorithm based on prior information to realize the online rapid dithering elimination of the observation image in the specimen in the fatigue test.The reference frame and target frames were The histogram was counted to form a histogram first.Then the Fourier transform was used to transform the histogram into the frequency domain,and the offset threshold was determined based on prior information.After that cross-correlation analysis was used to obtain a cross-correlation array.Finally the maximum value in the cross-correlation array was selected and the dithering quantity of the target frame was determined according to the maximum value position to eliminate image dithering.Cross-correlation frequency domain analysis based on prior information greatly reduced the computational quantity,so that the proposed method could achieve online rapid dithering elimination in high-resolution images.The test results showed that the computational efficiency of the proposed algorithm is increased by three times compared with the traditional method while keeping the dithering elimination effect unchanged for the image with a resolution of 4096×3000.The results could satisfy the application requirements of specimen observation in fatigue test.

关 键 词：疲劳试验 图像消抖 直方图匹配 互相关分析 傅里叶变换 

分 类 号：V216.3[航空宇航科学与技术—航空宇航推进理论与工程] TP391.4[自动化与计算机技术—计算机应用技术]