基于微流控的油液动态磨粒目标检测方法

Oil Detection Method for Dynamic Wear Particle Target Based on Microfluidics

设计了一种基于微流控的油液磨粒在线监测系统,提高了磨粒测量精度和检测效率,研究了图像退化模型,运用基于频域的方法进行估算运动模糊尺度,使用维纳滤波方法对磨粒运动模糊图像进行恢复;提出了一种基于显著性的油液动态磨粒目标检测方法,利用视觉显著性提取图像感兴趣区域的方法对动态磨粒进行监测。基于频域残差显著性计算,提取感兴趣区域,在此基础上构建评价指标,完成了磨粒目标检测。研究结果表明,相较于传统目标检测算法,基于显著性的磨粒目标检测算法简单有效,在保证较高检测准确率的前提下,提高了磨粒检测效率,减少目标检测时间,对实现颗粒在线监测具有较大意义。

Relying on the rapid development of digital camera technology and image processing technology,this paper designed an on-line monitoring system for oil abrasive particles based on microfluidics.Accuracy and detection efficiency,the monitoring system has high reliability.The image degradation model was studied.The frequency domain based method was used to estimate the motion blur scale.The Wiener filtering method was used to recover the abrasive image of the abrasive grain.A method based on significance for dynamic oil particle target detection was proposed.Visually significant methods of extracting image-sensing regions were used to monitor dynamic abrasive particles.Based on the frequency domain residual significance calculation,the region of interest is extracted,and the evaluation index is constructed on this basis to complete the abrasive target detection.The research results show that compared with the traditional target detection algorithm,the saliency-based abrasive particle target detection algorithm used in this paper is simple and effective.Under the premise of ensuring high detection accuracy,the abrasive particle detection efficiency can be greatly improved and the target can be reduced.It is of great significance for realizing on-line monitoring of particles.