【作者】 赵苓； 王涛； 彭光正； 赵彤；

【Author】 Zhao Ling Wang Tao Peng Guangzheng Zhao Tong (School of Automation,Complicated System of Intelligent Control and Decision,Key Lab of Education Ministry,Beijing Institute of Technology,Beijing 100081,China)

【机构】 北京理工大学自动化学院 复杂系统智能控制与决策 教育部重点实验室；

【摘要】 本文介绍了一种利用红外热像图对密闭容器进行泄漏点定位的装置,红外热像仪采集多帧被测容器冲入冷却气体后的红外图像,然后对每帧图像的ROI区域进行高斯滤波,再利用canny算子提取出图像边缘,利用hough圆检测方法检测出疑似泄漏点,最后通过疑似泄漏点出现概率大小判断是否有泄漏,如果泄漏,实现泄漏点的定位。实验证明该方法能够有效克服红外图像噪声较大,泄漏点信息容易被淹没的缺点,并能够准确的检测出泄漏点的位置。更多还原

【Abstract】 In this paper,a device for closed vessel’s leaking localization is introduced based on infrared image processing.After the infrared thermography captured multi-frame infrared images of the closed vessel with cooling gas,each frame of the ROI region is processed by Gauss filtering.The suspected leak points are detected by hough circle detection method after their edges in the ROI are extracted by using canny edge operator.Finally, the appearance probability of the suspected leak point is used to determine a leaking point.Experiments show that the proposed method overcomes the disadvantage of infrared image’s higher level noise effectively and is able to accurately localize the leak points.更多还原