A Method of Identifying Rust Status of Dampers Based on Image Processing

• Published in: IEEE transactions on instrumentation and measurement Vol. 69; no. 8; pp. 5407 - 5417
• Main Authors: Huang, Xinbo, Zhang, Xiaoling, Zhang, Ye, Zhao, Long
• Format: Journal Article
• Language: English
• Published: New York IEEE 01.08.2020; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Anisotropic Gaussian kernel (ANGK) directional derivative; Complexity; Computer vision; Corrosion; Dampers; Image acquisition; Image enhancement; Image processing; Image segmentation; Image transmission; Inspection; Mapping; Monitoring; Power transmission lines; Shock absorbers; the rust shade index; the rusty area ratio (RAR); Transmission lines
• ISSN: 0018-9456; 1557-9662
• DOI: 10.1109/TIM.2019.2963732

As one of the important antivibration fittings in transmission lines, the damper's accurate location, the quick detection of damper rust defect, and the timely remedial measures are all of great significance to ensure the reliable transmission of electric energy. Considering the background complexity of the damper image and harmfulness of its rust defect, a method for identifying the rust status of damper based on image processing is proposed in this article. First, the damper image acquired by the image acquisition device is grayed, and the feature enhancement of the damper region is completed by the local difference processing, the edge strength mapping based on the anisotropic Gaussian kernel directional derivative and the image fusion, and so on. Second, the segmentation of damper is completed by the threshold segmentation and the morphological treatment. Finally, the rust parameters of the damper area, which are the rusty area ratio (RAR) and the color shade index (CSI) are presented, by which the identification of the rust status can be defined as the rust-free, slight rust, medium rust, and severe rust. The results show that the segmentation and accurate identification of the rusted dampers can be completed by the proposed method. The total identification rate of the rusted dampers can reach 93% even under the complex background. The method is new and effective for identifying the rust status of damper.