Combined Fuzzy Local Binary Pattern and Wavelet Transform Features for Defect Detection of 11/33 kV Overhead Power Line Insulators
P. Surya Prasad1, B. Prabhakara Rao2
1P.Surya Prasad, Department of ECE, MVGR College of Engineering, Vizianagaram (Andhra Pradesh), India.
2B.Prabhakara Rao, Department of ECE, JNT University, Kakinada (Andhra Pradesh), India.
Manuscript received on 07 March 2019 | Revised Manuscript received on 20 March 2019 | Manuscript published on 30 March 2019 | PP: 1081-1085 | Volume-8 Issue-5, March 2019 | Retrieval Number: E3357038519/19©BEIESP
Open Access | Ethics and Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: The power line insulators are an important component of the power distribution system as consistent power delivery depends mainly on it. As the defected or damaged insulators on the electric poles leads to significant losses, there must be a regular monitoring system to check the insulator’s condition. This requires taking pictures of the poles, sending them for processing, and applying pattern recognition for classifying the status of insulator into healthy or defective, and replacing the damaged insulator. The breakage condition of the insulators can be determined by the structures derived from the insulator images. The insulator images can be obtained from the pole image captured using a video camera. The structures of corresponding insulator images are extracted from using Fuzzy Local Binary Pattern (FLBP), a variant module of the Local Binary Pattern as well as the wavelet transform. The obtained features are forwarded to SVM (Support Vector Machines) classifier which determines the status circumstance of the insulator and the efficacy of the proposed experimental results are validated. The hybrid model proposed in this paper, by combining both the feature vectors has resulted in better performance compared to when individual feature vectors are used for analysis. The automatic status determination of powerline insulators would reduce the human efforts to a larger extent and so the proposed insulator health condition monitoring system can be considered as a reliable method for insulator defect detection and the necessary follow-up mechanism.
Keyword: Feature Extraction, Classification, SVM, FLBP, Wavelet Transform, Pattern Recognition.
Scope of the Article: Classification