Research On Insulator Detection And Location Methods In Aerial Images

With the development of intelligent grid and power system automation,computer vision techniques are increasingly applied on the intelligent inspection and online monitoring for electric equipment on power system.The insulator is an important component in the transmission line.It is also a frequent failure component.Due to the long-term exposure to the field,the transmission lines are safe and reliable to be damaged by self-destruct,breakage,crack and foreign body hanging.Therefore,regular monitoring of insulator status,timely detection of insulator failure is essential.Through analyzing and processing the image data collected by the inspection of the transmission line,the method and technology of finding the insulator fault has become a hotspot in recent years.Based on the extraction and description of image features,this paper studies the detection and localization of insulator equipment in aerial image,and realizes the detection of the insulator drop fault,which is of great significance to realize the digitalization and intelligence of unmanned aerial vehicle inspection for overhead transmission line.Firstly,by analyzing the unique structural characteristics of the insulators,a method for recognition and locating insulators based on skeleton structure feature is presented.Based on the contour commonality of the target,the target inserts and the machine learning algorithm are combined to locate the insulators quickly.Then,the structure of the insulator is extracted and the skeleton structure of the insulator is defined.The structure is searched and the pseudo-target is eliminated to realize the accurate detection and location of the insulator.Secondly,to achieve more accurate and effective recognition and location,continue to extract a deeper level of insulator image features.a new method of insulator recognition based on the spatial configuration consistency is put forward.The candidate regions of the insulators are determined by fusion detection method of color contrast features and gradient contrast features.Then,the spatial form consistency features are defined and described by formula.The feature descriptors are used to search the insulators in the candidate regions and realize accurate identification of insulators.Then,based on the accurate recognition of the insulator,the unique features of the insulator dropout fault are defined,and a fault detection and location method of the insulator string based on the adaptive morphology is proposed.According to the color characteristics of the glass insulator,the color model is established,and then the image is segmented in the insulator localization area.Then the morphological processing is carried out according to the area of the insulator,and the fault areas is highlighted.Finally,the paper summarizes the work of this paper,and points out the researchwork that needs to be further carried out.