Localisation of insulator strings' images based on colour filtering and texture matching

Abstract

With the development of image processing technology and the improvement of unmanned aerial vehicle (UAV) manufacture, it has been becoming a hot research topic to inspect the transmission line's failure. Here, a method of locating the area of insulator strings in the aerial image captured by UAV system is been introduced. At first, the aerial image is converted to HSV colour space model, and the components of H and S are filtered to segment the insulator strings and background areas. Considering some background areas have the similar colour with the insulator string, the result of segmentation will contain some background areas. Then, the image of result is divided into small pieces with non-overlapping windows. The information of texture on each window is used to remove the remaining background area. The outline of the insulator strings in the image can be obtained. It is easily to detect the edge of the insulator string and locate the area of the insulator string using the outline image. At last, some insulator string images have been used to illustrate the processing of the method. Compared with the method using RGB colour space model, the results of method here is effective and reliable.

Keywords

• feature extraction
• image segmentation
• insulators
• image texture
• image colour analysis
• autonomous aerial vehicles
• image matching
• image filtering
• texture matching
• insulator string images
• outline image
• insulator string areas
• background areas
• HSV colour space model
• aerial image
• unmanned aerial vehicle manufacture
• digit image processing technology
• colour filtering
• insulator strings