Transmission line extraction method based on Chebyshev polynomial descriptor

Abstract

The invention discloses a transmission line extraction algorithm based on a Chebyshev polynomial descriptor, which includes: preprocessing the input image, using random Hough transform to extract straight lines and curves in the image, and using the Chebyshev polynomial descriptor curve Describe the advantages, describe the features of the extracted objects, and use the K-means clustering analysis method to train the visual dictionary, and extract the transmission lines in the image through the comparison of the visual dictionary. The invention can effectively improve the accuracy of power transmission line extraction under complex background.

Description

technical field [0001] The invention relates to the fields of power system, pattern recognition and classification, especially the classification and detection of specific targets. Background technique [0002] In the power system, overhead line transmission is the main transmission method adopted since the development of the electric power industry. Relying on manual inspection and maintenance of transmission lines not only has low efficiency and high cost, but also has great safety problems. With the development of intelligent technology, the use of computer vision and image processing technology to detect transmission lines plays a very important role in ensuring the safe operation of power systems. Since the acquisition of image data is easily affected by various factors such as sensors, shooting environment, and shooting angle, the obtained images are not pure transmission lines, and most of them contain complex backgrounds. It is important to study the extraction algo...

AI Technical Summary

Problems solved by technology

Literature ("Li Chaoyang, Yan Guangjian, etc. Automatic extraction of high-voltage power lines in high-resolution aerial images [J], Chinese Journal of Image and Graphics, 2007, 12(6): 1041-1047.") proposed a method for complex natural background Under certain conditions, the Ratio operator and segmented Radon transform are used to automatically extract 550kV high-voltage power lines from aerial images. This method can automatically extract high-voltage power lines under complex natural backgrounds. Straight line, it is impossible to extract the curved power line with large curvature; literature ("Li Cailin, Feng Zhaohui, etc. Power line extraction method under the background of complex ground objects [J], Computer Engineering and Application, 2016, 52(22): 198-202.") On the basis of using the Canny operator to extract the image edge and the Hough transform to generate the broken straight line segment, an algorithm to restore the complete power line according to the geometric characteristics of the power line is proposed. power lines, but this method deals with high-altitude images, and can only extract straight line power lines, but cannot extract curved power lines; literature ("Zhang Shaoping, Yang Zhong et al. A method for extracting power lines from aerial images based on feature detection [J], Applied Science and Technology, 2012, 39(5):36-39.") On the basis of analyzing the characteristics of the power line in the aerial image of the UAV inspection system of the transmission line, using Ratio operator and Hough transform, a feature detection-based Power line extraction algorithm, but this method deals with images taken by drones, and can only extract straight line power lines, but cannot extract curved power lines; literature ("Qirong Ma, Darren S. Goshi, Yi-Chi Shih, Ming-Ting Sun , An Algorithm for Power Line Detection and Warning Based on a Millimeter-Wave Radar Video [J], IEEE Transactions on Image Processing, 2011, 20(12): 3534-3543.”) proposed a method based on Hough transform and support vector machine ( SVM) transmission line detection algorithm, which is used to detect video images acquired by millimeter-wave radar, but this method cannot deal with the problem of power line extraction in visible light images with complex backgrounds; literature ("Wei Song, Xiaorui Hu, Jin Fu, Qing Zhou, Ting Zhou, Pengju Si, The Method of Hybrid-laser Image Spot Extracts Based on HSV Space SVD for Power Transmission Line Detecti on[C],2016IEEEInternational Conference on Information and Automation(ICIA),2016,1361-1364") Using multiple lasers for positioning, a transmission line detection algorithm based on HSV space singular value decomposition is proposed for complex environments and transmission line detection under background conditions, but this method still cannot handle the problem of curved power line extraction

[0004] Most of the existing power line detection work focuses on straight line detection, which cannot be applied to the detection of power line images with large curvature; at the same time, the extraction of power lines in complex backgrounds is still incomplete, and the algorithm is complex and unstable. Waiting for deficiencies

Images