Unmanned aerial vehicle vision wire patrol method based on gradient constraint Radon transform

A UAV and gradient technology, applied in the field of image processing, can solve the problems of inaccurate edge information, inconspicuous edges of high-voltage transmission lines, difficulty in judgment and identification, etc., to achieve high resolution, improved accuracy, and accurate detection. Effect

CN106056619AInactive Publication Date: 2016-10-26CHANGAN UNIV
3 Cites 21 Cited by

Patent Information

Authority / Receiving Office
CN · China
Current Assignee / Owner
Publication Date
2016-10-26
Estimated Expiration
Not applicable · inactive patent

Smart Images

  • Figure 1
    Figure 1
  • Figure 2
    Figure 2
  • Figure 3
    Figure 3
Patent Text Reader

Abstract

The invention discloses an unmanned aerial vehicle vision wire patrol method based on gradient constraint Radon transform. The method comprises the following steps: 1) obtaining a high-voltage line image to be processed by utilizing an unmanned aerial vehicle image acquisition device and converting the obtained degraded image into a grayscale image F; 2) carrying out edge extraction on the grayscale image F to obtain an edge image F'; 3) carrying out gradient calculation on the edge image F' to obtain gradient magnitude Gk; 4) setting eight direction pixel points of the kth pixels in the edge image F' being ki, wherein i=0,1,...,7, and when the gradient magnitudes Gk of the adjacent two pixels ki are same, grouping the pixels to the same line support region; 5) for the line support region obtained in the step 4), carrying out fitting to obtain a straight line according with conditions, and carrying out follow-up processing on the fit straight line through mathematical morphology; and 6) detecting a power transmission line through Radon transform according to the fitting result obtained in the step 5). The method realizes detection accuracy of the power transmission line in the image, removes redundant edge information and improves accuracy of power transmission line identification.
Need to check novelty before this filing date? Find Prior Art

Description

technical field

[0001] The invention belongs to the technical field of image processing, and in particular relates to an unmanned aerial vehicle visual wire inspection method based on gradient constrained Radon transformation. Background technique

[0002] At present, the main methods of long-distance power transmission and distribution in my country are high-voltage and ultra-high-voltage overhead transmission lines. Under the action, it will produce local fatigue damage and cause partial strand breakage. If the transmission line with broken strand defect is not repaired and replaced in time, it will not only increase the power loss of the transmission line and cause power waste, but in severe cases, it will cause the wire to break, which directly threatens the safety of human life; and the transmission line usually passes through hills, mountains and bad Geological areas, etc., sometimes need to cross large rivers, lakes and straits. Therefore, my country's existing low-eff...

Examples

Embodiment Construction

[0026] Below in conjunction with accompanying drawing, the present invention is described in further detail:

[0027] 1. Obtain the high-voltage line image in step 1: use the UAV image acquisition device to obtain the image of the high-voltage line to be processed, and convert the obtained degraded image into a grayscale image.

[0028] 2. In step 2, the Canny operator is used to extract the edge of the image: among all kinds of edge detection operators, the Roberts operator has a better edge detection effect on the edge gray value and the low-noise image. The sub-extracted straight line is discontinuous; the advantages of the Sobel operator are simple method, fast operation speed and smooth and continuous image edge. The edge of the image is rough; the Prewitt edge operator will get better results in detecting the edge of the image. In terms of noise suppression characteristics, it is not as strong as the anti-noise ability of the Canny operator; the effect of the edge detect...