Image mosaic method of aerial high-voltage power tower based on edge feature and point feature

An edge feature and image stitching technology, which is used in graphics and image conversion, image data processing, instruments, etc., can solve the problem of not considering the linear characteristics and integrity characteristics of high-voltage power towers, increasing the amount of algorithm calculation, and mismatching points with the same name. The problem is to ensure the integrity of connectivity and topological relationships, reduce mismatches, and improve robustness.

CN104881841BActive Publication Date: 2019-01-04ELECTRIC POWER RESEARCH INSTITUTE, CHINA SOUTHERN POWER GRID CO LTD +1
4 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Publication Date
2019-01-04

Smart Images

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

Abstract

The invention provides an aerial high voltage electric tower image splicing method based on edge characteristics and point characteristics. The method comprises the following steps of A, inputting two to-be-spliced images and removing backgrounds of the two to-be-spliced images; B, extracting the edge characteristics and a multi-scale tower skeleton; C, performing rough matching of the images to obtain a rough corresponding relation, and obtaining a rough overlap area of the two images on the basis of a roughly-matched image; D, performing surf characteristic point extraction in the overlap area in the step C; E, performing matching of surf characteristic points under the constraint of the rough corresponding relation in step C, so as to achieve homonymy points matching; F using surf homonymy points obtained in step E and a least square algorithm to obtain image affine transformation parameters, and performing accurate registration of the images; G, using a linear fusion method to achieve seamless splicing of registered images, multiplying the images with corresponding weighting functions respectively to achieve smooth transition of the overlap area of the images and the image fusion; and H, generating a spliced image.
Need to check novelty before this filing date? Find Prior Art

Description

technical field

[0001] The invention relates to an image splicing method of aerial high-voltage power towers based on edge features and point features. Background technique

[0002] Due to long-term exposure to the external environment, electrical equipment not only bears normal mechanical loads and electrical loads, but also is often violated by external forces such as pollution, lightning strikes, strong winds, rain, snow, and bird damage. Multiple factors cause various components on the line to age, Fatigue, oxidation and corrosion, if not detected and eliminated in time, may develop into various faults, posing a threat to the safety and stability of the power system, and the main components of the transmission line are located on the power tower, therefore, the inspection and maintenance of the tower It is a main work content of power equipment inspection.

[0003] With the rapid development of aerial remote sensing technology, electric power related departments have wi...

Examples

Embodiment 1

[0044] This embodiment provides an image stitching method for aerial high-voltage power towers based on edge features and point features, such as figure 1 As shown, it includes the following steps:

[0045] A. Input two images to be stitched, and remove the background of the two images to be stitched: Based on the color characteristics of the high-voltage power tower, by setting the gray threshold of each color component, the ground objects with obvious color differences from the tower in the image are processed Removed to reduce the interference of background objects on the subsequent extraction of towers.

[0046]First, the visible light image is converted from the RGB color space to the HSV color space, and the H color component of the HSV color space is used to set the threshold for background removal. The specific threshold is set as follows:

[0047]

[0048] Among them, f RGB (x, y) is the pixel value of the image at (x, y) position in RGB space, g H (x, y) is the...