Unlock instant, AI-driven research and patent intelligence for your innovation.

A method for detecting the inclination of intelligent towers based on machine vision

A technology of tilt detection and machine vision, which is applied in the field of image recognition, can solve problems such as high material cost, installation cost and maintenance cost, failure to find a suitable installation position, unfavorable tilt detection, etc., to achieve high safety, improve safety, and improve The effect of measurement accuracy

Active Publication Date: 2022-02-08
涵涡智航科技(玉溪)有限公司
View PDF10 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] 1. It is necessary to equip each tower with a corresponding camera (for complete monitoring of the tower, each tower needs to install cameras in at least two directions perpendicular to each other), the transmission of monitoring data requires additional hardware support, and it is installed in the field The camera itself is vulnerable to damage, resulting in extremely high material costs, installation costs and maintenance costs
[0006] 2. Since the height of the tower is often very high and much greater than the width, in order to obtain a complete tower image from the horizontal direction, the camera must be installed relatively far away, which will result in the most critical image of the tower width direction in the image (used to calculate the tower inclination angle ) is narrow, in the case of limited camera resolution and long distance, it is not conducive to tilt detection, which eventually leads to low detection accuracy or even no detection
[0007] 3. In complex environments such as forests and cities, there are usually a large number of occluders at the lower part of the tower, which will bring great challenges to the selection of the camera installation location. For example, if you want to install the camera to a higher and suitable shooting position location, it will further increase the cost, and even cannot find a suitable installation location

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • A method for detecting the inclination of intelligent towers based on machine vision
  • A method for detecting the inclination of intelligent towers based on machine vision
  • A method for detecting the inclination of intelligent towers based on machine vision

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0053]Firstly, the applicable platform of the method will be explained. In order to complete the necessary detection, the system needs to include a camera 2. The actual mobile platform that can be carried can be a lifting mechanism or a fixed set of equipment for each tower, but considering the cost and the need for unmanned and automated operation In this embodiment, the above-mentioned shooting device is preferably mounted under the aircraft 1; specifically, a multi-axis drone is selected as a mobile platform, and each camera 2 is connected through a pan-tilt, and the plane of the camera 2 is kept parallel to the horizontal plane.

[0054] A machine vision-based intelligent tower tilt detection method, such as figure 1 As shown, the main steps are as follows:

[0055] First of all, there is not only one way to manually control the aircraft 1 to move to the sky above the tower; it includes using GNSS positioning technology to navigate to the sky above the tower, and using vis...

Embodiment 2

[0080] This embodiment adjusts for S4 on the basis of Embodiment 1. Preferably, in S4, it is found that there will still be a small amount of raised plane in the top plane image of the tower that has the greatest influence, and the minimum value is selected as ; so in order to improve the solution The accuracy of the inclination angle is extended in this embodiment to select N depth values ​​in the plane area in addition to selecting the depth maximum point A and the minimum value point B, and use N+2 values ​​to fit the plane. Fitting methods include but not limited to the least squares method; obtain the equation of the fitting plane, and obtain the angle (face-to-face angle) between the plane and the horizontal plane to be the inclination of the tower.

Embodiment 3

[0082] First, machine vision is used for target recognition. After the tower is recognized, the relative position of the tower and the carrier is calculated using visual SLAM technology and path planning is carried out. Finally, the aircraft 1 is guided to move autonomously with the camera 2 to the sky above the tower, and is controlled by the flight control system. After the aircraft 1 reaches the sky above the tower, flip down the camera 2 through the gimbal to take the image of the top of the tower, and then proceed to step S1.

[0083] S1, take a top image of the tower to be measured 3 by the camera 2 positioned above the tower to be measured 3;

[0084] S2, based on the plane recognition model of machine learning training, identify the plane area at the top of the tower 3 to be measured in the image taken by S1; from the general structure of the tower, there must be at least one plane formed by points (or lines) at the top of the tower, specifically as The partition / inter...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention discloses a method for detecting the inclination of an intelligent tower based on machine vision. The specific steps include: S1, taking a top image of the tower to be tested with a camera located above the tower to be tested; S2, training a plane recognition model based on machine learning , identifying the plane area at the top of the tower to be measured in the image taken by S1; S3, obtaining the depth map of the plane area in S2 through the depth camera; S4, calculating the inclination angle of the plane area according to the depth map in S3 , that is, the inclination angle of the tower to be tested; the advantage of the present invention is that, combined with machine learning, it can automatically identify and complete the detection, which improves the detection efficiency; The image of the tower can greatly improve the measurement accuracy.

Description

technical field [0001] The invention relates to the technical field of image recognition, in particular to a machine vision-based intelligent tower tilt detection method. Background technique [0002] In today's society, the importance of power and communication in daily life and industrial production is self-evident, which puts forward high requirements on the stability of power grid and communication network. As the foundation of the aerial power grid and communication network, the tower plays a vital role in the power grid and communication network. Due to the influence of external forces, climate change, geological disasters and other factors, towers are prone to tilt, which brings potential safety hazards to power grids and communication networks, and requires timely investigation and repair. [0003] Existing tower inclination measurement methods can be roughly divided into two categories: one is methods that require manual operation and intervention, such as plumb bo...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Patents(China)
IPC IPC(8): G06T7/66G06T7/50G06V10/774G06K9/00G06T7/70
CPCG06T7/50G06T7/66G06T7/70G06V20/13G06F18/214
Inventor 岳欣欣张圣超
Owner 涵涡智航科技(玉溪)有限公司