Aerial image-based method for detecting bird nest in power transmission line

A technology for power transmission lines and aerial images, which is applied in the directions of instruments, character and pattern recognition, scene recognition, etc., can solve the problem of ineffective detection of bird's nests, achieve unconventional scale and proportion improvement, low missed detection rate, and large amount of calculation small effect

Inactive Publication Date: 2017-07-21
TIANJIN UNIV
View PDF3 Cites 29 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

The shape of the bird's nest is often related to its location. There are irregular hemispheres and cubes, etc. Existing

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
  • Aerial image-based method for detecting bird nest in power transmission line
  • Aerial image-based method for detecting bird nest in power transmission line
  • Aerial image-based method for detecting bird nest in power transmission line

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0022] In order to make the technical solution of the present invention clearer, the present invention will be further elaborated below in conjunction with the accompanying drawings. The present invention is concretely realized according to the following steps:

[0023] The first step is to prepare the dataset.

[0024] (1) Prepare image data and label data.

[0025] Aerial pictures of power transmission lines, including aerial pictures of towers, obtained by drone patrol and helicopter shooting are collected, and images with bird's nests in power transmission lines are selected and manually marked. Record the coordinates of the upper left corner and the lower left corner of the bird's nest in the entire image as the label data of the image. In order to adapt to the structure of the neural network, the size of all pictures is normalized to obtain a set of pictures with a size of 600*400, which are used to train the bird rough selection network and the fine selection network ...

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 present invention relates to an aerial image-based method for detecting a bird nest in a power transmission line. The method includes the following steps that: a picture database with annotations and labels is constructed; two deep convolutional neural networks are trained separately: an obtained data set is utilized to modify a classification network VGG16, so that two detection networks can be formed, the two detection networks are trained separately, so that a bird nest roughly selected neural network and a bird nest finely selected neural network are obtained, the roughly selected neural network can obtain a bird nest roughly selected region through using a window sliding method in a feature graph, and the roughly selected region is utilized to train the bird nest finely selected neural network, and therefore, an optimal bird nest region can be obtained; and cross-training and fine-tuning are performed on the networks so as to construct a detection model: a cross-training method is adopted to re-train the obtained roughly selected neural network and finely selected neural network, so that the two networks can share the parameters of the first 13 convolutional layers, and the two networks are combined into an end-to-end convolutional neural network, so that the bird nest detection model can be constructed.

Description

technical field [0001] The invention belongs to the field of object detection, and relates to a method for detecting a bird's nest in an aerial image of a power transmission line by using a deep convolutional neural network. Background technique [0002] Transmission lines play a very important role in the power system and are directly related to the electricity consumption of thousands of households. Large-scale power outages will bring immeasurable losses to the national economy. Therefore, the safety of transmission lines is one of the issues of great concern to the power sector [1] . The natural activities of birds often interfere with the normal operation of transmission lines. For example, birds often nest on transmission line towers, and the falling of their nest materials can easily cause short circuits around the towers. [0003] Therefore, in order to ensure the safe and reliable operation of the power grid, it is necessary to control the nesting of birds on the ...

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
IPC IPC(8): G06K9/00G06K9/62
CPCG06V20/13G06F18/2415G06F18/214
Inventor 侯春萍管岱杨阳章衡光郎玥
Owner TIANJIN UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products