Power transmission line fitting detection method and system
A technology for transmission lines and detection methods, applied in neural learning methods, instruments, biological neural network models, etc., can solve problems such as sample imbalance, inability of deep detection models to detect key components, and failure to integrate models with business knowledge in the power field. , to improve the detection effect, alleviate the problem of sample imbalance and long-tailed distribution, and reduce the number of samples required.
Active Publication Date: 2021-05-28
NORTH CHINA ELECTRIC POWER UNIV (BAODING) +4
View PDF10 Cites 1 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
However, the algorithm based on deep learning only applies and improves the applicability of the target detection model based on the characteristics of the transmission line fittings itself, and fails to effectively integrate the model with business knowledge in the power field
At the same time, limited by the particularity of the working environment of fittings, there are often serious sample imbalances among various fittings. For some fittings with fewer samples, a single depth detection model cannot accurately detect key components
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 moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment
[0059] figure 1 It is a flow chart of the detection method for transmission line fittings in the embodiment of the present invention, as figure 1 As shown, a transmission line fittings detection method, comprising:
[0060] Step 101: Obtain a data set of fittings; the data set of fittings includes a plurality of aerial images of power transmission lines.
[0061] Step 102: According to the aerial image of the power transmission line, the Faster R-CNN algorithm is used to obtain visual features.
[0062] Step 102 specifically includes:
[0063] Extract the multi-channel features of the aerial image of the transmission line to obtain the image feature map;
[0064] Slide the image feature map according to various anchor frames of preset sizes and ratios to generate multiple candidate frames;
[0065] Use the non-maximum value suppression algorithm to screen the candidate frames to obtain multiple target candidate areas;
[0066] The target candidate area is divided into n×n...
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
Login to View More Abstract
The invention discloses a power transmission line fitting detection method and system. The method comprises the following steps: acquiring a hardware fitting data set, and obtaining visual features by adopting a Faster R-CNN algorithm according to an aerial image of a power transmission line; learning by adopting a multilayer perceptron algorithm according to the aerial image and the visual features of the power transmission line to obtain a co-occurrence graph adjacency matrix after learning; performing information transmission on the visual features according to the learned concurrence graph adjacency matrix to obtain enhanced features; and cascading the visual feature and the enhanced feature to obtain a fusion feature, and performing full connection on the fusion feature to obtain a fitting type and a fitting position. By adopting the method and the system, the requirement of the traditional deep learning model on the sample number of each fitting in the data set can be reduced, the problems of sample imbalance and long tail distribution of the aerial photography data of the power transmission line are relieved, and the detection effect of the fitting of the power transmission line is improved.
Description
technical field [0001] The invention relates to the technical field of transmission line fitting detection, in particular to a detection method and system for transmission line fittings. Background technique [0002] In recent years, with the rapid development and comprehensive coverage of the power grid, transmission lines are the core system in power transmission, and their stable operation has a vital impact on the security of the power grid. Among them, as an important accessory of the transmission line, the fittings play a role in fixing, protecting and connecting, and maintaining the stable operation of the entire line. Since the working environment of fittings is often in a complex and harsh field environment, it is prone to defects such as corrosion, deformation, and damage. Therefore, regular inspections of transmission lines will greatly reduce the occurrence of transmission line failures. [0003] With the development of digital image processing and UAV monitorin...
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
Login to View More Patent Type & Authority Applications(China)
IPC IPC(8): G06K9/00G06K9/62G06N3/04G06N3/08
CPCG06N3/04G06N3/08G06V20/13G06V20/46G06V20/41G06F18/2415
Inventor 翟永杰杨旭王乾铭张效铭赵振兵
Owner NORTH CHINA ELECTRIC POWER UNIV (BAODING)