Laser data-based substation foreign object identification method

A recognition method and substation technology, applied in image data processing, character and pattern recognition, reradiation of electromagnetic waves, etc., can solve problems such as being easily affected by the surrounding environment, unstable recognition results, and decline in recognition results, and achieve color purity. , high power density, strong directional effect

Active Publication Date: 2018-06-01
国网黑龙江省电力有限公司超高压公司 +1
View PDF7 Cites 14 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0002] At present, in this field, foreign object recognition is mainly carried out through CCD camera acquisition and binocular camera method. However, when the target data collected by CCD camera is similar to the color of the surrounding environment, the recognition result drops sharply, which is v

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
  • Laser data-based substation foreign object identification method
  • Laser data-based substation foreign object identification method
  • Laser data-based substation foreign object identification method

Examples

Experimental program
Comparison scheme
Effect test

Example Embodiment

[0040] Example 1:

[0041] A method for identifying foreign objects in substations based on laser data. The steps of the method of identification are as follows: First, use laser radar to collect laser data from multiple areas, and build a model point cloud collection through analysis, such as figure 2 As shown, the laser radar is used to collect laser data from a certain area to be measured to form a test point cloud, such as image 3 As shown, the distribution information of the point cloud is filtered for isolated point removal by calculation, and then the distribution histogram of the point cloud data in the three directions of the xyz axis is counted, the key points are extracted through uniform sampling and filtering, and the feature vector is calculated to establish the kdtree matching pair index , After SVD, the conversion matrix is ​​solved, and the test point cloud coordinates are converted to the model point cloud for ICP iterative registration, and then the over-distan...

Example Embodiment

[0042] Example 2:

[0043] According to the laser data-based foreign body recognition method in substations according to the first embodiment, the method steps for the point cloud distribution information to remove and filter the isolated points are as follows: For each point P, using R as the search radius will get K neighbor points , Calculate the distance Di from all K neighboring points to P, and sum all the distances to get D,

[0044]

[0045] And the average distance M,

[0046]

[0047] Calculate the sum of squared distances S of K neighboring points,

[0048]

[0049] Calculate the covariance Var,

[0050]

[0051] And mean square error,

[0052]

[0053] After the above calculations are obtained, the filtering threshold T can be calculated:

[0054]

[0055] Bring back the threshold T and compare it with the previously calculated distance Di of K neighboring points. Set a reserved flag if it is less than the threshold, traverse all points once, and count the signs of the neighb...

Example Embodiment

[0056] Example 3:

[0057] According to the laser data-based foreign body identification method in substations according to embodiment 1 or 2, the method steps of the distribution histogram are as follows: obtain three histogram distributions Fx, Fy, Fz, and calculate the excess of the three histograms respectively. 75% of the test point cloud distribution range has been tested, and the size L1, L2, L3 required for dividing the range into 10 equal parts is obtained. After calculating the average value Lm of L1, L2, L3, the entire test point cloud data can be uniformly sampled The scale Lm, such as Figure 8 Shown.

[0058]

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 relates to a laser data-based substation foreign object identification method. At present, foreign object recognition is easily affected by environments, modeling is complex, and algorithms are time-consuming. The method of the present invention includes the following steps that: a laser radar is adopted to collect laser data in a plurality of regions; a model point cloud set is built through analysis and settlement; the laser radar is used to collect the laser data of a certain region to be measured, so that the laser data can form test point cloud; outlier removal filtering isperformed on point cloud distribution information through calculation; the distribution histograms of point cloud data in an x-axis direction, a y-axis direction and a z-axis direction are put into statistics; key points are extracted through uniform sampling filtering, and feature vectors are calculated; kdtree matching pair indexes are established; conversion matrix solving is performed throughSVD; the coordinates of the test point cloud are converted, and ICP iterative registration is performed on the converted coordinates of the test point cloud and model point cloud; point pair pairing distances are calculated through using the kdtree, and distance exceeding points are screened out; the density of a distance exceeding point set is calculated; and whether a foreign object exists is judged. The laser data-based substation foreign object identification method of the invention is used for substation foreign object identification.

Description

Technical field: [0001] The invention relates to a method for identifying foreign matter in a substation based on laser data. Background technique: [0002] At present, in this field, foreign object recognition is mainly carried out through CCD camera acquisition and binocular camera method. However, when the target data collected by CCD camera is similar to the color of the surrounding environment, the recognition result drops sharply, which is very vulnerable to the influence of the surrounding environment. The recognition results are unstable, and although the binocular camera can obtain three-dimensional information, the modeling is complicated, and the algorithm is time-consuming. The evaluation of practical use is not ideal. Invention content: [0003] The purpose of the present invention is to provide a method for identifying foreign objects in substations based on laser data. Test whether there is a foreign object in the point cloud, and introduce the spatial dept...

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): G01S17/88G06K9/62G06T7/33
CPCG06T7/33G01S17/88G06T2207/10028G06V10/758
Inventor 李海峰刘延辉阳薇孔波王国军侯木齐
Owner 国网黑龙江省电力有限公司超高压公司
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

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap