Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Component Detection Method Based on Multi-threshold Cascaded Detector in Dual-Channel Power Map

A component and dual-channel technology, applied in the field of dual-channel power map component detection, can solve problems such as difficulty in achieving higher accuracy and difficulty in detection, and achieve the effect of fast and accurate positioning and fast type recognition.

Active Publication Date: 2022-05-31
STATE GRID SHANDONG ELECTRIC POWER +2
View PDF8 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

The above methods and the target detection algorithms commonly used in the industry are usually difficult to detect targets with the above characteristics, and it is difficult to achieve high accuracy

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
  • Component Detection Method Based on Multi-threshold Cascaded Detector in Dual-Channel Power Map
  • Component Detection Method Based on Multi-threshold Cascaded Detector in Dual-Channel Power Map
  • Component Detection Method Based on Multi-threshold Cascaded Detector in Dual-Channel Power Map

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0054] The original atlas and sub-atlas of the power map are preferably preprocessed to meet preset size requirements, such as short

[0058] S1: First, the residual network ResNet101 is used as the network for feature extraction, and the feature pyramid structure is added in it.

[0062] S5: Use the original image training set and the sub-image training set to perform end-to-end joint training on the network architecture, namely

[0063] The present invention proposes a feature fusion and multi-threshold cascade detection algorithm based on convolutional neural network. first for

[0064] In the embodiment of the present invention, the residual network ResNet101 is used as the network for feature extraction, and the

[0065] S11: Construct a bottom-up feature extraction convolutional network to obtain enhanced semantic information, and obtain the ResNet101 network

[0066] S12: FPN uses a 1x1 convolution filter to reduce the number of channels of C5 to 256 dimensions to obtain an...

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 dual-channel power diagram component detection method based on a multi-threshold cascaded detector, which belongs to the technical field of power diagram intelligent identification. The method includes: obtaining the original atlas of the power diagram to be detected, The set is divided according to the distribution area of ​​the parts to obtain the sub-atlas, and the relative position of the sub-atlas to the original image is recorded; the original atlas and the sub-atlas are sent to the pre-trained dual-channel detection model to obtain the detection results of the original image and the sub-atlas. Figure detection results; according to the recorded relative positions of the sub-images, summarizing the original image detection results and the sub-image detection results to obtain the power diagram component detection results. The present invention utilizes the target detection technology based on the convolutional neural network to realize fast and accurate positioning and fast type identification of various types of power map components.

Description

Two-channel power map component detection method based on multi-threshold cascade detector technical field The present invention relates to the technical field of intelligent identification of power drawings, in particular to a kind of multi-threshold cascade detector based Two-channel power map component detection method. Background technique [0002] Massive power engineering drawings require enormous human resources for information statistics and content analysis. existing In the technology, by designing relevant algorithms, precise positioning and type identification of components in power diagrams can not only reduce labor costs. The pressure of the staff to analyze the image content is still the intelligent image recognition, the extraction of important image information, and the automatic error correction of the image content. an integral part of the technique. Object detection is a technique used to identify the location of objects in an image and determine t...

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): G06V30/422G06V10/26G06V10/774G06V10/764G06K9/62G06N3/04G06N3/08
CPCG06N3/08G06V30/422G06V10/26G06N3/045G06F18/214G06F18/24Y04S10/50
Inventor 李昊管荑王杉李慧聪石玮王黎陈静于嘉琳吴建龙
Owner STATE GRID SHANDONG ELECTRIC POWER
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

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

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com