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

Power transmission and transformation scene description automatic generation method fusing global information and local information

A technology for local information and scene description, applied in neural learning methods, computer components, character and pattern recognition, etc., can solve problems such as inability to specify scene information, inability to selectively describe specific objects, limited description ability, etc.

Active Publication Date: 2020-06-23
SHANDONG UNIV +3
View PDF11 Cites 21 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

The model mentioned in this patent document can only describe the objects in the picture, and its description ability is limited: neither can it specify the scene information corresponding to the picture, nor can it selectively describe specific objects

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
  • Power transmission and transformation scene description automatic generation method fusing global information and local information
  • Power transmission and transformation scene description automatic generation method fusing global information and local information
  • Power transmission and transformation scene description automatic generation method fusing global information and local information

Examples

Experimental program
Comparison scheme
Effect test

Embodiment

[0082] like figure 1 shown. A method for automatically generating a description of a power transmission and transformation scene by fusing global information and local information, comprising the following steps:

[0083] S1: Build a training set: collect pictures and manually mark the images in the pictures;

[0084] S2: Extract the global image features in the picture through the convolutional neural network;

[0085] S3: using the deep learning-based target detection algorithm Yolo network to extract local image features in the picture, and fused with the global image features in step S2 to obtain fusion features;

[0086] S4: Segment the Chinese description corresponding to the image according to the semantics, and construct a Chinese dictionary;

[0087] S5: Using the data after word segmentation in step S4 and the fusion features to train an improved long-short-term memory network based on a two-layer attention mechanism;

[0088] S6: In the use stage, the steps S1-S...

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 power transmission and transformation scene description automatic generation method fusing global information and local information, and aims to perform image Chinese description generation on a power transmission and transformation scene: extracting local and global feature information of an image by combining deep learning and a traditional computer vision technology; the semantic information of each Chinese segmented word is associated with the fused feature information; a long-term and short-term memory neural network is improved, a double-layer attention mechanism is added, an established sequence model generates distribution of an attention variable at each moment according to image feature information and semantic information, and finally Chinese description corresponding to an image is generated.

Description

technical field [0001] The invention discloses a method for automatically generating a power transmission and transformation scene description that integrates global information and local information, and belongs to the technical field of smart grids. Background technique [0002] The power system is a huge system covering the production, transmission and application of electric energy. It is composed of various power plants, power transmission and transformation lines, various substations and distribution stations, and users. Transmission lines play the role of connecting and connecting the upper and lower levels in the operation of the power system, and are very important for maintaining the stable and safe operation of the system. The harsh environment of power transmission and transformation lines makes it one of the links that are prone to failure in the power system, which will affect the safe and stable operation of the power grid. [0003] Traditional transmission a...

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 Applications(China)
IPC IPC(8): G06N3/04G06N3/08G06K9/46G06K9/62
CPCG06N3/08G06V10/40G06V10/44G06N3/045G06N3/044G06F18/253Y04S10/50
Inventor 聂礼强战新刚郑晓云姚一杨甘甜宓生润
Owner SHANDONG UNIV
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