Processing

Please wait...

Settings

Settings

Goto Application

1. CN110084165 - Intelligent identification and early warning method for abnormal events in open scene in power field based on edge calculation

Office
China
Application Number 201910319835.6
Application Date 19.04.2019
Publication Number 110084165
Publication Date 02.08.2019
Grant Number 110084165
Grant Date 07.02.2020
Publication Kind B
IPC
G06K 9/00
GPHYSICS
06COMPUTING; CALCULATING OR COUNTING
KRECOGNITION OF DATA; PRESENTATION OF DATA; RECORD CARRIERS; HANDLING RECORD CARRIERS
9Methods or arrangements for reading or recognising printed or written characters or for recognising patterns, e.g. fingerprints
G06N 3/04
GPHYSICS
06COMPUTING; CALCULATING OR COUNTING
NCOMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS
3Computer systems based on biological models
02using neural network models
04Architecture, e.g. interconnection topology
G06N 3/08
GPHYSICS
06COMPUTING; CALCULATING OR COUNTING
NCOMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS
3Computer systems based on biological models
02using neural network models
08Learning methods
CPC
G06K 9/00718
GPHYSICS
06COMPUTING; CALCULATING; COUNTING
KRECOGNITION OF DATA; PRESENTATION OF DATA; RECORD CARRIERS; HANDLING RECORD CARRIERS
9Methods or arrangements for reading or recognising printed or written characters or for recognising patterns, e.g. fingerprints
00624Recognising scenes, i.e. recognition of a whole field of perception; recognising scene-specific objects
00711Recognising video content, e.g. extracting audiovisual features from movies, extracting representative key-frames, discriminating news vs. sport content
00718Higher-level, semantic clustering, classification or understanding of video scenes, e.g. detection, labelling or Markovian modelling of sport events or news items
G06K 9/00771
GPHYSICS
06COMPUTING; CALCULATING; COUNTING
KRECOGNITION OF DATA; PRESENTATION OF DATA; RECORD CARRIERS; HANDLING RECORD CARRIERS
9Methods or arrangements for reading or recognising printed or written characters or for recognising patterns, e.g. fingerprints
00624Recognising scenes, i.e. recognition of a whole field of perception; recognising scene-specific objects
00771Recognising scenes under surveillance, e.g. with Markovian modelling of scene activity
G06K 2009/00738
GPHYSICS
06COMPUTING; CALCULATING; COUNTING
KRECOGNITION OF DATA; PRESENTATION OF DATA; RECORD CARRIERS; HANDLING RECORD CARRIERS
9Methods or arrangements for reading or recognising printed or written characters or for recognising patterns, e.g. fingerprints
00624Recognising scenes, i.e. recognition of a whole field of perception; recognising scene-specific objects
00711Recognising video content, e.g. extracting audiovisual features from movies, extracting representative key-frames, discriminating news vs. sport content
00738Event detection
G06N 3/0454
GPHYSICS
06COMPUTING; CALCULATING; COUNTING
NCOMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS
3Computer systems based on biological models
02using neural network models
04Architectures, e.g. interconnection topology
0454using a combination of multiple neural nets
G06N 3/08
GPHYSICS
06COMPUTING; CALCULATING; COUNTING
NCOMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS
3Computer systems based on biological models
02using neural network models
08Learning methods
Applicants SHANDONG UNIVERSITY
山东大学
智洋创新科技股份有限公司
国网浙江省电力有限公司
Inventors NIE LIQIANG
聂礼强
SONG XUEMENG
宋雪萌
SUN TENG
孙腾
XU KE
许克
YAO YIYANG
姚一杨
SU SHIHUA
宿仕华
Agents 济南金迪知识产权代理有限公司 37219
Title
(EN) Intelligent identification and early warning method for abnormal events in open scene in power field based on edge calculation
(ZH) 基于边缘计算的电力领域开放场景下异常事件的智能识别与预警方法
Abstract
(EN)
An intelligent identification and early warning method for abnormal events in an open scene in the power field based on edge calculation compresses and transplants an improved SSD target detection model to a mobile terminal, gives full play to the advantages of the edge calculation, and takes an Android terminal as an optimal scheme through experiments. According to the method, the Conv4 _ x feature layer and the Conv5 _ x feature layer in the VGG16 network are fused, and then the fused feature layer directly acts on the final prediction layer, so that the accuracy of small target detection isimproved; meanwhile, various basic weather conditions such as sunny days, cloudy days, rainy days and foggy days are summarized, and the image enhancement technology is used for increasing training data under different scenes, so that the generalization ability of the model is improved.

(ZH)
一种基于边缘计算的电力领域开放场景下异常事件的智能识别与预警方法,本发明将改进的SSD目标检测模型压缩并移植到移动端,充分发挥边缘计算的优势,通过实验,本发明将安卓端作为一个优选方案;本发明将VGG16网络中的Conv4_x特征层和Conv5_x特征层相融合,再将融合之后的特征层直接作用到最后的预测层,以此来提高小目标检测的准确率;同时,本发明总结出多种基本的天气情况:晴天、阴天、雨天、雾天等,并使用图像增强技术来增加不同场景下的训练数据,以此提高模型的泛化能力。