An Intelligent Weed Recognition Method Based on Optical Patrol Image

Power inspection is currently carried out mainly by drones. The weeds around the power equipment may cause potential safety hazards when the patrol images acquired by drones are used for patrol inspection, it is therefore necessary to recognize the weeds in the image. In this paper, a method for int...

Full description

Saved in:
Bibliographic Details
Main Authors: Guoliang YUE, Yanqiao LU, Hao CHANG, Cuiying SUN
Format: Article
Language:zho
Published: State Grid Energy Research Institute 2019-11-01
Series:Zhongguo dianli
Subjects:
Online Access:https://www.electricpower.com.cn/CN/10.11930/j.issn.1004-9649.201902152
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850222298243530752
author Guoliang YUE
Yanqiao LU
Hao CHANG
Cuiying SUN
author_facet Guoliang YUE
Yanqiao LU
Hao CHANG
Cuiying SUN
author_sort Guoliang YUE
collection DOAJ
description Power inspection is currently carried out mainly by drones. The weeds around the power equipment may cause potential safety hazards when the patrol images acquired by drones are used for patrol inspection, it is therefore necessary to recognize the weeds in the image. In this paper, a method for intelligent recognition of weeds is proposed for power patrol inspection based on optical patrol images. Based on the feature of weeds in the optical images, and combined with the convolutional neural network method, the problem of weed recognition near the power equipment in optical patrol images is solved. By amplifying and preprocessing the sample data of the optical patrol images, and introducing the region proposal network, the image features of the fixed number of candidate frames are extracted from the images. Then the network is connected to the improved image classification network to obtain a final convolutional neural network model. The experiments show that the accuracy rate can reach 97.98%, and the average time taken for detecting a 600×600 image is around 0.256 seconds, which meets the requirements of efficient recognition while ensuring accuracy.
format Article
id doaj-art-ff51e7a6cb3f4032bc6e72201955f0f1
institution OA Journals
issn 1004-9649
language zho
publishDate 2019-11-01
publisher State Grid Energy Research Institute
record_format Article
series Zhongguo dianli
spelling doaj-art-ff51e7a6cb3f4032bc6e72201955f0f12025-08-20T02:06:23ZzhoState Grid Energy Research InstituteZhongguo dianli1004-96492019-11-01521113814410.11930/j.issn.1004-9649.201902152zgdl-52-11-yueguoliangAn Intelligent Weed Recognition Method Based on Optical Patrol ImageGuoliang YUE0Yanqiao LU1Hao CHANG2Cuiying SUN3State Grid Hebei Electric Power Co., Ltd., Shijiazhuang 050041, ChinaState Grid Hebei Electric Power Research Institute, Shijiazhuang 050021, ChinaMaintenance Branch of State Grid Hebei Electric Power Co., Ltd., Shijiazhuang 050070, ChinaState Grid Hebei Electric Power Research Institute, Shijiazhuang 050021, ChinaPower inspection is currently carried out mainly by drones. The weeds around the power equipment may cause potential safety hazards when the patrol images acquired by drones are used for patrol inspection, it is therefore necessary to recognize the weeds in the image. In this paper, a method for intelligent recognition of weeds is proposed for power patrol inspection based on optical patrol images. Based on the feature of weeds in the optical images, and combined with the convolutional neural network method, the problem of weed recognition near the power equipment in optical patrol images is solved. By amplifying and preprocessing the sample data of the optical patrol images, and introducing the region proposal network, the image features of the fixed number of candidate frames are extracted from the images. Then the network is connected to the improved image classification network to obtain a final convolutional neural network model. The experiments show that the accuracy rate can reach 97.98%, and the average time taken for detecting a 600×600 image is around 0.256 seconds, which meets the requirements of efficient recognition while ensuring accuracy.https://www.electricpower.com.cn/CN/10.11930/j.issn.1004-9649.201902152patrol imageconvolutional neural networkregion proposalimage classificationweed recognitionartificial intelligence and big data application
spellingShingle Guoliang YUE
Yanqiao LU
Hao CHANG
Cuiying SUN
An Intelligent Weed Recognition Method Based on Optical Patrol Image
Zhongguo dianli
patrol image
convolutional neural network
region proposal
image classification
weed recognition
artificial intelligence and big data application
title An Intelligent Weed Recognition Method Based on Optical Patrol Image
title_full An Intelligent Weed Recognition Method Based on Optical Patrol Image
title_fullStr An Intelligent Weed Recognition Method Based on Optical Patrol Image
title_full_unstemmed An Intelligent Weed Recognition Method Based on Optical Patrol Image
title_short An Intelligent Weed Recognition Method Based on Optical Patrol Image
title_sort intelligent weed recognition method based on optical patrol image
topic patrol image
convolutional neural network
region proposal
image classification
weed recognition
artificial intelligence and big data application
url https://www.electricpower.com.cn/CN/10.11930/j.issn.1004-9649.201902152
work_keys_str_mv AT guoliangyue anintelligentweedrecognitionmethodbasedonopticalpatrolimage
AT yanqiaolu anintelligentweedrecognitionmethodbasedonopticalpatrolimage
AT haochang anintelligentweedrecognitionmethodbasedonopticalpatrolimage
AT cuiyingsun anintelligentweedrecognitionmethodbasedonopticalpatrolimage
AT guoliangyue intelligentweedrecognitionmethodbasedonopticalpatrolimage
AT yanqiaolu intelligentweedrecognitionmethodbasedonopticalpatrolimage
AT haochang intelligentweedrecognitionmethodbasedonopticalpatrolimage
AT cuiyingsun intelligentweedrecognitionmethodbasedonopticalpatrolimage