Identity statement area
Reference TypeConference Paper (Conference Proceedings)
Last Update2021: (UTC)
Metadata Last Update2021: (UTC) administrator
Citation KeyRochaVPBAMSRSMN:2021:CoStMe
TitleA Comparative Study of Methods based on Deep Neural Networks for Self-reading of Energy Consumption in a Chatbot Application Context
Access Date2022, Jan. 24
Number of Files1
Size877 KiB
Context area
Author 1 Rocha, Carlos Vinicios Martins
 2 Vieira, Pedro Henrique Carvalho
 3 Pinto, Antonio Moreira
 4 Bernhard, Pedro Vinnicius
 5 Anchieta Junior, Ricardo José Fernandes
 6 Marques, Ricardo Costa da Silva
 7 Silva, Italo Francyles Santos da
 8 Rocha, Simara Vieira da
 9 Silva, Aristófanes Corrêa
10 Monteiro, Eliana Márcia Garros
11 Nogueira, Hugo Daniel Castro Silva
Affiliation 1 Federal University of Maranhão
 2 Federal University of Maranhão
 3 Federal University of Maranhão
 4 Federal University of Maranhão
 5 Federal University of Maranhão
 6 Federal University of Maranhão
 7 Federal University of Maranhão
 8 Federal University of Maranhão
 9 Federal University of Maranhão
10 Equatorial Energy Group
11 Equatorial Energy Group
EditorPaiva, Afonso
Menotti, David
Baranoski, Gladimir V. G.
Proença, Hugo Pedro
Junior, Antonio Lopes Apolinario
Papa, João Paulo
Pagliosa, Paulo
dos Santos, Thiago Oliveira
e Sá, Asla Medeiros
da Silveira, Thiago Lopes Trugillo
Brazil, Emilio Vital
Ponti, Moacir A.
Fernandes, Leandro A. F.
Avila, Sandra
Conference NameConference on Graphics, Patterns and Images, 34 (SIBGRAPI)
Conference LocationGramado (Virtual), Brazil
DateOctober 18th to October 22nd, 2021
PublisherIEEE Computer Society
Publisher CityLos Alamitos
Book TitleProceedings
Tertiary TypeIndustry Application Paper
History (UTC)2021-09-10 14:22:17 :: -> administrator ::
2021-11-12 11:47:13 :: administrator -> :: 2021
Content and structure area
Is the master or a copy?is the master
Content Stagecompleted
Content TypeExternal Contribution
energy consumption
chatbot application
image processing
deep learning
AbstractSelf-reading is a process in which the consumer is responsible for measuring his own energy consumption, which can be done through digital platforms, such as websites or mobile applications. The Equatorial Energy group's electric utilities have been working on developing a chatbot application through which consumers can send an image of their energy meter to a server that runs a method based on image processing and deep learning for the automatic recognition of consumption reading. However, the incorporation of these methods in a solution available to the public should consider factors such as response time and accuracy, so that it presents a satisfactory response time when it needs to handle a large number of simultaneous requests. Therefore, this paper presents a comparative study between approaches developed for the automatic recognition of consumption readings in images of electric meters sent to the server. Response time performances are analyzed through stress tests that simulate the real application scenario. The mean average precision (mAP) and the accuracy metrics of the methods are also analyzed in order to evaluate the generalization of the used convolutional neural networks. > SDLA > SIBGRAPI 2021 > A Comparative Study...
doc Directory Contentaccess
source Directory Contentthere are no files
agreement Directory Content
agreement.html 10/09/2021 11:22 1.3 KiB 
Conditions of access and use area
data URL
zipped data URL
Target FileAutoclara___Sibgrapi_2021___English__sem_subpastas_.pdf
Allied materials area
Next Higher Units8JMKD3MGPEW34M/45PQ3RS
Notes area
Empty Fieldsaccessionnumber archivingpolicy archivist area callnumber copyholder copyright creatorhistory descriptionlevel dissemination documentstage doi edition electronicmailaddress group holdercode isbn issn label lineage mark nextedition notes numberofvolumes orcid organization pages parameterlist parentrepositories previousedition previouslowerunit progress project readergroup readpermission resumeid rightsholder secondarydate secondarykey secondarymark secondarytype serieseditor session shorttitle sponsor subject tertiarymark type url versiontype volume
Description control area
e-Mail (login)