Submitted by Ricardo Cedraz Duque Moliterno (ricardo.moliterno@uefs.br) on 2016-08-29T21:08:03Z
No. of bitstreams: 1
Dissertacao-Leonardo_Melo-UEFS_PGCA-2.pdf: 1657755 bytes, checksum: 78e6ca37c9610faf17a158b94e25bcb8 (MD5) / Made available in DSpace on 2016-08-29T21:08:03Z (GMT). No. of bitstreams: 1
Dissertacao-Leonardo_Melo-UEFS_PGCA-2.pdf: 1657755 bytes, checksum: 78e6ca37c9610faf17a158b94e25bcb8 (MD5)
Previous issue date: 2016-04-08 / Funda??o de Amparo ? Pesquisa do Estado da Bahia - FAPEB / The Huanglongbing (HLB) is the disease of greatest concern for growers because they spread quickly and cause severe symptoms. The Diaphorina citri insect is the main vector of the HLB. The application of insecticides is a control measure of the vector insect of the HLB widely adopted. The amount of pesticides needed for an effective control of this insect is better estimated if such application is combined with a monitoring of its population by yellow sticky traps. These insects are captured for a manual count in research centers. So, this research aims to discover a computational approach of classification of Diaphorina citri insect images with higher accuracy rate that the classification rate currently used in manual counting procedure and thus enable the automation of this important counting procedure. For this, have been tried and combined computational methods for features extraction (ORB, SIFT, SURF, BRISK and FREAK), grouping of characteristics (Mini Batch K-Means) and features classification for machine learning (KNN and SVM), using a generated bank with 1152 images of insects. The best found classification approach (extractor SURF/SIFT, BoF with Diaphorina citri features and SVM with core RBF) generated classification performance results for the metric accuracy, which outperformed the best measured result in research that evaluated the counting manual process. In this approach, the highest achieved accuracy, in the cross validation process, was 98.17% and was 2.54% as standard deviation and the accuracy of the final test of generalization model was 99.14%. The achieved result is of great importance for the control of HLB. The achieved classification accuracy rates were higher than rates reported in the manual procedure, making possible the construction of computer systems to high accuracy for the control of this insect. This automated control can provide significant savings of funds. / O Huanglongbing (HLB) ? a doen?a de maior preocupa??o para os citricultores por se propagar com rapidez e provocar severos sintomas. O inseto Diaphorina citri ? o principal vetor do HLB. A aplica??o de inseticidas ? uma medida de controle do inseto vetor do HLB amplamente adotada. A quantidade de inseticidas necess?ria para um controle efetivo desse inseto ? melhor estimada se essa aplica??o for combinada a um monitoramento de sua popula??o por meio de armadilhas adesivas amarelas. Esses insetos s?o capturados para uma contagem manual em centros de pesquisa. Ent?o, esta pesquisa tem por objetivo descobrir uma abordagem computacional de classifica??o de imagens de insetos Diaphorina citri com taxa de acur?cia de classifica??o maiores que a taxa de classifica??o utilizada atualmente no procedimento manual de contagem e, assim, possibilitar a automa??o desse importante procedimento de contagem. Para isso, foram experimentados e combinados m?todos computacionais para a extra??o de caracter?sticas (ORB, SIFT, SURF, BRISK e FREAK), agrupamento de caracter?sticas (Mini Batch K-Means) e classifica??o de caracter?sticas por aprendizagem de m?quina (KNN e SVM), utilizando um banco gerado com 1152 imagens de insetos. A melhor abordagem de classifica??o encontrada (extrator SURF/SIFT, BoF com caracter?sticas do Diaphorina citri e SVM com n?cleo RBF) gerou resultados de desempenho de classifica??o, pela m?trica da acur?cia, que superaram o melhor resultado medido na pesquisa que avaliou o processo de contagem manual. Nessa abordagem, a maior acur?cia atingida no processo de valida??o cruzada foi de 98,17% e teve 2,54% como desvio padr?o e a acur?cia do teste final de generaliza??o de modelo foi de 99,14%. O resultado alcan?ado ? de grande import?ncia para o controle do HLB. As taxas de acur?cia de classifica??o alcan?adas foram superiores as taxas relatadas no procedimento manual, tornando vi?vel a constru??o de sistemas computacionais de alta acur?cia para o controle desse inseto. Esse controle automatizado pode proporcionar uma economia significativa de recursos financeiros.
| Identifer | oai:union.ndltd.org:IBICT/oai:tede2.uefs.br:8080:tede/377 |
| Date | 08 April 2016 |
| Creators | Melo, Jos? Leonardo dos Santos |
| Contributors | Angelo, Michele F?lvia |
| Publisher | Universidade Estadual de Feira de Santana, Mestrado em Computa??o Aplicada, UEFS, Brasil, DEPARTAMENTO DE TECNOLOGIA |
| Source Sets | IBICT Brazilian ETDs |
| Language | Portuguese |
| Detected Language | English |
| Type | info:eu-repo/semantics/publishedVersion, info:eu-repo/semantics/masterThesis |
| Format | application/pdf |
| Source | reponame:Biblioteca Digital de Teses e Dissertações da UEFS, instname:Universidade Estadual de Feira de Santana, instacron:UEFS |
| Rights | info:eu-repo/semantics/openAccess |
| Relation | 303317282311144204, 600, 600, 600, 600, 4335108523020347051, 8930092515683771531, -8233071094704392586 |
Page generated in 0.0017 seconds