Estimativa da germinação de sementes de Pityrocarpa moniliformis (Benth.) Luckow & R. W. Jobson usando Deep Learning
The possibility of estimating the germination of a seed lot through the use of Deep Learning has shown potential as a complementary method to the analysis of seed quality. In this research, we evaluated the efficiency of using Deep Learning, with convolutional neural networks, to estimate the ger...
Na minha lista:
| Autor principal: | |
|---|---|
| Outros Autores: | |
| Formato: | Dissertação |
| Idioma: | pt_BR |
| Publicado em: |
Universidade Federal do Rio Grande do Norte
|
| Assuntos: | |
| Endereço do item: | https://repositorio.ufrn.br/handle/123456789/47623 |
| Tags: |
Adicionar Tag
Sem tags, seja o primeiro a adicionar uma tag!
|
| id |
ri-123456789-47623 |
|---|---|
| record_format |
dspace |
| institution |
Repositório Institucional |
| collection |
RI - UFRN |
| language |
pt_BR |
| topic |
Espécies florestais Inteligência artificial Redes convolucionais Aprendizado profundo |
| spellingShingle |
Espécies florestais Inteligência artificial Redes convolucionais Aprendizado profundo Andrade, Francisca Adriana Ferreira de Estimativa da germinação de sementes de Pityrocarpa moniliformis (Benth.) Luckow & R. W. Jobson usando Deep Learning |
| description |
The possibility of estimating the germination of a seed lot through the use of Deep Learning
has shown potential as a complementary method to the analysis of seed quality. In this
research, we evaluated the efficiency of using Deep Learning, with convolutional neural
networks, to estimate the germination of Pityrocarpa moniliformis (Benth.) Luckow & R. W.
Jobson seeds. 1000 seeds were randomly selected from four lots, 250 seeds from each,
which were used in germination tests and computational analyses. A scanner was used to
capture the images of the seeds, from which the images of each seed were obtained
individually. The seed images were used to implement, train and test the convolutional neural
networks in the computational algorithm created in this research, aiming at comparing the
results obtained from the computational analysis with those of the individual germination of
each seed. Therefore, after acquiring the images, the seeds were placed to germinate,
identifying each one individually. After the germination period, the seeds were divided into
two classes: germinated (0) and non-germinated (1). From the images of the seeds before
germination, and with the individual result of each seed, the computational analysis was
carried out. The pre-trained networks after five epochs of execution indicated a tendency to
improve accuracy, however, there were also signs of overfitting, since the performance in the
training data was better than the validation (test) data. The recall (sensitivity) was greater
than 90% in all models for the class of germinated seeds. The recall value was much lower
for the non-germinated class, both below 20%. For the model proposed, 85% of recall was
obtained for the class of non-germinated seeds and 18% for the germinated ones, which may
have occurred due to the overlap of the classes of interests. The results of the pre-trained
networks and the proposed model proved to be inefficient, as they cannot adequately
distinguish the classes of interest to assess the efficiency of using Deep Learning to estimate
the germination of Pityrocarpa moniliformis seeds, but the analyzes indicate the need to
improve and adjust the pre-processing of the images and require more investigation time and
more tests to configure the models. |
| author2 |
Pereira, Márcio Dias |
| author_facet |
Pereira, Márcio Dias Andrade, Francisca Adriana Ferreira de |
| format |
masterThesis |
| author |
Andrade, Francisca Adriana Ferreira de |
| author_sort |
Andrade, Francisca Adriana Ferreira de |
| title |
Estimativa da germinação de sementes de Pityrocarpa moniliformis (Benth.) Luckow & R. W. Jobson usando Deep Learning |
| title_short |
Estimativa da germinação de sementes de Pityrocarpa moniliformis (Benth.) Luckow & R. W. Jobson usando Deep Learning |
| title_full |
Estimativa da germinação de sementes de Pityrocarpa moniliformis (Benth.) Luckow & R. W. Jobson usando Deep Learning |
| title_fullStr |
Estimativa da germinação de sementes de Pityrocarpa moniliformis (Benth.) Luckow & R. W. Jobson usando Deep Learning |
| title_full_unstemmed |
Estimativa da germinação de sementes de Pityrocarpa moniliformis (Benth.) Luckow & R. W. Jobson usando Deep Learning |
| title_sort |
estimativa da germinação de sementes de pityrocarpa moniliformis (benth.) luckow & r. w. jobson usando deep learning |
| publisher |
Universidade Federal do Rio Grande do Norte |
| publishDate |
2022 |
| url |
https://repositorio.ufrn.br/handle/123456789/47623 |
| work_keys_str_mv |
AT andradefranciscaadrianaferreirade estimativadagerminacaodesementesdepityrocarpamoniliformisbenthluckowrwjobsonusandodeeplearning AT andradefranciscaadrianaferreirade estimationofpityrocarpamoniliformisbenthluckowrwjobsonseedgerminationusingdeeplearning |
| _version_ |
1773963571368034304 |
| spelling |
ri-123456789-476232022-06-09T20:18:00Z Estimativa da germinação de sementes de Pityrocarpa moniliformis (Benth.) Luckow & R. W. Jobson usando Deep Learning Estimation of Pityrocarpa moniliformis (Benth.) Luckow & R. W. Jobson seed germination using Deep Learning Andrade, Francisca Adriana Ferreira de Pereira, Márcio Dias http://lattes.cnpq.br/7590560029057579 http://lattes.cnpq.br/2969947409452499 Silva, Larecio Junio da Oliveira, Laura Emmanuella Alves dos Santos Santana de Pires, Raquel Maria de Oliveira Espécies florestais Inteligência artificial Redes convolucionais Aprendizado profundo The possibility of estimating the germination of a seed lot through the use of Deep Learning has shown potential as a complementary method to the analysis of seed quality. In this research, we evaluated the efficiency of using Deep Learning, with convolutional neural networks, to estimate the germination of Pityrocarpa moniliformis (Benth.) Luckow & R. W. Jobson seeds. 1000 seeds were randomly selected from four lots, 250 seeds from each, which were used in germination tests and computational analyses. A scanner was used to capture the images of the seeds, from which the images of each seed were obtained individually. The seed images were used to implement, train and test the convolutional neural networks in the computational algorithm created in this research, aiming at comparing the results obtained from the computational analysis with those of the individual germination of each seed. Therefore, after acquiring the images, the seeds were placed to germinate, identifying each one individually. After the germination period, the seeds were divided into two classes: germinated (0) and non-germinated (1). From the images of the seeds before germination, and with the individual result of each seed, the computational analysis was carried out. The pre-trained networks after five epochs of execution indicated a tendency to improve accuracy, however, there were also signs of overfitting, since the performance in the training data was better than the validation (test) data. The recall (sensitivity) was greater than 90% in all models for the class of germinated seeds. The recall value was much lower for the non-germinated class, both below 20%. For the model proposed, 85% of recall was obtained for the class of non-germinated seeds and 18% for the germinated ones, which may have occurred due to the overlap of the classes of interests. The results of the pre-trained networks and the proposed model proved to be inefficient, as they cannot adequately distinguish the classes of interest to assess the efficiency of using Deep Learning to estimate the germination of Pityrocarpa moniliformis seeds, but the analyzes indicate the need to improve and adjust the pre-processing of the images and require more investigation time and more tests to configure the models. Coordenação de Aperfeiçoamento de Pessoal de Nível Superior - CAPES A possibilidade de estimar a germinação de um lote de sementes por meio uso do Deep Learning tem apresentado potencial, como método complementar à análise de qualidade de sementes. Nesta pesquisa, avaliou-se a eficiência do uso do Deep Learning, com redes neurais convolucionais, para estimar a germinação de sementes de Pityrocarpa moniliformis (Benth.) Luckow & R. W. Jobson. Selecionou-se 1000 sementes aleatoriamente de quatro lotes, 250 sementes de cada, as quais foram usadas nos testes de germinação e análises computacionais. Para a captação das imagens das sementes utilizou-se um escâner, a partir do qual se obteve as imagens de cada semente, individualmente. As imagens das sementes foram usadas para implementar, treinar e testar as redes neurais convolucionais no algoritmo computacional criado nesta pesquisa, visando à comparação dos resultados obtidos a partir da análise computacional com os da germinação individual de cada semente. Para tanto, após a aquisição das imagens, as sementes foram colocadas para germinar, identificando-se cada uma individualmente. Após o período de germinação, as sementes foram divididas em duas classes: germinadas (0) e não germinadas (1). A partir das imagens das sementes antes da germinação, e com o resultado individual de cada semente, procedeu-se a análise computacional. As redes pré-treinadas após cinco épocas de execução indicaram tendência de melhora da acurácia, no entanto também observou-se indícios de sobreajuste, uma vez que o desempenho nos dados de treino foram melhores do que os dados de validação (teste). O recall (sensibilidade) foi superior a 90% em todos os modelos para classe de sementes germinadas. O valor do recall foi bem menor para a classe não germinadas, ambas, abaixo de 20%. Para o modelo proposto obteve-se 85% de recall para a classe de sementes não germinadas e 18% para as germinadas, o que pode ter ocorrido devido a sobreposição das classes de interesses. Os resultados das redes pré- treinadas e do modelo proposto mostraram-se não eficientes, pois não conseguem distinguir as classes de interesse de forma adequada para avaliar a eficiência do uso de Deep Learning na estimativa da germinação das sementes de Pityrocarpa moniliformis, porém as análises indicam necessidade de melhorar e ajustar os pré-processamentos das imagens e carece de mais tempo de investigação e mais testes de configuração dos modelos. 2022-06-09T20:17:21Z 2022-06-09T20:17:21Z 2022-03-03 masterThesis ANDRADE, Francisca Adriana Ferreira de. Estimativa da germinação de sementes de Pityrocarpa moniliformis (Benth.) Luckow & R. W. Jobson usando Deep Learning. 2022. 45f. Dissertação (Mestrado em Ciências Florestais) - Escola Agrícola de Jundiaí, Universidade Federal do Rio Grande do Norte, Natal, 2022. https://repositorio.ufrn.br/handle/123456789/47623 pt_BR Acesso Aberto application/pdf Universidade Federal do Rio Grande do Norte Brasil UFRN PROGRAMA DE PÓS-GRADUAÇÃO EM CIÊNCIAS FLORESTAIS |