<b>Image analysis of coffee seeds submitted to the LERCAFE test

Abstract

ABSTRACT. The aim of this experiment was to quantify the stained areas of coffee seeds submitted to the LERCAFE test using image analysis. The seeds used were of the Catuaí Vermelho IAC 99 and Paradise cultivars. The physiological quality of the lots was assessed using germination tests, moisture content and a germination speed index. The LERCAFE test was conducted using seeds without parchment immersed in a solution of 3% sodium hypochlorite for 3 hours. Color photographs (RGB) with a resolution of 5 MPx were taken of each seed. The seeds were visually evaluated, and the functions generated from the analyses of the stained and non-stained regions were quantified by the Matlab R2009b program. Classification models were developed based on the Fisher Linear Discriminant Function and the evaluation of the adequacy of the models confusion matrix between the visual references and the classification generated by the linear functions. The image analysis for the creation of Fisher's linear discriminant function in the development of classifiers for the coffee seeds submitted to the LERCAFE test is potentially efficient; however, it remains necessary to test other discriminant functions and quantification methodologies.Keywords: determinants functions, sodium hypochlorite, pixels. Análise de imagens de sementes de café submetidas ao Teste LERCAFÉRESUMO. Objetivou-se com essa pesquisa quantificar as áreas coradas de sementes de café submetidas ao teste LERCAFÉ por meio da análise de imagens. Foram utilizadas sementes das cultivares Catuaí Vermelho IAC 99 e Paraíso. Foi realizada a caracterização da qualidade fisiológica dos lotes com os testes de germinação, umidade e índice de velocidade de germinação. O teste LERCAFÉ foi conduzido utilizando semente sem o pergaminho com solução de hipoclorito de sódio 3% pelo período de 3 horas. Foram feitas fotografias coloridas (RGB) com resolução de 5 MPx de cada semente. As sementes foram avaliadas visualmente e por funções geradas a partir da análise de regiões coradas e não coradas quantificadas pelo programa Matlab R2009b. Foram desenvolvidos modelos de classificação com base na Função Linear de Fisher e avaliação da adequação dos modelos pela Matriz de Confusão entre as referências visuais e classificação gerada pelas funções lineares. A análise de imagem para criação da função discriminante linear de Fisher no desenvolvimento de classificadores para sementes de café submetidas ao teste LERCAFÉ é potencialmente eficiente sendo necessário testar outras funções discriminantes e metodologias para quantificação.Palavras-chave: funções determinantes, hipoclorito de sódio, pixels.

FITTING Data Mining Settings for Ranking Seed Lots

“…Thus, a confusion matrix evaluates an algorithm's performance based on the classifier's errors and successes (Ribeiro et al, 2016). As recently performed research in the seed sector, a 10-fold cross-validation technique was used here, with a confusion matrix of the same classifier using this training method (Table 6).…”

“…In this case, errors were higher; therefore, accuracy and precision for prediction and ranking of seed lots also depend on the computational model applied, not only the training technique used as a benchmark for results. According to Ribeiro et al (2016), the machine learning process must consider the relationship between the technique used and the goal aimed at. This is because once a classifier is established, its method must be analyzed.…”

Separation of coriander seeds by Red, Green and Blue image processing

“…In order to quantify the areas that present stains on coffee seeds, it is necessary to perform the LERCAFE test, where the seeds are immersed in a sodium hypochlorite solution for a present time and temperature. As this is a visual assessment, RIBEIRO et al (2016) used image processing to make this classification through Fisher's linear discriminant function, concluding the method's potential efficiency.…”