Francisco de Assis de Carvalho Pinto

Definition of management zones in coffee production fields based on apparent soil electrical conductivity

Fertilizer application at variable rates requires dense sampling to determine the resulting field spatial variability. Defining management zones is a technique that facilitates the variable-rate application of agricultural inputs. The apparent electrical conductivity of the soil is an important factor in explaining the variability of soil physical-chemical properties. Thus, the objective of this study was to define management zones for coffee (Coffea Arabica L.) production fields based on spatial variability of the apparent electrical conductivity of the soil. The resistivity method was used to measure the apparent soil electrical conductivity. Soil samples were collected to measure the chemical and physical soil properties. The maps of spatial variability were generated using ordinary kriging method. The fuzzy k-means algorithm was used to delimit the management zones. To analyze the agreement between the management zones and the soil properties, the kappa coefficients were calculated. The best results were obtained for the management zones defined using the apparent electrical conductivity of the soil and the digital elevation model. In this case, the kappa coefficient was 0.45 for potassium, which is an element that is associated with quality coffee. The other variable that had a high kappa coefficient was remaining phosphorous; the coefficient obtained was 0.49. The remaining phosphorus is an important parameter for determining which fertilizers and soil types to study.

The relationship between apparent soil electrical conductivity and soil properties

Precision agriculture that is based on the physical and chemical properties of soil requires a dense sampling to evaluate spatial variability in the field. This dense sampling is often expensive and time consuming. One technique to reduce the number of samples is to define management zones based on information that is collected in the field. Some researchers have demonstrated the importance of the electrical properties of soil in defining management zones. Thus, the objective of this study was to evaluate the relationship between the apparent soil electrical conductivity and soil properties in mountainous areas of coffee production. The electrical conductivity of soil was evaluated at soil depths ranging from 0.00-0.20 m (EC20) and 0.00-0.40 m (EC40) using a portable meter. The mean values of EC20 and EC40 were 1.80 mS m-1 and 1.22 mS m-1, respectively. Both EC20 and EC40 exhibited comparatively low correlations with the soil properties, whereas higher correlations were obtained for measurements of remaining phosphorus, wherein values of 0.427 and 0.465, respectively, were obtained.

Efeito da frequência e amplitude de vibração sobre a derriça de frutos de café

Quality parameters influence directly the coffee price. However, selective coffee harvesting is frequently associated to good quality of this product. This procedure can be performed by mechanical vibration. Therefore, the study of the frequency and amplitude parameters is important for the design of a specific harvesting machine. The objective of this work was to evaluate the effect of the frequency and amplitude of vibration, the coffee variety and the ripeness condition of the fruits upon the harvesting efficiency. The vibration tests were done in laboratory using an electromagnetic shaker. The tests were done using amplitudes in the range of 3.75 to 7.50 mm and frequencies in the range of 13.33 to 16.67 Hz. Branches of coffee plants of Mundo Novo variety were tested. The highest harvesting efficiency was obtained when using the 26.67 Hz frequency of vibration. The highest harvesting efficiency was obtained when an amplitude of 7.5 mm was used. It was also observed that the number of fruits per stem influences the harvesting efficiency of the coffee fruits of the Mundo Novo variety.

Análise discriminante paramétrica para reconhecimento de defeitos em tábuas de eucalipto utilizando imagens digitais

The lumber industry has given special attention for lumber grading and selection stages. Machine Vision Systems have been proposed as a technological solution for automation of these stages. The proper feature selection for discriminating defect and clear wood is one of the most challenging in the development of such technology. The objective of this work was to evaluate, using multivariate analysis, the discriminating power of color images percents. In this work, linear and quadratic discriminant analysis were accomplished for classification of defects and clear wood in digital images of eucalyptus lumber. The percent features of the histogram for the red, green and blue bands, from two sizes of image blocks were used for developing and testing the discriminant functions. 492 blocks were used, containing the 12 studied defects and clear wood, derived from images of 40 lumbers randomly sampled. The features were analyzed with their original values, scores of the principal components and scores of the canonical variables. The smallest global misclassification errors were 19% and 24% for linear discriminant function with the canonical variable scores using block sizes of 64x64 and 32x32 pixels, respectively. The percent features were considered appropriate to discriminate defects and clear wood in digital images.

Algoritmo para classificação de plantas de milho atacadas pela lagarta do cartucho (Spodoptera frugiperda, Smith) em imagens digitais

An image processing and analysis algorithm was developed to identify the fall armyworm damage on corn plants. The developed program segmented the larvae damage on the image in two stages: a coarse and fine classification. The coarse stage applied a threshold technique on image blocks of 60 x 60 pixels. The fine stage was based on a neural network classifier which classifies image blocks of 3 x 3 pixels. The algorithm accuracy was accessed by evaluating the error matrix based on 80 and 75 image blocks of the coarse and fine stages, respectively. The algorithm presented an overall accuracy of 80.74%.

Determination of nitrogen and chlorophyll levels in bean-plant leaves by using spectral vegetation bands and indices

Objetivou-se desenvolver classificadores com base em diferentes combinacoes de bandas e indices de vegetacao espectrais de imagens originais, segmentadas e reflectâncias, para discriminacao de teores de nitrogenio e clorofila foliares do feijoeiro, definindo a melhor epoca e as melhores variaveis. Foi utilizado um sistema de sensoriamento remoto constituido por um balao a gas helio e duas câmeras digitais de pequeno formato. Alem das bandas isoladamente, foram testados quatro indices de vegetacao: da razao simples, da diferenca normalizada, da diferenca normalizada utilizando a banda do verde e o da absorcao de clorofila modificado. Os classificadores demonstraram serem eficientes na discriminacao de teores de nitrogenio e clorofila foliares. A melhor epoca para discriminar teor de nitrogenio foliar foi aos 13 DAE (estadio V4). Os melhores classificadores para esta epoca utilizaram como entrada dois indices em imagens reflectância segmentada, um indice relacionado com a estrutura do dossel e outro relacionado com a clorofila, com Kappa variando entre 0,26 a 0,31. Para discriminar teor de clorofila foliar, a melhor epoca foi aos 21 DAE (estadio V4). O melhor classificador utilizou como entrada duas imagens originais, uma da banda vermelha e outra da banda azul, com Kappa de 0,47.

Vegetation spectral indices to discriminate nitrogen rates in tanzania grass

The objective of this study was to determine, among five vegetation indices, calculated based on spectral reflectance data, the one that best discriminated nitrogen rates, and presented highest correlation with chlorophyll readings and tanzania grass (Panicum maximum Jacq.) dry matter. The tested vegetation indices were NDVI (normalized difference vegetation index), VARI (visible atmospherically resistant index using red edge and green bands), WDRVI (wide dynamic range vegetation index using three weighted coefficients, 0.05, 0.1 and 0.2). Four nitrogen levels (0, 80, 160 and 320 kg/ha) were evaluated in a randomized complete block design with three replications and three subsamplings per block. The indices VARI, using red edge band, and WDRVI, using weighted coefficients 0.05 and 0.1, were the best indices to discriminate nitrogen rates. The index that presented the highest correlation with chlorophyll readings and dry matter was the VARI using red edge band.

Determinação da cobertura do solo por análise de imagens e redes neurais

An image classification algorithm was developed to estimate the soil cover based on artificial neural networks (ANN) trained by back-propagation algorithm. The learning data sets were obtained from digital normalized images. Five ANN architectures of the type 25-n1-n2-2 were tested. The architecture 25-20-10-2 presented the best result and therefore, it was used in the image classification program. The classification presented an overall accuracy of 82.10%. This result shows that ANN may be applied for separating features when the pixel brightness does not provide enough information to apply the threshold technique.

Detecção do efeito da adubação nitrogenada em Brachiaria decumbens Stapf. utilizando um sistema de sensoriamento remoto

The objective of the present work was the detection of different nutritional statuses in Brachiaria decumbens pasture using remote sensing techniques. The area was treated with five rates of nitrogen fertilizer (0, 50, 100, 150 and, 200 kg ha-1) with six repetitions and evaluated in a completely randomized statistical design. A remote sensing system composed of digital cameras, cables, a framegrabber and a computer was used with a three meter metallic support to position the cameras. The system acquired images in two spectral bands simultaneously in two phases. The first phase occurred from February to March 2006 at 15, 21 and, 32 days after fertilization and the second from March to May of 2006 at 28, 36, 45 and, 53 days after reapplication of the same N rates. Vegetation indices were evaluated from the original images, and the data was submitted to regression and correlation analyses. Estimate values of chlorophyll content using the chlorophyll meter SPAD 502 and values of the leaf N content were also acquired. First or second degree models were adjusted to the experimental data for all periods. The indices using the green band proved more efficient to detect the relationship with the estimated chlorophyll values, the leaf N content and the dry mass yield than the red band in all studied periods. Thus, the used remote sensing system technique allowed for the identification of different effects of nitrogen fertilization in the forage.

Spatial and Temporal Variability of Coffee Quality

The objective of this work was to analyze the coffee quality by using the spatial correlation. The work was done in the 2004 and 2005 harvesting seasons at Brauna Farm, located in Araponga, Minas Gerais state, Brazil. Coffee quality maps were generated. From the quality maps, the spatial autocorrelation analyses were performed by using the Moran Index. The results obtained indicated that the coffee quality has spatial dependence.