ABSTRACT Phosphorus concentration is one of the main attributes determined in laboratory analyses of soil samples collected in the field. The objective is to develop a soil phosphorus test using a low-cost spectrophotometer and a machine learning technique. For reflectance measurements, a low-cost system consisting of a Sparkfun AS7625x spectrophotometer and an Arduino Uno is used. Ion exchange resins under standard saturated solutions and modified conditions are used to extract phosphorus ions from the soil samples. Reflectance and phosphorus concentrations determined by the reference method are used in the training and testing of a machine learning. A modification procedure of the ion-exchange resin saturation solution allows the establishment of a strong correlation between the reflectance in 18 spectral bands and P concentration of the soil samples. The obtained model uses five reflectance of the modified resins at wavelengths of 410, 460, 560, 705, and 645 nm to predict the phosphorus concentration. This model presents an R2t accuracy of 0.97 in the training stage with an R2v of 0.96, RMSEv of 9.05, and ratio of prediction to deviation) of 3.81 in the test step.

RESUMO Os índices de vegetação são importantes indicadores da saúde e produtividade das culturas agrícolas. Dentre os tipos de sensores, os sensores proximais podem ser utilizados para tomada de decisão em tempo real. Dessa forma, o objetivo deste estudo foi desenvolver um sistema sensor proximal, com base em fototransistor, para obtenção e armazenamento dos índices de vegetação: NDVI (Índice de Vegetação da Diferença Normalizada), SR (Razão Simples), WDRVI (Índice de Vegetação de Larga Escala Dinâmica), SAVI (Índice de Vegetação Ajustado ao Solo) e OSAVI (Índice de Vegetação Ajustado ao Solo Otimizado). O sistema sensor foi desenvolvido utilizando um circuito analógico para realizar a aquisição de dados de reflectância das bandas do vermelho e infravermelho próximo. O sistema sensor foi calibrado, utilizando grama Zoysia japonica, para ajustar seus valores aos de um espectroradiômetro. Desenvolveu-se um algoritmo que calcula os índices de vegetação e armazena os resultados em um arquivo. Avaliou-se a correlação de Pearson entre os índices de vegetação obtidos com o sistema sensor e o espectroradiômetro. Os índices de vegetação apresentaram correlação de Pearson superiores a 0,92 com os valores estimados pelo espectroradiômetro. Nas condições analisadas, o sistema sensor proposto pode ser utilizado na determinação de todos os índices de vegetação abordados.

ABSTRACT Vegetation indexes are important indicators of the health and yield of agricultural crops. Among the sensors used to evaluate vegetation indexes, proximal sensors can be used for real-time decision-making. Thus, the objective of this study was to develop a proximal sensor system based on phototransistors to acquire and store the following vegetation indexes: normalized difference vegetation index, simple ratio, wide dynamic range vegetation index, soil-adjusted vegetation index, and optimized soil-adjusted vegetation index. The sensor system was developed using an analog circuit to acquire reflectance data from red and near-infrared bands. The sensor system was calibrated according to the results of a spectroradiometer, using Zoysia japonica grass as the target. An algorithm that calculates and stores vegetation indexes in a file was developed. The Pearson correlation between the vegetation indexes obtained with the sensor system and the spectroradiometer was evaluated. The vegetation indexes presented a Pearson correlation higher than 0.92 to the estimated values by the spectroradiometer. Under the evaluation conditions, the proposed sensor system could be used to determine all vegetation indexes evaluated.

ABSTRACT There exist two options for digital cameras that can capture the near-infrared (NIR) band. Conventional red–green–blue (RGB, visible bands) cameras with a single sensor provide NIR band visibility based on the removal of the internal NIR-blocking filter. Alternatively, multisensor cameras exist that have a specific sensor for each band. The modified RGB cameras are of a lower price. In this context, the objective of this study was to compare the performance of a modified RGB camera with that of a multisensor camera for obtaining the normalized difference vegetation index (NDVI) in an area with coffee cultivations. A multispectral camera with five sensors and another camera with only one sensor were used. The NDVI of the coffee field was also measured using the GreenSeeker handheld NDVI sensor manufactured by Trimble. The images were calibrated radiometrically based on the targets in shades of gray made of napa, and the NDVI was calculated after image calibration. The calibration curves showed a high coefficient of determination. The NDVI value obtained with the calibrated images from the cameras showed a significant correlation with the values obtained by the GreenSeeker NDVI sensor, making it possible to obtain the variability pattern of the vegetation index. However, the NDVI obtained using the multisensor camera was closer to the NDVI obtained by the GreenSeeker NDVI sensor.

ABSTRACT The delimitation of management classes is critical for successful precision agriculture. This process involves choosing the variables to use and analyzing the spatial variability of the variables. Thus, the objective of this study was to analyze the correlation between management class maps generated from orbital images and yield maps. A 95-hectare area of rainfed grain was evaluated. Yield maps were obtained for the 2015/2016 and 2016/2017 soybean crops. Orbital images were used from two dates for each crop to generate vegetation index maps. The spatial correlation between the vegetation indices and the yield maps was obtained using a bivariate Moran index. The delineated management classes were compared using the Kappa index. This study demonstrated that the Kappa values resulting from the comparison between the management class maps generated from the soybean yield and the vegetation index ranged from 5% to 67% depending on the number of delineated classes. The highest Kappa values were observed when the area was delineated into three classes.

ABSTRACT The global navigation satellite system (GNSS) is the basis for localized crop management by allowing the georeferencing of collected data and the generation of maps by different systems that compose precision agriculture. There is a demand for low-cost navigation systems to enable their use in agriculture. Therefore, the objective of this study is to integrate a low-cost GNSS module to a single-board computer using Kalman filtering to obtain navigation data. The system was evaluated by performing one static and two kinematic experiments, with three repetitions each. In the static experiment, the mean error was 3.25 m with a root mean square error (RMSE) of 3.73 m. In the first kinematic experiment, data variability was lower at a velocity of 1.39 m s−1. In the second kinematic experiment, the mean error was 1.26 and 1.13 m, and the RMSE was 1.45 and 1.27 m for data obtained before and after filtering, respectively. In conclusion, the system reduces the lateral errors in linear sections but is not indicated for sections that change direction. Moreover, this system can be used in agricultural applications such as soil sampling and crop yield monitoring.

A agricultura de precisão baseada nas propriedades físicas e químicas do solo exige uma amostragem densa para se determinar a variabilidade espacial no campo. Essa amostragem densa, muitas vezes, apresenta custo e tempo consumido elevados. Uma das técnicas para reduzir o número de amostras é definir zonas de manejo em função de informações coletadas no campo. Alguns pesquisadores têm demonstrado a importância desempenhada pelas variáveis elétricas do solo para definir zonas de manejo. Dessa forma, este trabalho teve como objetivo avaliar a relação entre a variabilidade espacial da condutividade elétrica aparente e as propriedades do solo em regiões de produção de cafés de montanha. Os mapas de variabilidade espacial foram gerados utilizando-se de técnicas de geoestatística. Com base nos resultados espacializados, foi realizada uma análise de correlação, utilizando-se do índice de Moran bivariado, entre o mapa de condutividade elétrica e as propriedades do solo. Observou-se que os mapas de potássio (K) e fósforo remanescente (P-rem) foram os que mais se aproximaram do padrão de variabilidade espacial da condutividade elétrica aparente do solo.

Precision agriculture based on the physical and chemical properties of soil requires dense sampling to determine the spatial variability of these properties. This dense sampling is often expensive and time-consuming. One technique used to reduce sample numbers involves defining management zones based on information collected in the field. Some researchers have demonstrated the importance of soil electrical variables in defining management zones. The objective of this study was to evaluate the relationship between the spatial variability of the apparent electrical conductivity and the soil properties in the coffee production of mountain regions. Spatial variability maps were generated using a geostatistical method. Based on the spatial variability results, a correlation analysis, using bivariate Moran's index, was done to evaluate the relationship between the apparent electrical conductivity and soil properties. The maps of potassium (K) and remaining phosphorus (P-rem) were the closest to the spatial variability pattern of the apparent electrical conductivity.

Vários trabalhos de análise de custos em unidades armazenadoras de produtos agrícolas foram conduzidos; no entanto, poucos abordam as análises dos efeitos dos fatores associados a cada uma das etapas do pré-processamento e armazenagem. Desse modo, o objetivo deste trabalho foi desenvolver um sistema de apoio à decisão para determinação de custos e tarifas em unidades armazenadoras de produtos agrícolas. Uma unidade da CONAB, localizada em Ponta Grossa - PR, foi usada para geração de resultados conforme suas características específicas, tais como quantidade de produto recebido e estocado ao longo do ano, capacidade horária de secagem, limpeza, recepção e expedição. Ao empregar o sistema de apoio à decisão, foi constatado que os custos de recepção e expedição diminuem exponencialmente com o aumento do índice de rotatividade da unidade armazenadora; que os custos de limpeza e secagem foram aproximadamente lineares e crescentes com o aumento do teor de água inicial do produto, e que o custo de armazenagem apresentou comportamento exponencial crescente com a redução do índice de ocupação da unidade armazenadora.

Many research works have being carried out on analyzing grain storage facility costs; however a few of them had taken into account the analysis of factors associated to all pre-processing and storage steps. The objective of this work was to develop a decision support system for determining the grain storage facility costs and utilization fees in grain storage facilities. The data of a CONAB storage facility located in Ponta Grossa - PR, Brazil, was used as input of the system developed to analyze its specific characteristics, such as amount of product received and stored throughout the year, hourly capacity of drying, cleaning, and receiving, and dispatch. By applying the decision support system, it was observed that the reception and expedition costs were exponentially reduced as the turnover rate of the storage increased. The cleaning and drying costs increased linearly with grain initial moisture. The storage cost increased exponentially as the occupancy rate of the storage facility decreased.