The aim of this paper is to classify the land covered with oat crops, and the quantification of frost damage on oats, while plants are still in the flowering stage. The images are taken by a digital colour camera CCD-based sensor. Unsupervised classification methods are applied because the plants present different spectral signatures, depending on two main factors: illumination and the affected state. The colour space used in this application is CIELab, based on the decomposition of the colour in three channels, because it is the closest to human colour perception. The histogram of each channel is successively split into regions by thresholding. The best threshold to be applied is automatically obtained as a combination of three thresholding strategies: (a) Otsu’s method, (b) Isodata algorithm, and (c) Fuzzy thresholding. The fusion of these automatic thresholding techniques and the design of the classification strategy are some of the main findings of the paper, which allows an estimation of the damages and a prediction of the oat production.

No abstract

Abstract:One of the challenges faced by subwatershed hydrology is the discovery of patterns associated with climate and landscape variability with the available data. This study has three objectives: (1) to evaluate the annual recession curves; (2) to relate the recession parameter (RP) with physiographic characteristics of 21 Mexican subwatersheds in different climate regions; and (3) to formulate a Baseflow (BF) model based on a top-down approach. The RP was calibrated utilizing the largest magnitude curves. The RP was related to topographical, climate and soil variables. A non-linear model was employed to separate the baseflow which considers RP as a recharge rate. Our results show that RP increases with longitude and decreases with latitude. RP displayed a sustained non-linear behavior determined by precipitation rate and evapotranspiration ( P E ) over years and subwatersheds. The model was fit to a parameter concurrent with invariance and space-time symmetry conditions. The dispersion of our model was associated with the product of ( P E ) by the aquifer's transmissivity. We put forward a generalized baseflow model, which made the discrimination of baseflow from direct flow in subwatersheds possible. The proposed model involves the recharge-storage-discharge relation and could be implemented in basins where there are no suitable ground-based data.

Los métodos de detección de deficiencias de hierro (Fe) en cultivos como el frijol (Phaseolus vulgaris L.) constituyen una herramienta valiosa en la toma de decisiones porque pueden utilizarse para predecir el estado nutrimental de las plantas en etapas tempranas. En esta investigación se usaron redes neuronales bayesianas regularizadas (BRNN, por sus siglas en inglés) y árboles de clasificación para llevar a cabo la predicción de dichas deficiencias basados en lecturas del SPAD 502, el cual se empleó para medir el índice de verdor de las hojas en el frijol. Se llevó a cabo un experimento con ocho tratamientos con diferentes variaciones en la concentración de Fe (0, 20, 40, 60, 80, 100, 150 y 200 %) en la solución nutritiva. Durante siete semanas se tomaron las mediciones promedio del índice de verdor de los tres foliolos de cinco repeticiones correspondientes a los ocho tratamientos y posteriormente los datos fueron utilizados para ajustar los modelos estadísticos antes mencionados. Con las BRNN, la correlación entre valores observados y predichos fue de 0.77 para el conjunto de datos en entrenamiento y de 0.54 a 0.71 para prueba. Para el caso de los árboles de clasificación, en etapa de entrenamiento el porcentaje de clasificaciones correctas fue 56.25 % y disminuyó casi 30 % cuando se llevó a cabo el procedimiento de validación. Por lo que para la presente investigación, el uso de BRNN constituye una herramienta valiosa para la predicción de deficiencias tempranas de Fe en el cultivo de frijol.

The information about where crops are distributed is useful for agri-environmental assessments, but is chiefly important for food security and agricultural policy managers. The quickness with which this information becomes available, especially over large areas, is important for decision makers. Methodologies have been proposed for the study of crops. Most of them require field survey for ground truth data and a single crop map is generated for the whole season at the end of the crop cycle and for the next crop cycle a new field survey is necessary. Here, we present models for recognizing maize (Zea mays L.), beans (Phaseolus vulgaris L.), and alfalfa (Medicago sativa L.) before the crop cycle ends without current-year field survey for ground truth data. The models were trained with an exhaustive field survey at plot level in a previous crop cycle. The field surveys begin since days before the emergence of crops to maturity. The algorithms used for classification were support vector machine (SVM) and bagged tree (BT), and the spectral information captured in the visible, red-edge, near infrared, and shortwave infrared regions bands of Sentinel 2 images was used. The models were validated within the next crop cycle each fifteen days before the mid-season. The overall accuracies range from 71.9% (38 days after the begin of cycle) to 87.5% (81 days after the begin cycle) and a kappa coefficient ranging from 0.53 at the beginning to 0.74 at mid-season