Search

Nowadays, diseases are one of the major threats to sustainable viticulture. Manual detection through visual surveys, usually done by agronomists, relies on symptom identification and requires an enormous amount of time. Detection in field conditions remains difficult due to the lack of infrastructure to perform detailed and rapid field scouting covering the whole vineyard. In general, symptoms of grapevine diseases can be seen as spots and patterns on leaves. In this sense, computer vision technologies and artificial intelligence (AI) provide an excellent alternative to improve the current disease detection and quantification techniques using images of leaves and canopy. These novel methods can minimize the time spent on symptom detection, which helps in the control and quantification of the disease severity. In this article, we present some results of deep learning-based approaches used for detecting automatically leaves with downy mildew symptoms from RGB images acquired under laboratory and field conditions. The results obtained so far with AI approaches for detecting leaves with downy mildew symptoms are promising, and they put in evidence of the huge potential of these techniques for practical applications in the context of modern and sustainable viticulture.

Resumen La estimación del rendimiento es muy importante para la industria vitivinícola, ya que proporciona información útil para la gestión de viñedos y bodegas. Los efectos del cambio climático, como el aumento de las temperaturas y la menor disponibilidad de agua, pueden afectar a los componentes del rendimiento del viñedo. En general, la previsión tradicional del rendimiento se basaba en el recuento manual y destructivo de los racimos y el peso de las bayas. Los métodos convencionales no proporcionan una estimación precisa y son lentos, caros y laboriosos. En este estudio, se utilizaron métodos novedosos que emplean tecnologías digitales como el uso de la detección próxima, la visión por ordenador y la inteligencia artificial para la estimación del rendimiento en viñedos comerciales. La visión por ordenador se utilizó para la detección automática de diferentes características del dosel y para la calibración de ecuaciones de regresión para la predicción del rendimiento por cepa. La inteligencia artificial se utilizó para el recuento automático de racimos. Los resultados mostraron que el algoritmo de aprendizaje profundo fue capaz de detectar racimos con una alta precisión. En conclusión, nuestros resultados demostraron la aplicabilidad de estos nuevos métodos para evaluar los componentes del rendimiento en viñedos comerciales.

Ocean energy sources are a promising source of energy. However, simulating a hydrokinetic farm with multiple units requires significant computational resources, while physical experimentation on site is expensive. Therefore, the scientific challenge is to develop analytical and experimental tools that consider real aspects of areas with generation potential in a controlled laboratory environment. This paper presents a theoretical and experimental tool for analysing the interconnection of a hydrokinetic energy farm comprising 20 generation units. The test bench is a Power Hardware in the Loop type, consisting of one physical prototype generator to scale and 19 discrete averaged models operating in real-time. The system allows generators to interact through an amplifier, emulating the impact of power injection in a small electrical network. This is based on the variability of the marine resource, specifically the current velocities in the Cozumel-Mexico channel. Unlike other publications, the most significant contribution of this work is a complete feasible emulation of a marine current plant interconnected to an electrical grid, where the objective is to have a global analysis of the operation of each generation unit and the impact of the interconnection as a whole, considering that such information is of utmost importance for the execution of future projects of power generation from the sea. [ABSTRACT FROM AUTHOR]

The evapotranspiration ( E T ) process is an essential component in many agricultural water management systems, and its estimation is even more determinant when crops are grown under water-limited environments. The traditional canopy resistance ( r c ) approaches were evaluated to simulate potato evapotranspiration ( E T c p ) using the original Penman–Monteith equation under different irrigation levels. A field study was carried out on a drip-irrigated potato crop (var. Puyehue INIA) located in the Research Center Carillanca (INIA), La Araucanía Region, Chile (38°41′ S, 72°24′ W, 188 m above sea level) during the 2018/2019 and 2019/2020 growing seasons. The different irrigation levels were full irrigation ( I L 1 ), 75% of I L 1 ( I L 2 ), and 60% of I L 1 ( I L 3 ). The soil water content, morphological, physiological, meteorological, and micrometeorological variables were measured to calculate the different r c approaches and estimate ET for both growing evaluated seasons. The final values of estimated E T c p were compared to the soil water balance method ( E T c p W B ). The use of amphistomatous (LA) and hypostomatous (LH) r c approaches are the best alternative to estimate the E T c p on potato crops. The best estimation of ET was found for E T c p L A with an overestimation of 0.6% for I L 1 , 7.0% for I L 2 , and 13.0% for I L 3 , while for E T c p L H with underestimations of 12.0, 11.0 and 31.0% for I L 1 , I L 2 , and I L 3 , respectively. The lowest average values of root mean square error (RMSE), mean absolute error (MAE), and index of agreement (d) were observed for E T c p L A in both I L 1 and I L 2 conditions, with values of 4.4 and 3.2 mm, 3.2 and 2.5 mm, and 0.82 and 0.87, respectively. More investigation is necessary on the plasticity of the morphological features of potato leaves and canopy geometry, as the stomatal water vapor flowing on the canopy surface could be affected, which is a key factor in the canopy resistance model for accurate E T estimation under soil-water-limited conditions. [ABSTRACT FROM AUTHOR]

This article revisits the idea of parallelism between the two semiotic planes – expression and content – focusing on two seemingly contradictory analogies found in the structural literature, namely expression figuræ : sign expressions :: content figuræ : sign contents and phonic features : phonemes :: semantic features : signifieds. It is shown that, though it may not seem to be the case at first glance, both analogies describe the same types of relations and can be regarded as equally valid from the point of view of their respective theoretical frameworks and levels of analysis. However, a functional synthesis of the two views ultimately reveals an asymmetry in the categorical structure of the two planes, a situation that is explained by the nature of the relationship between expression and content. [ABSTRACT FROM AUTHOR]

The determination of bunch features that are relevant for bunch weight estimation is an important step in automatic vineyard yield estimation using image analysis. The conversion of 2D image features into mass can be highly dependent on grapevine cultivar, as the bunch morphology varies greatly. This paper aims to explore the relationships between bunch weight and bunch features obtained from image analysis considering a multicultivar approach. A set of 192 bunches from four cultivars, collected at sites located in Portugal and South Africa, were imaged using a conventional digital RGB camera, followed by image analysis, where several bunch features were extracted, along with physical measurements performed in laboratory conditions. Image data features were explored as predictors of bunch weight, individually and in a multiple stepwise regression analysis, which were then tested on 37% of the data. The results show that the variables bunch area and visible berries are good predictors of bunch weight (R2 ranging from 0.72 to 0.90); however, the simple regression lines fitted between these predictors and the response variable presented significantly different slopes among cultivars, indicating cultivar dependency. The elected multiple regression model used a combination of four variables: bunch area, bunch perimeter, visible berry number, and average berry area. The regression analysis between the actual and estimated bunch weight yielded a R2 = 0.91 on the test set. Our results are an important step towards automatic yield estimation in the vineyard, as they increase the possibility of applying image-based approaches using a generalized model, independent of the cultivar. [ABSTRACT FROM AUTHOR]

Table grape browning is a complex physiological disorder that occurs during cold storage. There is a need to investigate novel and innovative ways to manage the problem that hampers the progressive and sustainable growth of table grape industries. Given the complex nature of the browning phenomenon, techniques such as near-infrared (NIR) spectroscopy can be utilized for the non-destructive classification of different browning phenotypes. In this study, NIR coupled with partial least squares discriminant analysis (PLS-DA) and artificial neural networks (ANN) were used to classify bunches as either clear or as having chocolate browning and friction browning based on the spectra obtained from intact 'Regal Seedless' table grape bunches that were cold-stored over different periods. Friction browning appears as circular spots close to the pedicel area that are formed when table grape berries move against each other, and chocolate browning appears as discoloration, which originates mostly from the stylar-end of the berry, although the whole berry may appear brown in severe instances. The evaluation of the models constructed using PLS-DA was done using the classification error rate (CER), specificity, and sensitivity and for the models constructed using ANN, the kappa score was used. The CER for chocolate browning (25%) was better than that of friction browning (46%) for weeks 3 and 4 for both class 0 (absence of browning) and class 1 (presence of browning). Both the specificity and sensitivity of class 0 and class 1 for friction browning were not as good as that of chocolate browning. With ANN, the kappa score was tested to classify table grape bunches as clear or having chocolate browning or friction browning and showed that chocolate browning could be classified with a strong agreement during weeks 3 and 4 and weeks 5 and 6 and that friction browning could be classified with a moderate agreement during weeks 3 and 4. These results open up new possibilities for the development of quality checks of packed table grape bunches before export. This has a significant impact on the table grape industry for it will now be possible to evaluate bunches non-destructively during packaging to determine the possibility of these browning types being present when reaching the export market. [ABSTRACT FROM AUTHOR]