RESUMEN
Wireless sensor networks (WSNs) have gained a positive popularity for structural health monitoring (SHM) applications. The underlying reason for using WSNs is the vast number of devices supporting wireless networks available these days. However, some of these devices are expensive. The main objective of this paper is to develop a cost-effective WSN based on low power consumption and long-range radios, which can perform real-time, real-scale acceleration data analyses. Since a detection system for vibration propagation is proposed in this paper, the synchronized monitoring of acceleration data is necessary. To meet this need, a Pulse Per Second (PPS) synchronization method is proposed with the help of GPS (Global Positioning System) receivers, representing an addition to the synchronization method based on real-time clock (RTC). As a result, RTC+PPS is the term used when referring to this method in this paper. In summary, the experiments presented in this research consist in performing specific and synchronized measurements on a full-scale steel I-beam. Finally, it is possible to perform measurements with a synchronization success of 100% in a total of 30 samples, thereby obtaining the propagation of vibrations in the structure under consideration by implementing the RTS+PPS method.
RESUMEN
Precision agriculture has the objective of improving agricultural yields and minimizing costs by assisting management with the use of sensors, remote sensing, and information technologies. There are several approaches to improving crop yields where remote sensing has proven to be an important methodology to determine agricultural maps to show surface differences which may be associated with many phenomena. Remote sensing utilizes a wide variety of image sensors that range from common RGB cameras to sophisticated, hyper-spectral image cameras which acquire images from outside the visible electromagnetic spectrum. The NDVI and NGBVI are computer vision vegetation index algorithms that perform operations from color masks such as red, green, and blue from RGB cameras and hyper-spectral masks such as near-infrared (NIR) to highlight surface differences in the image to detect crop anomalies. The aim of the present study was to determine the relationship of NDVI and NGBVI as plant health indicators in tomato plants (Solanum lycopersicum) treated with the beneficial bacteria Bacillus cereus-Amazcala (B. c-A) as a protective agent to cope with Clavibacter michiganensis subsp. michiganensis (Cmm) infections. The results showed that in the presence of B. c-A after infection with Cmm, NDVI and NGBVI can be used as markers of plant weight and the activation of the enzymatic activities related to plant defense induction.
RESUMEN
Recently, deep-learning techniques have become the foundations for many breakthroughs in the automated identification of plant diseases. In the agricultural sector, many recent visual-computer approaches use deep-learning models. In this approach, a novel predictive analytics methodology to identify Tobacco Mosaic Virus (TMV) and Pepper Huasteco Yellow Vein Virus (PHYVV) visual symptoms on Jalapeño pepper (Capsicum annuum L.) leaves by using image-processing and deep-learning classification models is presented. The proposed image-processing approach is based on the utilization of Normalized Red-Blue Vegetation Index (NRBVI) and Normalized Green-Blue Vegetation Index (NGBVI) as new RGB-based vegetation indices, and its subsequent Jet pallet colored version NRBVI-Jet NGBVI-Jet as pre-processing algorithms. Furthermore, four standard pre-trained deep-learning architectures, Visual Geometry Group-16 (VGG-16), Xception, Inception v3, and MobileNet v2, were implemented for classification purposes. The objective of this methodology was to find the most accurate combination of vegetation index pre-processing algorithms and pre-trained deep- learning classification models. Transfer learning was applied to fine tune the pre-trained deep- learning models and data augmentation was also applied to prevent the models from overfitting. The performance of the models was evaluated using Top-1 accuracy, precision, recall, and F1-score using test data. The results showed that the best model was an Xception-based model that uses the NGBVI dataset. This model reached an average Top-1 test accuracy of 98.3%. A complete analysis of the different vegetation index representations using models based on deep-learning architectures is presented along with the study of the learning curves of these deep-learning models during the training phase.
RESUMEN
Plant transpiration is considered one of the most important physiological functions because it constitutes the plants evolving adaptation to exchange moisture with a dry atmosphere which can dehydrate or eventually kill the plant. Due to the importance of transpiration, accurate measurement methods are required; therefore, a smart sensor that fuses five primary sensors is proposed which can measure air temperature, leaf temperature, air relative humidity, plant out relative humidity and ambient light. A field programmable gate array based unit is used to perform signal processing algorithms as average decimation and infinite impulse response filters to the primary sensor readings in order to reduce the signal noise and improve its quality. Once the primary sensor readings are filtered, transpiration dynamics such as: transpiration, stomatal conductance, leaf-air-temperature-difference and vapor pressure deficit are calculated in real time by the smart sensor. This permits the user to observe different primary and calculated measurements at the same time and the relationship between these which is very useful in precision agriculture in the detection of abnormal conditions. Finally, transpiration related stress conditions can be detected in real time because of the use of online processing and embedded communications capabilities.