Forest fires are still a large concern in several countries due to the social, environmental and economic damages caused. This paper aims to show the design and validation of a proposed system for the classification of smoke columns with object detection and a deep learning-based approach. This approach is able to detect smoke columns visible below or above the horizon. During the dataset labelling, the smoke object was divided into three different classes, depending on its distance to the horizon, a cloud object was also added, along with images without annotations. A comparison between the use of RetinaNet and Faster R-CNN was also performed. Using an independent test set, an F1-score around 80%, a G-mean around 80% and a detection rate around 90% were achieved by the two best models: both were trained with the dataset labelled with three different smoke classes and with augmentation; Faster R-CNNN was the model architecture, re-trained during the same iterations but following different learning rate schedules. Finally, these models were tested in 24 smoke sequences of the public HPWREN dataset, with 6.3 min as the average time elapsed from the start of the fire compared to the first detection of a smoke column.
The sitting position has become one of the most common postures in developed countries. However, assuming a poor sitting posture leads to several health problems, namely back, shoulder and neck pain. In a previous work, an intelligent chair was developed and was shown to classify and correct the seating position. This work describes improvements on this intelligent chair prototype culminating with the development of a new prototype. The improvements of this new prototype are presented, resulting in new studies for posture identification. Pressure maps for 12 sitting postures were gathered in order to automatically detect user's posture through a neural network algorithm, obtaining an overall posture classification of around 81%.
Background and Aims
The evaluation of genotype‐by‐environment interaction and the genotypic correlations between important economic traits are two relevant issues of the methodology of grapevine selection that remain insufficiently explored. The aim of this study is to provide methodological tools to: (i) assess the genotype‐by‐environment interaction and (ii) evaluate the genetic correlations between traits and their practical impacts on genetic selection.
Methods and Results
A multi‐environment analysis for each trait and multi‐trait analyses were applied for the evaluation of genotype‐by‐environment interaction and genetic correlation between pairs of traits, respectively. The presence of genotype‐by‐environment interaction for all traits was detected. Genetic correlation between traits varied from non‐existent to strong.
Conclusions
This study supports and recommends the applicability of multi‐environment and multi‐trait analyses to grapevine clonal selection data. The detection of genotype‐by‐environment interaction in grapevine clones was successfully assessed. When multi‐trait models were applied they provided greater accuracy and precision in selection.
Significance of the Study
This study shows the importance of the mixed multivariate models for the development of the grapevine clonal selection. They should be implemented in a grapevine selection program.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.