Principais Coleções do Repositório
Submissões Recentes
Artigo
At-site ood frequency analysis in Brazil
(2023-10-16) Zegarra Rodríguez, Demóstenes; Okey, Ogobuchi Daniel; Maidin, Siti Sarah; Udo, Ekikere Umoren; Kleinschmidt, João Henrique
ecent years have witnessed an in-depth proliferation of the Internet of Things (IoT) and Industrial Internet of Things (IIoT) systems linked to Industry 4.0 technology. The increasing rate of IoT device usage is associated with rising security risks resulting from malicious network flows during data exchange between the connected devices. Various security threats have shown high adverse effects on the availability, functionality, and usability of the devices among which denial of service (DoS) and distributed denial of service (DDoS), which attempt to exhaust the capacity of the IoT network (gateway), thereby causing failure in the functionality of the system have been more pronounced. Various machine learning and deep learning algorithms have been used to propose intelligent intrusion detection systems (IDS) to mitigate the challenging effects of these network threats. One concern is that although deep learning algorithms have shown good accuracy results on tabular data, not all deep learning algorithms can perform well on tabular datasets, which happen to be the most commonly available format of datasets for machine learning tasks. Again, there is also the challenge of model explainability and feature selection, which affect model performance. In this regard, we propose a model for IDS that uses attentive mechanisms to automatically select salient features from a dataset to train the IDS model and provide explainable results, the TabNet-IDS. We implement the proposed model using the TabNet algorithm based on PyTorch which is a deep-learning framework. The results obtained show that the TabNet architecture can be used on tabular datasets for IoT security to achieve good results comparable to those of neural networks, reaching an accuracy of 97% on CIC-IDS2017, 95% on CSE-CICIDS2018 and 98% on CIC-DDoS2019 datasets.
dissertação
Associação entre caracteres agronômicos e índices de vegetação na seleção de linhagens de soja
(Universidade Federal de Lavras, 2026-07-23) De Paula, Allyson Gabriel Santos; Bruzi, Adriano Teodoro; https://orcid.org/0000-0003-2617-9101; Alves, Rodrigo Silva; https://orcid.org/0000-0002-3038-6210; Pereira, Jose Luiz de Andrade Rezende; https://orcid.org/0000-0002-6343-6931
The selection of superior soybean lines is a crucial process and involves the evaluation of multiple traits. In this context, high-throughput phenotyping represents an important tool in breeding programs. Vegetation indices constitute a promising technique for predicting grain yield and traits associated with photosynthetic efficiency. However, the adoption of these tools must result in representative and accurate information. In this sense, the repeatability coefficient is a parameter that allows the assessment of the consistency and reliability of trait estimates to be transferred to the offspring. Given the above, the objectives were: (i) to evaluate the repeatability of the vegetation indices NDVI and GLI; (ii) to investigate the potential of these indices in the multi-trait selection of superior soybean lines using the MGIDI index; and (iii) to compare selection criteria with and without the inclusion of spectral indices. Data from Value for Cultivation and Use (VCU) trials conducted in the municipalities of Ijaci, Campanha, Nazareno, and São João del-Rei, during the 2023/24 and 2024/25 growing seasons, were used. The experiments were carried out in a randomized complete block design with three replications. The following traits were evaluated: days to maturity, plant height, lodging, and grain yield. Vegetation indices were obtained from images captured by multispectral (NDVI) and RGB (GLI) cameras mounted on a Mavic drone. The experimental data were subjected to individual and joint analyses of variance. A resampling simulation was also performed to evaluate the effect of combining environments on repeatability estimates. Genotype selection was conducted using a multi-trait approach through the MGIDI index, and the agreement among different selection strategies was assessed using the coincidence index (CI). The integration of vegetation indices with agronomic traits via MGIDI is efficient for selecting superior soybean lines. NDVI shows higher repeatability than GLI. However, GLI stands out as a viable and lower-cost alternative, with high agreement with NDVI. Greater selection gains were obtained in multi-trait analyses involving multiple environments and growing seasons, highlighting that vegetation indices are reliable complementary tools in multi-environment evaluations.
Artigo
Atributos físicos de um Latossolo Vermelho-Amarelo distrófico sob diferentes usos do solo e floresta nativa
(2023-08-06) Silveira, André Ribeiro; Dias Junior, Moacir de Souza; Rocha, Wellington Willian; Leite, Ângelo Márcio Pinto; Faria, Vanessa Lopes de; Santos, Josiel de Jesus
The physical attributes of a soil are important for the sustainability of agricultural and forestry systems, making it necessary to monitor its quality. In recent years, the deterioration of soil quality has been a subject of paramount importance in soil science, as the removal of vegetation cover and the intensive use of mechanization in all cultivation operations can result in a decrease in soil quality productive capacity. In this sense, the aim of this work was to analyze the physical attributes of a dystrophic RedYellow Oxisol subjected to different land use alternatives and under native forest. Sample collection and field trials were made up of undisturbed and deformed samples from eucalyptus plantation (PE), cerrado (CE), integrated crop, livestock and forest (ILPF), conventional corn plantation (PCM), extensive pasture (PEX) and native forest (FN). Normal Proctor tests were carried out to evaluate the maximum density and optimum compaction moisture, soil resistance to penetration and basic infiltration velocity (VIB). The alternative uses CE and PEX were the ones that resulted in the highest soil densities; with the NF having the lowest density, due to the greater presence of organic matter in this environment, generating a decrease in the pressure load applied to the soil structure. The CE and PEX alternatives also presented the lowest Optimal Humidity (Uot) values, given their state of consolidation due to non-returning of the soil and trampling by animals, as well as the highest Compaction Degree (GC), indicating a greater structural alteration of the soil when compared to the FN. All systems showed values of Basic Infiltration Rate (VIB) lower than the NF, indicating that anthropic activity interfered in the soil structure.
Dados de pesquisa
Diário de campo Semear Digital
(2024) Caroline Mendonça Nogueira Paiva
This document presents an anonymized version of the field diary produced during the first technical visit to the Ingaí Agrotechnological District (DAT), Minas Gerais, Brazil, conducted in March 2024 as part of the Semear Digital project. The record contains observations on the territory’s socioeconomic and productive context, the dairy farming value chain, available infrastructure, connectivity conditions, technology use, and interactions with rural producers and institutional stakeholders. The field diary was part of the qualitative stage of the territorial diagnosis and supported the planning of research and intervention activities, as well as scientific publications related to the establishment and development of the Ingaí DAT. For public dissemination, names, contact details, properties, specific locations, and other information that could lead to the identification of participants were removed or generalized.
Artigo
Pathways to rural sustainability: opportunities and challenges in the creation of an agrotechnological district in Ingaí City, Brazil
(2024) Caroline Mendonça Nogueira Paiva; Derick David Quintino; Thacyo Bruno Custódio de Morais; Elisa Guimarães Cozadi; Jaqueline Severino da Costa; Paulo Henrique Montagnana Vicente Leme; José Roberto Soares Scolforo
An agrotechnological district (ATD) integrates sustainable agricultural practices and technologies, aiming to transform rural communities by stimulating socioeconomic development and addressing the UN Sustainable Development Goals. Ingaí, a dairy-producing municipality in Minas Gerais, Brazil, stands to benefit from the implementation of an ATD. This study aimed to identify the opportunities and challenges for the implementation of an ATD in the municipality, considering its socioeconomic characteristics, the level of producers’ technological adoption, connectivity infrastructure, support networks, and rural and market management. The method was based on an exploratory case study, using semi-structured group interviews for data collection, conducted in March 2024, with agricultural stakeholders. As a result, key challenges emerged: limited connectivity infrastructure; ineffective property management; rural family succession problems; barriers in accessing credit; resistance to the adoption of technologies; and dependence on the local government. Opportunities included the strengthening of support networks among the agents of rural productive activity; the promotion of technology adoption by producers; and creating training and rural extension programs. The adoption of these initiatives can stimulate Ingaí’s economic and social development, generate multiplier effects in its region of influence, and serve as a model for other initiatives in the state of Minas Gerais, the main dairy hub in the country, and one of the primary drivers of Brazilian agribusiness.