Aquatic products are popular among consumers, and their visual quality used to be detected manually for freshness assessment. This paper presents a solution to inspect tuna and salmon meat from digital images. The solution proposes hardware and a protocol for preprocessing images and extracting parameters from the RGB, HSV, HSI, and L*a*b* spaces of the collected images to generate the datasets. Experiments are performed using machine learning classification methods. We evaluated the AutoML models to classify the freshness levels of tuna and salmon samples through the metrics of: accuracy, receiver operating characteristic curve, precision, recall, f1-score, and confusion matrix (CM). The ensembles generated by AutoML, for both tuna and salmon, reached 100% in all metrics, noting that the method of inspection of fish freshness from image collection, through preprocessing and extraction/fitting of features showed exceptional results when datasets were subjected to the machine learning models. We emphasize how easy it is to use the proposed solution in different contexts. Computer vision and machine learning, as a nondestructive method, were viable for external quality detection of tuna and salmon meat products through its efficiency, objectiveness, consistency, and reliability due to the experiments’ high accuracy.
2 January 2024 in Revista de Gestão Social e Ambiental
Objective: The objective is to develop a Geographic Information System for Monitoring (SIGFIS) for the oversight of water resource use in the State of Pernambuco, aiming to improve the efficiency of monitoring operations and water sustainability. Literature Review: The study is based on theories and applications of Geographic Information Systems (GIS) in water resource management. Method: The methodology includes an applied and descriptive study, with data collection from sources such as APAC, IBGE, and CPRM. Documental research methods and field surveys were used, along with geoprocessing. The data were standardized and integrated into a geographic database, with the creation of thematic maps and spatial analyses. Results and Discussion: The results include an improvement in the real-time monitoring capacity of water resources, allowing for a faster response to irregularities and optimization of monitoring resources. Research Implications: Practical implications include improving the efficiency of monitoring operations and APAC's response capacity, while theoretical implications reinforce the importance of GIS in water resource management. It is expected that SIGFIS will contribute to the formulation of more robust and scientifically-based public policies. Originality/Value: This study contributes to the literature by developing a specific GIS for the monitoring of water resources in Pernambuco, demonstrating the relevance of GIS technology in promoting water sustainability and improving environmental governance.
A dengue, uma doença viral de grande impacto global, é transmitida pelo mosquito Aedes aegypti, especialmente em regiões tropicais e subtropicais onde este encontra condições ideais para reprodução. A fêmea dessa espécie, responsável pela transmissão, é um alvo crucial para o controle eficaz da disseminação da doença. Contudo, diferenciá-la de outras espécies e gêneros de mosquitos é um desafio devido às semelhanças morfológicas, sendo necessária uma precisão elevada para evitar erros nas estratégias de controle. Neste contexto, este trabalho propõe uma abordagem automatizada para identificar a fêmea do Aedes aegypti utilizando redes neurais convolucionais aplicadas a uma base de dados de imagens de alta resolução coletadas por microscopia digital. O objetivo desta pesquisa é implementar um modelo CNN robusto e preciso, capaz de identificar a fêmea do Aedes aegypti e, assim, contribuir para o monitoramento e controle dos surtos de dengue. A metodologia adotada inclui o pré-processamento das imagens com técnicas de normalização e aumento de dados, o que permite ao modelo extrair características relevantes e melhorar a acurácia. Foram configurados e testados dois modelos de CNN: uma rede baseada na MobileNetV2 e outra na VGG16, ambas adaptadas para classificar entre as classes macho e fêmea. O modelo baseado na MobileNetV2 obteve 93% de acurácia, contra 89% do modelo baseado na VGG16, mostrando como é promissor o reconhecimento do sexo do mosquito através desta abordagem.
The gross revenue indicator contributes to the understanding of the company’s situation, and generating sales revenue forecasts is a strategy that helps the manager in directing the business. This work aims to develop a set of Machine Learning (ML) models to forecast sales in physical retail. Methodology – To carry out this work, a methodology was proposed to create, compare and evaluate ML models. Findings – When analyzing the forecast scenarios, it was observed that Hourly forecasts performed better than Day forecasts. We highlight the LIGHTGBM model, which presented the best scores in the F1-score metric with 82.95%, 79.26% and 76.53% scenario representing one hour, two hours and three hours ahead, respectively. Value – It is expected that the forecast models will help managers to find insights to support the operational decisions of physical retail contributing to carry out actions to optimize companies’ processes.
Cardiovascular disorders pose a significant global health challenge, resulting in a substantial number of annual deaths. Early and accurate prediction of heart disorders is crucial to mitigate their impact on individuals and healthcare systems. In this study, we explore the potential of Convolutional Neural Network in automating heart disease prediction using spectrogram data. The dataset comprises audio recordings collected from the general public via an iPhone app and a clinical trial using a digital stethoscope. We preprocess the data to obtain spectrograms and design a Convolutional Neural Network architecture to classify heart sounds into distinct categories. The Convolutional Neural Network exhibits promising performance, achieving an accuracy of approximately 77%. Our research highlights the opportunity to leverage Convolutional Neural Network in this context, paving the way for advanced automated cardiac diagnostics. Keywords: Cardiovascular disorders,Convolutional neural network DOI: 10.7176/RHSS/13-14-02 Publication date:August 31st 2023
This study analyzes and compares the Environmental, Social, and Governance (ESG) strategies adopted by Renault and Volkswagen, focusing on how these practices are integrated into their global operations and their impact on the Sustainable Development Goals (SDGs). Renault and Volkswagen demonstrate significant commitments to reducing CO2 emissions and promoting sustainable practices. Renault focuses on expanding its electric vehicle lineup and improving energy efficiency, while Volkswagen emphasizes the goal of 100% renewable energy usage by 2025. Social initiatives include Renault's diversity and inclusion programs and Volkswagen's community development efforts. Both companies maintain strong governance systems to ensure transparency and ethics, aligning with various SDGs. This analysis highlights the different approaches each automaker takes towards achieving sustainability goals, reflecting their specific priorities and contexts within the automotive industry.