Wiliam Regone
Wiliam Regone has a total of 5 co-authors. Here are the top co-authors by number of shared publications:
- Anderson Vicente Borille
Instituto Tecnológico de Aeronáutica – ITA, Brazil
1 shared publication - André Dorigueto Canal
Instituto Tecnológico de Aeronáutica – ITA, Brazil
1 shared publication Luciel Henrique De Oliveira
Pontifícia Universidade Católica de Minas Gerais - Campus Poços de Caldas
1 shared publication- Luciel Henrique de Oliveira
PUC Minas
1 shared publication Vitoria Caprioglio Oliveira
Pontifícia Universidade Católica de Minas Gerais - Campus Poços de Caldas
1 shared publication
Latest Publications
- Open Access
- 3 January 2025
- Published by Universidade Federal de Vicosa in The Journal of Engineering and Exact Sciences
- Vol. 11 (1) , 22798
Predictive Modeling of Surface Roughness in Machining Processes Using Random Forest Regressor
Development of machine learning, specifically the Random Forest Regressor (RFR), to predict surface roughness (Ra) in machining processes. The study considered machining forces (Fx, Fy, Fz, and the resultant force F) and the main cutting parameters: depth of cut (ap), cutting speed (vc), and feed per revolution (f). The methodology involved a set of 612 experimental data points, distributed between Experiment 1 (New Tool) and Experiment 2 (Worn Tool). Statistical analyses, including Pearson correlation, were performed to assess the relationships between the variables. The model was trained and validated using statistical metrics widely applied in regression problems, along with graphical visualizations that aid in interpreting its predictive performance. The prediction algorithm achieved high coefficients of determination R², being 0.9534 for Experiment 1 and 0.9643 for Experiment 2. Furthermore, it showed low predictive errors, with MAE of 0.0406 (Exp1) and 0.0271 (Exp2), and MAPE of 6.43% (Exp1) and 4.63% (Exp2). The model's performance was evaluated using scatter plots, which demonstrated a strong alignment between the predicted values and the actual roughness values. The histograms of the residual errors showed an approximately normal distribution centered at zero, indicating a well-fitted model without significant biases. The analysis of variable importance revealed that feed per revolution (f) was the most influential factor in predicting roughness, especially in Experiment 1, where it represented 58.08% of the model's total contribution. In Experiment 2, due to tool wear, the influence of feed was reduced to 38.16%, while the machining forces, used as factors in the model, particularly the components of feed (Fy) and passive (Fz), had, collectively, a greater impact on the final roughness. The results demonstrate that RFR is an effective tool for predicting surface roughness.
- 3 January 2025
- Published by Universidade Federal de Vicosa in The Journal of Engineering and Exact Sciences
- Vol. 11 (1) , 22798
Open AccessPredictive Modeling of Surface Roughness in Machining Processes Using Random Forest Regressor
Development of machine learning, specifically the Random Forest Regressor (RFR), to predict surface roughness (Ra) in machining processes. The study considered machining forces (Fx, Fy, Fz, and the resultant force F) and the main cutting parameters: depth of cut (ap), cutting speed (vc), and feed per revolution (f). The methodology involved a set of 612 experimental data points, distributed between Experiment 1 (New Tool) and Experiment 2 (Worn Tool). Statistical analyses, including Pearson correlation, were performed to assess the relationships between the variables. The model was trained and validated using statistical metrics widely applied in regression problems, along with graphical visualizations that aid in interpreting its predictive performance. The prediction algorithm achieved high coefficients of determination R², being 0.9534 for Experiment 1 and 0.9643 for Experiment 2. Furthermore, it showed low predictive errors, with MAE of 0.0406 (Exp1) and 0.0271 (Exp2), and MAPE of 6.43% (Exp1) and 4.63% (Exp2). The model's performance was evaluated using scatter plots, which demonstrated a strong alignment between the predicted values and the actual roughness values. The histograms of the residual errors showed an approximately normal distribution centered at zero, indicating a well-fitted model without significant biases. The analysis of variable importance revealed that feed per revolution (f) was the most influential factor in predicting roughness, especially in Experiment 1, where it represented 58.08% of the model's total contribution. In Experiment 2, due to tool wear, the influence of feed was reduced to 38.16%, while the machining forces, used as factors in the model, particularly the components of feed (Fy) and passive (Fz), had, collectively, a greater impact on the final roughness. The results demonstrate that RFR is an effective tool for predicting surface roughness.
- Open Access
- 28 June 2024
- Published by Sociedade Brasileira de Computação in Revista Eletrônica de Iniciação Científica em Computação
- Vol. 22 (1) , 101-109
Aplicação da rede neural convolucional para determinar e classificar em imagens de raio-X de tórax a presença ou não do vírus da COVID-19
Imagens radiográficas, como a radiografia de tórax, são ferramentas utilizadas no processo de diagnóstico de doenças, incluindo pneumonias causadas por infecções virais. Por meio de modelos baseados em inteligência artificial, é possível desenvolver ferramentas computacionais que permitam o processo de classificação de imagens a partir de uma série de características. Com base nisso, a presente pesquisa propõe um algoritmo estruturado na forma de rede neural convolucional para codificar e treinar a classificação de imagens de raios X de tórax com vírus COVID-19 e normais. Os resultados foram obtidos usando dois modelos. O primeiro modelo usou o ResNet50 com uma precisão de 99% para imagens com COVID-19 e 92% para imagens normais. O segundo modelo usou o processamento de imagens e obteve uma precisão de 96% para imagens com COVID-19 e 92% para imagens normais.
- 28 June 2024
- Published by Sociedade Brasileira de Computação in Revista Eletrônica de Iniciação Científica em Computação
- Vol. 22 (1) , 101-109
Open AccessAplicação da rede neural convolucional para determinar e classificar em imagens de raio-X de tórax a presença ou não do vírus da COVID-19
Imagens radiográficas, como a radiografia de tórax, são ferramentas utilizadas no processo de diagnóstico de doenças, incluindo pneumonias causadas por infecções virais. Por meio de modelos baseados em inteligência artificial, é possível desenvolver ferramentas computacionais que permitam o processo de classificação de imagens a partir de uma série de características. Com base nisso, a presente pesquisa propõe um algoritmo estruturado na forma de rede neural convolucional para codificar e treinar a classificação de imagens de raios X de tórax com vírus COVID-19 e normais. Os resultados foram obtidos usando dois modelos. O primeiro modelo usou o ResNet50 com uma precisão de 99% para imagens com COVID-19 e 92% para imagens normais. O segundo modelo usou o processamento de imagens e obteve uma precisão de 96% para imagens com COVID-19 e 92% para imagens normais.
- Open Access
- 26 December 2023
- Published by A Fundacao para o Desenvolvimento de Bauru (FunDeB) in Revista Gestão da Produção, Operações e Sistemas
- Vol. 18
Harnessing Deep Learning in Computer Vision for Effective Face Mask Detection in the Global COVID-19 Crisis
Purpose: This study aims to scrutinize the integration of deep learning algorithms within the sphere of computer vision, with a concentrated focus on proficiently detecting face mask usage amidst the global COVID-19 pandemic. Theoretical Framework: The research is grounded in the theoretical underpinnings of deep learning, a branch of artificial intelligence, and its application in computer vision. It explores the advancements in machine learning algorithms capable of complex image processing and pattern recognition, essential for identifying face mask usage in various settings. Methodology/Approach: The research adopts a methodological approach involving the design and development of a deep learning model. This model is trained on a diverse dataset encompassing images of individuals with and without face masks. Python, along with libraries such as OpenCV, Keras, and TensorFlow, forms the backbone of the implementation, facilitating the processing and analysis of image data. Findings: The study's findings reveal that the developed model demonstrates a high degree of accuracy, with a 99% success rate in test image predictions, showcasing the effectiveness of deep learning in image recognition tasks. This underscores the model's proficiency in identifying face mask usage, a critical factor in controlling the spread of airborne viruses like COVID-19. Research, Practical & Social Implications: This research contributes significantly to the field of computer vision, offering practical applications in public health monitoring and societal well-being. The model's ability to accurately detect face mask usage paves the way for enhanced pandemic management strategies and reinforces the role of technology in public health initiatives. Originality/Value: This study innovates within existing research by applying deep learning in computer vision for addressing the COVID-19 crisis. It uniquely focuses on developing technological solutions for efficient and cost-effective monitoring of face mask usage, emphasizing prevention. Keywords: Computer Vision, Convolutional Neural Network, Face Mask Detection.
- 26 December 2023
- Published by A Fundacao para o Desenvolvimento de Bauru (FunDeB) in Revista Gestão da Produção, Operações e Sistemas
- Vol. 18
Open AccessHarnessing Deep Learning in Computer Vision for Effective Face Mask Detection in the Global COVID-19 Crisis
Purpose: This study aims to scrutinize the integration of deep learning algorithms within the sphere of computer vision, with a concentrated focus on proficiently detecting face mask usage amidst the global COVID-19 pandemic. Theoretical Framework: The research is grounded in the theoretical underpinnings of deep learning, a branch of artificial intelligence, and its application in computer vision. It explores the advancements in machine learning algorithms capable of complex image processing and pattern recognition, essential for identifying face mask usage in various settings. Methodology/Approach: The research adopts a methodological approach involving the design and development of a deep learning model. This model is trained on a diverse dataset encompassing images of individuals with and without face masks. Python, along with libraries such as OpenCV, Keras, and TensorFlow, forms the backbone of the implementation, facilitating the processing and analysis of image data. Findings: The study's findings reveal that the developed model demonstrates a high degree of accuracy, with a 99% success rate in test image predictions, showcasing the effectiveness of deep learning in image recognition tasks. This underscores the model's proficiency in identifying face mask usage, a critical factor in controlling the spread of airborne viruses like COVID-19. Research, Practical & Social Implications: This research contributes significantly to the field of computer vision, offering practical applications in public health monitoring and societal well-being. The model's ability to accurately detect face mask usage paves the way for enhanced pandemic management strategies and reinforces the role of technology in public health initiatives. Originality/Value: This study innovates within existing research by applying deep learning in computer vision for addressing the COVID-19 crisis. It uniquely focuses on developing technological solutions for efficient and cost-effective monitoring of face mask usage, emphasizing prevention. Keywords: Computer Vision, Convolutional Neural Network, Face Mask Detection.