Engenharia de Controle e Automação
URI permanente desta comunidadehttps://ri.ifam.edu.br/handle/4321/970
Navegar
9 resultados
Resultados da Pesquisa
Trabalho de Conclusão de Curso Métodos para reconhecimento de resíduos recicláveis através de visão computacional(2025-02-10) Silva, Vitor Arlinson Rodrigues da; Pereira, Micila Sumária Medeiros; http://lattes.cnpq.br/7877913821937987; Santos, Alyson de Jesus dos; http://lattes.cnpq.br/5998752909180697; Fialho, Edevaldo Albuquerque; http://lattes.cnpq.br/1681351413618686; Santos, Renan Cavalcante; http://lattes.cnpq.br/6930748017205035This work proposes the study of methods that seek to unite concepts from society 5.0 and industry 4.0 to be used in favor of sustainability. With the increase in the production of plastic waste around the world and its low recycling rate, there was a need to reduce the impact on the ecosystem. The methodology used was applied, explanatory research with a technological design. As a result, three methods were developed for recognizing recyclable waste. Finally, the method selected used a convolutional neural network to create and train a classifier for a machine, powered by plastic and metal lids. The aim of this system is to promote sustainability and reduce the amount of recyclable materials disposed of inappropriately in the environment. Therefore, this work aims to develop low-cost sorters and select the best one to be used in an machine powered by recyclable waste.Trabalho de Conclusão de Curso Uso do drone Tello Edu como ferramenta integrada para o ensino de programação, visão computacional e inteligência computacional(2025-03-13) Alves Júnior, Washington Henrique; Frota, Vítor Bremgartner da; http://lattes.cnpq.br/6100146230873494; Velozo, Hugo Alves; http://lattes.cnpq.br/8351107136518878; Santos, Alyson de Jesus dos; http://lattes.cnpq.br/5998752909180697This study addresses the utilization of the Tello EDU drone as an integrated educational tool for teaching programming, computer vision, and computational intelligence at the Instituto Federal do Amazonas (IFAM), Manaus Distrito Industrial campus. The general aim is to demonstrate the educational potential of the Tello EDU drone, highlighting its contribution to developing computational thinking and facilitating the practical application of related disciplinary content. The adopted methodology comprises a literature review and a critical analysis of data and previous experiments. Initially, the research focused on visual programming through Scratch, revealing significant student engagement and interest due to the ease and interactivity of the activities. Subsequently, the project introduced Python as a textual programming language, enabling students to execute more complex tasks, such as autonomous flights and advanced command sequences using the DJITelloPy library. Additionally, the study explored practical applications using OpenCV and MediaPipe libraries to implement gesture-based control systems. Results from prior experiments confirmed the viability of this integrated approach, providing students with an intuitive and dynamic learning interface, promoting practical and meaningful learning experiences in programming, computer vision, and computational intelligence. In conclusion, the Tello EDU drone proved to be an effective resource for fostering computational thinking and enhancing practical understanding and application in programming, computer vision, and computational intelligence. This approach significantly contributed to interactive and meaningful learning experiences.Trabalho de Conclusão de Curso Uso de visão computacional para detecção de quedas em tempo real(2025-01-07) Chung, Minjae Lins; Santos, Alyson de Jesus dos; http://lattes.cnpq.br/5998752909180697; Santos, Alyson de Jesus dos; http://lattes.cnpq.br/5998752909180697; Rodrigues, Marlos André Silva; http://lattes.cnpq.br/0650250324042531; Ribeiro, Ewerton Andrey Godinho; http://lattes.cnpq.br/0187470129136467The aging of the global population brings numerous challenges, both for society as a whole and on a personal/family level. Families face the dilemma of protecting elderly family members without compromising their autonomy and independence during this vulnerable phase of their lives. Along with the physical and psychological changes involved in the natural aging process, domestic accidents, particularly falls, become common events for most of the elderly population. In this context, a fall detection system was presented, employing pose recognition (poseNet) and action recognition (actionNet) technologies, capable of automatically sending alerts to relatives or caregivers. The results demonstrated an accuracy of 82.5% in the proposed model, with near-instantaneous time between alert delivery and receivement. Thus, the system proved to be efficient, meeting the outlined objectives and serving as a foundation for future projects.Trabalho de Conclusão de Curso Uso de redes neurais para classificação de resíduos sólidos(2024-11-26) Maciel, Kainy Medeiros; Santos, Alyson de Jesus dos; http://lattes.cnpq.br/5998752909180697; Santos, Alyson de Jesus dos; http://lattes.cnpq.br/5998752909180697; Rodrigues, Marlos André Silva Rodrigues; http://lattes.cnpq.br/0650250324042531; Nascimento, José Fábio de Lima; ttp://lattes.cnpq.br/0056522906128231; Alencar, Sérgio Costa Martins de; http://lattes.cnpq.br/7936103793279068Trabalho de Conclusão de Curso Análise comparativa de algoritmos de reconhecimento facial(2023-02-17) Ricardo, Misael Marques; Santos, Alyson de Jesus dos; http://lattes.cnpq.br/5998752909180697; Santos, Alyson de Jesus dos; http://lattes.cnpq.br/5998752909180697; Costa, Jaidson Brandão da; http://lattes.cnpq.br/4553321582341998; Compto, Gabriel Pinheiro; http://lattes.cnpq.br/5432787843953143THE PURPOSE OF THIS PAPER IS TO CARRY OUT A STUDY ON THE OPERATION OF FACE RECOGNITION ALGORITHMS DESCRIBED IN THE LITERATURE. THREE OF THE FACIAL RECOGNITION ALGORITHMS DESCRIBED IN THE LITERATURE WERE USED AND A TEST SET IS PROPOSED IN ORDER TO EVALUATE THE RESULT AND VERIFY THE POSSIBILITY OF USING THE ALGORITHM IN AN ACCESS CONTROL SYSTEM TO PHYSICAL SPACES. THE EIGENFACES, FISHERFACES AND LBPH ALGORITHMS WERE USED. TO THE TESTS THREE DATABASES WERE USED: THE FIRST WAS CONSTITUTED WITH PHOTOS OF THE AUTHOR AND HIS FAMILY, THE SECOND WAS THE DATABASE OF PEOPLE WITH AND WITHOUT MASKS AND LASTLY THE DATABASE CONSISTED OF PEOPLE WITH AND WITHOUT GLASSES AND THE RESULTS THE ACCURACY OF RECOGNITION WERE REPORTED AND PRESENTED IN THIS PAPER.Trabalho de Conclusão de Curso Detecção de defeitos em placas de circuito impresso aplicando visão computacional(2023-02-17) Monteiro, Lucas Paiva; Santos, Alyson de Jesus dos; Santos, Alyson de Jesus dos; Costa, Jaidson Brandão da; http://lattes.cnpq.br/4553321582341998; Compto, Gabriel Pinheiro; http://lattes.cnpq.br/5432787843953143This work consists of the development of a vision system capable of identifying faults in printed circuit boards. The system consists of a 2MP high-speed camera, a camera controller to receive the image and apply computational vision tools, a PLC to process camera information and an HMI to serve as an interface for the test operator, allowing a monitoring of results and control of the camera trigger and reset of post-shift information. The vision system is composed of the camera and the controller and works applying computer vision tools, in this work we use a self-learning pattern identification tool that allowed us to obtain a higher quality degree in the test results from a group of images.Trabalho de Conclusão de Curso Desenvolvimento de um sistema inteligente OCR utilizando visão computacional para leitura de etiquetas de roteador(2022-09-29) Xavier, Giovane Taveira de Souza; Santos, Alyson de Jesus dos; Santos, Alyson de Jesus dos; Compto, Gabriel Pinheiro; Costa, Jaidson BrandãoThis final paper consists of the development of an intelligent OCR system using computer vision to read router labels, to extract information to be inserted into programs that perform router tests. The system uses a Raspberry Pi 4 as the system's processing center, coupled to it is a Logitech C270 webcam, responsible for capturing the images of the labels. The vision system uses OpenCV to process the image so that the OCR algorithm extracts the text from the image, after which the program stores and displays this information on screens simulating the test software.Trabalho de Conclusão de Curso Desenvolvimento de um sistema de inspeção de componentes utilizando técnicas de visão computacional(2022-04-29) Prado, Wendel da Costa; Reis, Ailton Gonçalves; Reis, Ailton Gonçalves; Queiroga, Alberto Luis Fernandes; Santos, Alysson Jesus dosThis course conclusion paper (CCP) aims to present a system that, through Computer Vision, can identify the presence of electronic components on electronic boards, without human intervention, and make decisions by triggering a panel that displays the localized error. Thus, to develop a system that can capture data such as perimeter, part area, position, color, using only one sensor, and from these characteristics the system will be able to show if a specific component is missing or if it is showing some type of failure, such as: absence of a terminal or displacement.Trabalho de Conclusão de Curso Sistema de detecção de objeto personalizado utilizando inteligência artificial(2020-11-06) Marques, Lucas de Souza; Frota, Vitor Bremgartner da; Frota, Vitor Bremgartner da; Fernandes, Priscila Silva; Netto, José MagalhãesThis work aims to develop an automatic object verification system on a production line. Currently, there is a market for object detection that comprises values above R$ 100 thousand reais for the implementation of the same project. The project is based on being developed on a completely Open Source platform. The proposal consists of creating a low-cost Python language system that solves the problems of instantaneous line stops and field problems related to the lack of the object and make it portable to any computer. Due to the fast time of the line some boxes go unnoticed in the eyes of the operator and with the implemented system it is expected that all boxes are controlled and detect all manuals and take their own actions autonomously without the need for human intervention following the concept of industry 4.0. The developed system had quick responses (about 40 milliseconds) for the evaluation of each image and is already implemented in the factory.