01.1 - Graduação (Sede)
URI permanente desta comunidadehttps://arandu.ufrpe.br/handle/123456789/2
Navegar
6 resultados
Resultados da Pesquisa
Item Rastreabilidade na cadeia produtiva do atum (Thunnus sp.) comercializado fresco – revisão de literatura(2024-08-20) Alves, Adryanne Marjorie Souza Vitor; Moura, Andrea Paiva Botelho Lapenda de; http://lattes.cnpq.br/6414540974581675; http://lattes.cnpq.br/0158550024403180The Mandatory Supervised Internship is a required course in the 11th semester of the Veterinary Medicine Program at UFRPE, with a total workload of 420 hours. Its purpose is to provide students with practical experience necessary to develop the essential skills and competencies needed to become qualified veterinarians. The internship was completed at Produmar Exportadora de Produtos do Mar LTDA and in the Imaging Diagnostics Department at Veterinarii Recife from April 1 to June 19, 2024, under the supervision of veterinarians Ana Carolina Baptista Freitas Braga and Yannike Lourenço Maciel, and under the guidance of Professor Dr. Andrea Paiva Botelho Lapenda de Moura. The internship provided hands-on experience in animal product technology and imaging diagnostics, successfully fulfilling the objectives of the Mandatory Supervised Internship.Item Representação virtual para segurança de espaços através de detecção de objetos, pessoas e suas relações(2022-10-07) Torres, Lucas Amorim Vasconcelos; Simões, Francisco Paulo Magalhães; http://lattes.cnpq.br/4321649532287831; http://lattes.cnpq.br/8237338186784482The detection of risk situations is something that has been improved every year. This work presents the prototype of a system for monitoring the risk of accidents in industrial environments based on tracking objects and people using computer vision. In this work, visualization tools are used in virtual environments to detect collisions, verifying when a large object is close to colliding with people. The central idea is to make a spatialanalysis of a tractor, or some object similar to this type of vehicle, and of the people who transit in that place. Through this, it is possible to create visualization methods so that the end user, whether a work safety inspector or an industry 4.0 system, can understand what is happening in the surroundings of the place and the relationships between objects. In addition to viewing distances, the prototype allows changing the distances considered safe between objects and people, making it possible to also test different types of tool configuration.Item Avaliação de algoritmos de rastreamento no problema de detecção de pessoas no mar(2023-09-13) Nascimento, Ramicés Moisés do; Macário Filho, Valmir; http://lattes.cnpq.br/4346898674852080; http://lattes.cnpq.br/0247140467691140It is known that shark attacks are a constant fear for the population on the beaches of Pernambuco. Many of these attacks are fatal, which calls for some action to be taken, considering that Pernambuco beaches attract thousands of tourists each year. Therefore, researchers from UFRPE initiated a study aiming to develop a system for tracking people in the sea, which would make it possible to alert lifeguards when individuals exceed a designated safe area on the beach, as well as allocate a greater number of these professionals in areas with a higher concentration of people. The system was divided into three stages: image segmentation, detection of beachgoers, and tracking of individuals. This work focuses on the third stage. Tracking people is a complex task with high computational costs. Problems such as changes in lighting conditions, alterations in the direction of targets, and variations in the background are just a few of the difficulties that can be mentioned. Thus, the objective of this research is to evaluate six people tracking algorithms found in the literature using beach images. Firstly, a database of ten videos recorded at Boa Viagem beach in Pernambuco was manually labeled. Then, six algorithms were selected for evaluation. Subsequently, the output of each frame provided by the algorithm was compared with the previously labeled data, and an average was calculated. Overall averages were then obtained to assess the algorithm’s accuracy and execution time. Finally, the best algorithm was chosen for optimization using a genetic algorithm, and any improvements in the results were verified. CSRT was the algorithm that obtained the best result and after optimization with the genetic algorithm, an improvement of 20% in its accuracy was obtained.Item Raspagem de Dados Jurídicos Utilizando Scrapy(2021-12-20) Barbosa, Jadiel Eudes Mendonça; Bocanegra, Silvana; http://lattes.cnpq.br/4596111202208863; http://lattes.cnpq.br/8044959053132773Web scraping is a computational technique that uses a program to extract data that are hidden in web pages. In this way, this academic work aims to use how web scraping techniques to extract data from legal processes from the websites of the courts in order to help contracting companies to take strategic decisions with their legal departments.Item Rastreamento de pedestres 3D multi-câmera usando redes neurais de grafos(2022-05-27) Andrade, Isabella Stefanny Fernandes de; Lima, João Paulo Silva do Monte; http://lattes.cnpq.br/1916245590298485; http://lattes.cnpq.br/5529506615862118Tracking the position of pedestrians over time through camera images is a rising computer vision research topic. In multi-camera settings, the researches are even more recent. Many solutions use supervised neural networks to solve this problem, which can require a lot of effort to annotate the data in addition to a lot of time spent to train the network. The goals of this work are: develop variations of pedestrian tracking algorithms, being desirable to avoid the need to have annotated data; and compare the results obtained through accuracy metrics. Therefore, this work proposes an approach for tracking pedestrians in 3D space in multi-camera environments using the Message Passing Neural Network framework inspired by graphs. We evaluated the solution using the WILDTRACK dataset and a generalizable detection method, reaching 77.1% of MOTA when training with data obtained by a generalizable tracking algorithm. The algorithm can track at a 40 frames per second rate.Item Métodos para estudos de área de uso de serpentes da Floresta Atlântica(2020-02-03) Lima, Luiz Filipe Lira; Santos, Ednilza Maranhão dos; http://lattes.cnpq.br/5812920432455297; http://lattes.cnpq.br/8555166529926373