Detecção e rastreamento de meteoros: abordagem em operações morfológicas

• Autor(a) principal: Campanini, Eduardo Felipe
• Data de Publicação: 2023
• Tipo de documento: Trabalho de conclusão de curso
• Idioma: por
• Título da fonte: Repositório Institucional da UFSCAR
• Texto Completo: https://repositorio.ufscar.br/handle/20.500.14289/18950
• topic: Visão computacional; Análise de vídeo; Meteoro; BRAMON; Processamento de imagem; OpenCV; Rastreamento de meteoro; Rastreamento de objetos; Detecção de objetos; Computer vision; Video analysis; Meteor; Image processing; Meteor tracking; object tracking; object detection; ENGENHARIAS::ENGENHARIA ELETRICA

The object detection and tracking system implemented in this project consists of a series of steps that involve capturing and preprocessing video frames, object detection through image segmentation, object tracking using a region correlation algorithm, and finally outputting the results in Excel and XML files. The implementation makes use of image processing and computer vision libraries such as OpenCV and Pandas. Object detection is performed through adaptive thresholding and image segmentation, while tracking is done through the KCF algorithm, which uses region correlation to track objects over video frames. The tracking results are stored in a dataframe, which is later sorted and saved to Excel and XML files. The Excel file contains information such as the object identifier, centroid coordinates, and the frame number in which the object was detected. The XML file contains similar information, but in a more structured format. The system showed good results in different test scenarios, being able to detect and track objects with high precision and reliability. The code is highly configurable, allowing adjustments for different types of objects, sizes and speeds of movement, as well as for different video recording environments.