Uso de visão computacional para otimização do tempo

• Autor(a) principal: Hass, Cristofer Viante
• Data de Publicação: 2023
• Outros Autores: Nunes Junior, Leonaldo Ferreira
• Tipo de documento: Trabalho de conclusão de curso
• Idioma: por
• Título da fonte: Repositório Institucional da UTFPR (da Universidade Tecnológica Federal do Paraná (RIUT))
• Texto Completo: http://repositorio.utfpr.edu.br/jspui/handle/1/35363
• topic: Python (Linguagem de programação de computador); Interface de usuário baseada na Web; Processamento eletrônico de dados em tempo real; Python (Computer program language); Web-based user interfaces; Real-time data processing; CNPQ::CIENCIAS EXATAS E DA TERRA::CIENCIA DA COMPUTACAO

The present study focuses on the use of computer vision for time optimization in registration operations, aiming to explore the use of computer vision, specifically optical character recognition (OCR), as a solution to reduce the time required for registration operations. The proposal aims to enhance the experience of individuals who frequently face long queues during these operations. To achieve the stated objective, a web interface and a Python API, along with OpenCV and pytesseract tools, were utilized. A comparison between manual and automatic operations was conducted, with the time spent on each method being measured. The goal was to determine whether the use of OCR leads to a significant reduction in the time required to perform these operations. Based on the conducted experiments, it was observed that the utilization of computer vision for optical character recognition can indeed save time in registration operations. However, it was noted that the application of this method in chaotic environments can be challenging. Therefore, despite the identified benefits, careful consideration of the implementation context is necessary, taking into account factors such as image quality and the stability of the operational environment.