Real-time Video Fusion for Surveillance Applications

• Autor(a) principal: André da Cunha Trancoso Ranito
• Data de Publicação: 2019
• Tipo de documento: Dissertação
• Idioma: eng
• Título da fonte: Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)
• Texto Completo: https://hdl.handle.net/10216/122468
• topic: Engenharia electrotécnica, electrónica e informática; Electrical engineering, Electronic engineering, Information engineering

Surveillance is becoming a really important element in our daily lives. From urban environments as crime prevention, detection and resolution, vandalism prevention and traffic flow control to more remote environments such as military applications, for instance, to identify and locate enemy forces. As technology develops, the surveillance subject is also getting more sophisticated. The systems are improving quality wise, are getting safer, costs are getting lower, there is higher scalability of systems and there is better integration between different types of surveillance systems. One of the main types of surveillance is known as video surveillance. As the name states, this technique consists of a constant capture of images in order to obtain a sequence of events happening in a given location. However, one of the main disadvantages of these systems is the dependency on the visibility conditions available in the location. In the scope of this dissertation a real-time system capable of capturing images containing useful information even in low visibility conditions, such as nighttime, fog or smoke, was developed. For this purpose, a technique known as image fusion was used. In this case, a fusion between an image contained in the infrared spectrum and another contained in the visible spectrum. Sensing a complementary image of the environment in the infrared spectrum will provide extra information, such as the temperature. This extra information will then be fused with the visible spectrum image, generating just one image containing the information from both the visible and infrared images.