Detection and identifitication of registration and fishing gear in vessels

• Autor(a) principal: Branquinho, Jorge Miguel de Albuquerque
• Data de Publicação: 2017
• Tipo de documento: Dissertação
• Idioma: eng
• Título da fonte: Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)
• Texto Completo: http://hdl.handle.net/10071/15093
• topic: Embarcação de pesca; Sistema tecnológico; Monitorização; Fishing; Vessel; CCTV; Image processing; Plate recognition; Machine learning; Computer vision; Machine vision

Illegal, unreported and unregulated (IUU) fishing is a global menace to both marine ecosystems and sustainable fisheries. IUU products often come from fisheries lacking conservation and management measures, which allows the violation of bycatch limits or unreported catching. To counteract such issue, some countries adopted vessel monitoring systems (VMS) in order to track and monitor the activities of fishing vessels. The VMS approach is not flawless and as such, there are still known cases of IUU fishing. The present work is integrated in a project PT2020 SeeItAll of the company Xsealence and was included in INOV tasks in which a monitoring system using video cameras in the Ports (Non-boarded System) was developed, in order to detect registrations of vessels. This system registers the time of entry or exit of the vessel in the port. A second system (Boarded System) works with a camera placed in each vessel and an automatic learning algorithm detects and records fishing activities, for a comparison with the vessel’s fishing report.