A Machine Learning Approach to Enhance Real-Time Harbor Management

Abstract

Fisheries industry is a vital sector of Sri Lanka’s economy and each departing and arriving fishing vessel should have gone through ample security check by the harbor authorities. But with the COVID 19 pandemic and social distancing procedure, harbor authorities are facing difficulties detecting and recognizing fishing vessels by getting on the boats as usual. Also, currently harbors are using a paper-based system for recording the information on boat departures and arrivals. This leads to the inefficiency of harbor management process, delays in rescue missions and failures of security missions. To solve these problems, this paper introduces a Boat Recognition and Automated Harbor Management System (BRAHMS) which is based on YOLO v5 algorithm. In this research, a novel de-skewing method is discovered for the slanted license plate recognition process. The de-skewing process aims for three main approaches: auto de-skewing, manual de-skewing and a hybrid de-skewing which uses both auto and manual processes together.