Fishing boat Activity Clustering and Timeline Studying (FACTS)
Abstract : Design of fishing boat for Pelabuhanratu fisherman as one of effort to increase production of capture fisheries. The fishing boat should be proper for the characteristic of its service area, as ;capacity of fishing boat up to 60 GT, the fishing boat has minimum 6 fish holds and location of fish hold in the middle body, the fishing boat hull has the bilge keel plate, and the material of hull fishing boat to be made of wooden, steel, aluminium, or fiberglass. Main dimesion of fishing boat is Length Over All = 25.436 m, Breadth = 4.55 m, Draft = 1.6 m, Speed = 12.5 knots. The research had been known every thing that will be supporting the production of capture fisheries like ; amount of fish production = 25.030 ton per day, the fishing port capacity approximately 268.957GT per day, the area of fishing port < 30 hectares, the zone of fish processing industry had not completed, therefore all data research result less than standard of Oceanic Fising Port. So Pelabuhanratu National Fishing Port can not be changed to Oceanic Fishing Port.
? Although existing studies have demonstrated some progress, most of these methods rely on speed and its related information to identify the type of fishing.
? It has also become important to rationalize scattered data collection efforts, as existing data are often poorly integrated in national systems, remaining buried in computer spreadsheets or paper files and thus unavailable for analysis or reporting.
? Food security exists when all people, at all times, have physical, social and economic access to sufficient, safe and nutritious food that meets their dietary needs and food preferences for an active and healthy life.
? Moreover, in high traffic density areas there is the problem of message collisions with the result of making the satellite receiver almost “blind” in such areas.
? In order to solve problems of misspelling a matching algorithm was implemented using the Levenshtein and Jaro strings matching distances functions.
? The value of using effort data at these finer time and space resolutions when evaluating environmental impacts of fishing activity.
• A novel algorithm for nearshore ship detection based on multidirectional information from high-resolution SAR images was proposed.
• Cross-validation (CV) was used to evaluate the performance of the proposed fishing vessel type classification model.
• We compare the proposed method with other advanced machine learning methods, such as logistic regression (LR), support vector machine (SVM), XGBoost, k-nearest neighbor (KNN) and CatBoost.
? In particular the performance resulted higher for otter trawler, bottom and mid-water, pair trawler, purse seiner and Danish seiner and lower in the case of gillnetter, long-liner and fishing pots.
? At the individual vessel level higher differences in the validation score between fishing and non-fishing and therefore better performance of the classification method were recorded for vessels having more AIS messages available to fit the characteristic bi-modal speed profiles.
? The use of EM algorithm to find the parameters of the speed profiles of individual vessels proved to be efficient in computational terms and suitable for the fast processing of large AIS dataset.
|