Seaweed Growth Detection in Aquaculture Environment Using Simple Linear Iterative Clustering Method
Estimating the total biomass of cultivates in aquaculture plantations (fisheries, mussel plants, seaweed farms and compound sites) remains to be an issue for the industry and the researchers alike. There has been a diverse array of approaches towards this issue, like using markers, manually stapling the leaflets, weighting the actual mass of the organism and calculating the total mass by extrapolation. Seaweed growth detection is a subset of this problem. Our goal is to introduce a solution by automatically detecting the ratio of the target object in images of seaweed taken from an underwater environment. Researchers/operators then can evaluate the total mass of seaweed. This study aimed to function as a decision support system. The system is built based on an image segmentation algorithm named Simple Linear Iterative Clustering (SLIC) which is a kind of superpixel segmentation. This paper conveys the results obtained from our approach towards the seaweed growth detection, elaborates on the usage and feasibility of our solution in seaweed sites and showcase the economic impact in the industry. Other dimensions of the growth detection methods in current practice for seaweed growth is also discussed, such as lack of automation in the current best-practices while focusing on the difficulties accompanying this status-quo.