Sea Cense - A Commercially Valued Sea Cucumber Analyzer

In Sri Lanka, sea cucumbers are a highly sought-after marine product that is traded abroad. It has gained significant attention in recent years due to its high demand in the seafood market and its medicinal properties. Yet, the Sri Lankan Sea cucumber industry has not been able to fully realize its potential due to barriers like inaccurate methods for identifying and predicting the quality of sea cucumber species and the inability to provide a proper market value. This study offers an innovative solution that automates the evaluation and classification of sea cucumbers utilizing advanced computer vision and machine learning approaches. The proposed analyzer incorporates a multi-step approach, including image preprocessing, feature extraction, and classification. A set of preprocessed images are used to extract textural and distinguished color-based characteristics. These characteristics are then used to accurately classify sea cucumber species and grade them according to quality criteria using a machine learning algorithm. This project introduces a novel approach to analyzing sea cucumber length to calculate the price using a mobile application, making the task more efficient and accessible. The proposed solution “Sea Cense” comprises 4 models that are developed using the CNN algorithm. These models' experimental findings demonstrated accuracy levels exceeding 85%.