Development of an Automated System for Detecting Cannibalism and Water Ecosystem Monitoring for Crabs

Abstract

Cannibalism and crucial water parameters of crabs are two major obstacles for the crab farmers to gain maximum yield when fattening in a pond. Our project aims to help minimize the cannibalism rate of crabs as well as to give farmers more control over their water. We created a machine learning model to identify possible crab sounds that are produced by crabs when they feel threatened or in danger. The model will notify the farmer when it detects crab sounds continuously. A water parameter monitoring system is also implemented which uses various sensors (temperature, pH, salinity, DO) to monitor the water parameters. When a parameter reaches or exceeds its tolerance level, the system notifies the farmer. There is a user interface that can be accessed to view all the system status and sensor readings. The project aims to give farmers more control over when they are fattening in ponds.