An Explainable Real-time Decision Support System for Identifying Fish Diseases and Analysing Water Quality

Abstract

Fishes account for approximately 15% of the animal protein intake of the human population globally and is the main source of income for some developing countries like Cambodia. Fish is a traditional staple in the Cambodian diet and vital to nutrition and food security. Due to a lack of human resources with expertise in research into detecting fish diseases, breeding, and raising technologies, most farmers are struggling to increase the productivity of fish. In this research work, an intelligent classification framework is proposed for increasing the productivity of fish farmers by providing them with a facility to detect fish diseases along with analyzing the water quality. To understand the problems faced by the local farmers, the research team visited the six fisheries farms in the Takeo Province and collected the requirements. After the identification of the key challenges, a general framework is proposed for the early detection of fish diseases. The proposed framework will help the local farmers to educate them about the fish diseases and help them to improve the fish productivity.