Application of artificial intelligence-based techniques for sustainable handling and processing of Katamba fish in household-scale industries

This study explores the application of artificial intelligence (AI)-based methods to address quality and sustainability challenges in small-scale Katamba fish processing. As an essential source of protein and income for local communities, Katamba fish is traditionally processed through manual methods that often lead to inefficiencies, high spoilage rates, and inconsistent product quality. The objective of this research is to evaluate the potential of AI, specifically ResNet-based models, in enhancing the quality and consistency of Katamba fish processing, while minimizing environmental impact and supporting sustainable resource use. By tailoring AI solutions to the domestic sector, the study aims to provide a scalable model for improving processing standards across similar small-scale food industries. The research employed a mixed-methods approach, combining quantitative analysis of processing metrics (such as salt and moisture content) with qualitative insights from interviews and observations. Thirty household fish processors participated, representing a diverse range of resources and skills. Results indicate that AI-based processing methods can effectively reduce salt content and improve moisture retention, both crucial for product quality. Sensory analysis further demonstrated high consistency in color, though aroma and texture showed moderate variability, suggesting areas for improvement. These findings highlight the potential of AI to support consistent quality control and promote sustainable practices in small-scale fish processing, with implications for broader food security and public health initiatives.