<![CDATA[Gayo Arabica coffee is one of Indonesia's leading commodities that currently faces serious threats in the form of counterfeiting practices and declining quality standards. Determining coffee quality and purity, which still relies on human sensory testing (sight, smell, and taste), is subjective, slow, and carries the risk of misjudgment, especially when performed by inexperienced assessors. This study aims to design and assess the feasibility of ArabiScan, a purity detection and quality classification system for Gayo Arabica coffee based on digital image processing and the K-Nearest Neighbor (K-NN) algorithm. The study employed a qualitative descriptive method through literature review and conceptual system design. The feasibility of its implementation was then analyzed using a SWOT (Strengths, Weaknesses, Opportunities, and Threats) analysis. The results indicate that the ArabiScan system operates through four stages: image acquisition, image processing, K-NN-based classification, and output presentation in the form of quality categories and purity percentages. Based on previous research, digital image-based methods and learning algorithms such as K-NN can achieve 94-95% classification accuracy for coffee, significantly more objective and consistent than conventional methods. A SWOT analysis shows that the system's main strengths lie in its objectivity and speed of assessment, while its weaknesses are the need for a comprehensive standard database and initial investment. It can be concluded that ArabiScan has the potential to be an effective Smart Farming technology solution for maintaining the authenticity, quality, and competitiveness of Gayo Arabica coffee in both national and international markets.]]>
