In-line detection of foreign particles in sugar and color measurement using AI-based methodology

This study focuses on the in-line detection of foreign particles in dry white sugar on a conveyor belt, with an added attention on color measurement. The presence of non-sugar particles in sugar products can lead to customer complaints and quality concerns, making rigorous quality control essential. To address this issue, an innovative approach combines area cameras with light sources covering the entire width of the conveyor belt, along with an artificial intelligence (AI) based methodology for foreign particle classification. The primary objective is to enhance the understanding of foreign particle occurrences within sugar to silo, while simultaneously monitoring the sugar color. Area cameras capture real-time images of the sugar on the conveyor belt, allowing for continuous monitoring and analysis. An AI-based approach is used to identify and classify foreign particles as well as to estimate sugar color. Monitoring of these parameters helps to find root causes regarding product quality issues and appearance of non-sugar contaminants.