Machine learning applied to emerald gemstone grading: framework proposal and creation of a public dataset
Researchers have developed a novel machine learning framework to automate the grading of emerald gemstones, moving away from subjective human evaluation. This system integrates image acquisition with processing to categorize stones, achieving a 98% accuracy rate. The proposed method reportedly outperforms a deep learning approach and includes a newly created public dataset of 192 emerald images with extracted features. AI
IMPACT Automates a subjective industry process, potentially setting a precedent for AI in specialized grading and authentication.