Main Article Content
Keywords
Abstract
Coffee consumption has increased worldwide, and the rise in coffee trading has served as a key economic motivator for producers to add value and expand their market presence, thereby enhancing the value of their products. Among the implemented strategies to increase the value of coffee, the free pesticide product label stands out due to the market's tendency to consume this product. In this work, artificial intelligence techniques were employed to analyze the photoacoustic spectra of cherry coffee to determine if chemical-fertilizer traces are present in the coffee cherries. Spectra were divided into bands to improve the artificial algorithm's response. Subsequently, logistic regression, random forest classifier, support vector machines, and decision tree classifier algorithms were applied. The results indicate that principal component analysis offers the highest accuracy in detecting inorganic fertilizers at wavelengths of 320-330 nm and 560-620 nm in the analyzed samples. Consequently, photoacoustic spectrum analysis using artificial intelligence techniques represents a viable option for detecting inorganic fertilizers in coffee at the cherry stage rather than beans or powder.