Automatic Monitoring Cheese Ripeness Using Computer Vision and Artificial Intelligence

Ripening is a very important process that contributes to cheese quality, as its characteristics are determined by the biochemical changes occur during this period. Therefore, monitoring ripening time fundamental task market quality product in timely manner. However, it difficult accurately determine degree of ripeness. Although some scientific methods have also been proposed literature, conventional adopted dairy industries typically based on visual and weight control. This study proposes novel approach aimed at automatically analysis images acquired photo camera. Both computer vision machine learning techniques used deal with task. The dataset 195 (specifically collected from an Italian industry), which represent Pecorino forms four degrees All stages but one labeled “day 18”, has 45 images, consist 50 images. These handled image processing then classified according ripening, i.e., 18, 22, 24, 30 days. A 5-fold cross-validation strategy was empirically evaluate performance models. During phase, each training fold augmented online. allowed use 624 for training, leaving 39 original per testing. Experimental results demonstrated validity approach, showing good most trained