“…Activity Recognition (AR) is one of these areas where different sports activities such as walking, pedaling, climbing and swimming are recognized by smart wearable devices [20]. Learning algorithms paired to operate together with classification models such as k-Nearest Neighbors (kNN), Multilayer perceptron (MLP), Decision Tree (DT) and Support Vector Machine (SVM) have been proven effective in a variety of applications ranging from dry bean crops to cherry tomatoes and egg volume estimation [21,[23][24][25]. Most of the studies published in the field of vision-based product classification agree that in object categorization and grading, image resolution, sorting method and color represent the critical parameters of the process [13,[26][27].…”