Analysis of Different Naïve Bayes Methods for Categorizing Spices Through Photo using the Speeded-up Robust Feature

Spices are biological natural resources that have long been used in human life. Spices are highly valued in the European market due to their flavor, aroma, and delicacy. Spices come in a variety of shapes and sizes, each with its own set of characteristics. Because there are so many different types of spices, many people are unfamiliar with their names and forms. As a result, this study discusses how to classify spices using the Nave Bayes method and the Speeded-up Robust Features feature extraction method. According to the results of the tests conducted in this study, experiments with 5 types of spices produced better results with an accuracy of 77.3%, precision of 77.5%, recall of 77.5%, f1 score of 76.4%, f beta score of 76.8%, and Jaccard score of 63.3%, whereas experiments with 10 types of spices and 15 types of spices produced less than the maximum. The findings revealed that the number of spice species used in extracting information is greatly influenced by feature extraction. Speeded-up Robust features that have been accelerated Feature Extraction works best when the number of spices extracted is small, and it performs poorly when used in a large number of classification types.