<![CDATA[As living things, humans need to survive. One of the basic need human's bodies require to survive is food. Foods provide nutrients that contain carbohydrates, protein, minerals, fats, and vitamins for boosting endurance. Basically, foods can be easily identified with human's eyes. But it is not like the brain-computer that require the introduction or features extraction from food objects for classification. The features extraction used are HSV Color Moment for color features and Haralick for texture features. Then, the results of the features extraction will be classified using the Naive Bayes classifier method. The data set used are based of the primary data that contains pictures and the pictures were taken by the smartphone camera consist of 276 foods images.. This research uses 2 testing processes, that are the comparison of the amount of the training data and testing data, and the testing of the used features. Based on the testing of the comparison of the amount of the training data and the training data using K-Fold Cross Validation, it showed that the best accuracy is 61,95% that using 166 training data images and 110 training data images. Then, the accuracy from the features test that was just using the HSV Color Moment feature is about 57,66%. The accuracy from test that using the Haralick feature is 36,67%. The accuracy from the combination of 2 features of the HSV Color Moment and Haralick are better than only using the texture features with the 56,33% accuracy. The image processing technique using HSV Color Moment and Haralick features extraction can be used for foods image classification using the Naive Bayes classifier method.]]>
Copyrights © 2019
