Klasifikasi Makanan Tradisional Mandar Menggunakan Ekstraksi Fitur Warna Dan Tekstur Dengan Metode K-Nearest Neighbour

Abstract

People who are seeing it for the first time or tourists from outside the area will find it difficult to distinguish the name, taste, or type of traditional food in an area just by looking ti t without tasting the food. One solution to make it easier for the public to know and monitor the types of food consumed is to create an intelligent system. To support this solution, research was conducted to identify types of food using color feature extraction and texture features. The initial stage in conducting the introduction by classifying the types of food. The classification process is carried out based on the value of the normalized feature extraction results. The recognition process begins with a preprocessing process to obtain an image object which is then followed by feature extraction. The feature extraction used is the Color Histogram and the Gray Level Co-occurrence Matrix. Feature extraction uses the Color Histogram using 3 color channels namely red, green, blue with each color having mean, standard deviation, and skewness features. In addition, feature extraction using the Gray Level Co-occurrence Matrix has 6, namely contrast, dissimilarity, homogeneity, angular second moment, energi, and correlation with the angle of taking pixel values ​​of 0°, 45°, 90°, and 135°. The method for classifying values ​​from feature extraction results uses the K-Nearest Neighbor. The best accuracy result is 82% at k=1 with a comparison of training data and test data of 60:40%.