Classification of Tempeh Maturity Using Decision Tree and Three Texture Features

Tempe is an average food from Indonesia, eaten in Indonesia. Even today, tempe around the world, and vegans world use tempeh as a meat substitute. This study plans to work on accuracy of characterization by utilizing three-element extraction technique choice tree arrangement strategy. research uses decision method with three texture features its classification. The results obtained indicate that this has highest Gabor channel level, including extraction, which 71% accuracy, split proportion 10;90 lowest 60% parted balance 90:10. most important level value GCLM precision 86% 90;10 price ratio for Wavelet rate 77%. It can be said elements, GLCM element Wavelet, at 10:90 86%. test shows Featured Tree highlight designation. was superior different strategies interaction development quality. In next research, improve performance so it reach 100% using CNN deep learning method. Then you also add Support Vector Machine (SVM) Naive Bayes methods based Extraction feature.