%0 Thesis
%9 Skripsi
%A Reyhan Sapta Anggara, NIM.: 18106020036
%B FAKULTAS SAINS DAN TEKNOLOGI
%D 2025
%F digilib:72895
%I UIN SUNAN KALIJAGA YOGYAKARTA
%K CNN, Fluoroscience Spectro-Imaging System, Methanyl Yellow, Tahu Kuning
%P 128
%T DISKRIMINASI CITRA TAHU KUNING MURNI DAN TAHU KUNING TERKONTAMINASI METHANYL YELLOW MENGGUNAKAN THIRD GENERATION OF UIN SUNAN KALIJAGA’S HIGH POWER  UV-LED FLUOROSCIENCE SPECTRO-IMAGING SYSTEM  TERKOMBINASI DEEP LEARNING BERALGORITMA CONVOLUTIONAL NEURAL NETWORK (CNN)
%U https://digilib.uin-suka.ac.id/id/eprint/72895/
%X This research was motivated by the potential fraudulent practices of yellow tofu producers in using hazardous textile dyes, one of which is methanyl yellow. The objective of this research is to apply the Third Generation of UIN Sunan Kalijaga’s High Power UV-LED Fluorescence Spectro-Imaging System as a method to discriminate between pure yellow tofu and methanyl yellow-contaminated yellow tofu, and to classify their fluorescence images using deep learning with a Convolutional Neural Network (CNN) algorithm. The research method consisted of two main stages: data acquisition and data processing. The data acquisition stage included sample preparation and image acquisition, while the data processing stage involved three processes: preprocessing, model development and training, and validation, resulting in a CNN-based deep learning model.The results demonstrated that fluorescence images of pure yellow tofu and methanyl yellow-contaminated yellow tofu were successfully discriminated using the CNN-based deep learning approach. Furthermore, the CNN model achieved high performance in distinguishing between pure and contaminated samples, with an accuracy of 96% in image classification and 100% in the confusion matrix.
%Z Frida Agung Rakhmadi, M.Sc