Pengembangan Aplikasi Klasifikasi Alat Transportasi Berdasarkan Citra Digital untuk Pencatatan Aset Studi Kasus: PT. Pulo Mas Jaya
This research aims to develop desktop-based software that is useful for classifying digital images of transportation equipment for asset recording using the Convolutional Neural Network method to classify image data. The data used as training data is 15000 data which is divided into 3 data groups. While the test data used in this study were 3000 data. The convolutional neural network method is implemented using the TensorFlow software library. In this study, using the incremental model of software development. This incremental stage is divided into two stages, namely the first iteration focuses on making the classification model of transportation equipment and the second iteration focuses on making GUI-based applications. From the research results, iteration one produces a model with an accuracy value of 92.17%. for training data at 40 epochs and 92.16% on test data. Then, iteration two creates a desktop- based user interface built using the Tkinter framework.