| 2 |
Smart Access by Face Recognition |
Penelitian ini mengembangkan sistem buka pintu otomatis menggunakan sistem pengenalan wajah. |
|
|
2024 |
2026 |
Berjalan
|
| 2 |
Otomatisasi Klasifikasi Kematangan Buah Berdasarkan Semantic Template Warna, Tekstur, dan Shape deng |
In this study, we have created a fruit ripeness dataset for 8 categories, namely Ripe Mango, Ripe Tomato, Ripe Orange, Ripe Apple, Unripe Mango, Unripe Tomato, Unripe Orange, and Unripe Apple. Based on the fruit ripeness dataset, we build a classification model of fruit ripeness using the SVM algorithm. Color feature extraction implemented in this study is RGB, HSV, HSL, and L * a * b *. To determine fruit ripeness, we done by predict image input to the model generated. Based on the experiment result, we have found that the best SVM model in determining fruit ripeness is the 6thdegree polynomial kernel and by extracting HSV color features. We evaluated the model generated based on the value of accuracy, precision, recall, and F-Measure. The best performance of our system for accuracy, precision, recall, and F-Measures are 0.76, 0.80, 0.76, and 0.78, respectively. |
|
Skema: Penelitian Dasar - Dikti |
2018 |
2019 |
Selesai
|
| 2 |
Optimasi Mikroemulsi sebagai Nanoreaktor untuk Sintesis Nanopartikel: Pendekatan Machine Learning da |
Microemulsion is a stable nanoreactor system, where the type of phase formed is strongly influenced by the composition of oil, surfactant, and metal used. Determination of the microemulsion phase class is generally carried out by laboratory tests, thus requiring large materials and costs as well as a long time. The type of microemulsion phase is very important in the ideal nanoparticle synthesis process. The type of microemulsion phase plays a very important role in producing the size and morphology of nanoparticles. This study aims to predict the phase of microemulsion formation based on the composition of the given emulsion compound. This study proposes 4 (four) methods in predicting the prediction of microemulsion phase formation, namely: K-Nearest Neighbors (KNN), Naive Bayes (NB), Decision Tree (DT), and Support Vector Machine (SVM). In SVM, 4 (four) kernels are proposed, namely: linear, sigmoid, rbf, and poly. To determine the best features of microemulsion phase formation, 3 (three) types of feature selection are used, namely: mutual information (MI), selection feature, and SHAP. To ensure the stability and consistency of the model, 5-fold cross validation is proposed. The dataset used in this study is a primary dataset from experimental results collected directly from various types and amounts of emultant compound compositions for microemulsion phase formation. There are 8 emultant compound features with a total of 162 data. The best model for predicting the microemulsion phase is the SVM kernel poly. The accuracy, precision, recall, and f1-score of the best SVM poly model are each valued at 1.0. Three important features that influence the formation of the microemulsion phase based on MI are surfactant, co-surfactant, and metal. Three important features of microemulsion phase formation based on feature selection are co-surfactant, metal, and oil. Meanwhile, three important features of microemulsion phase formation based on SHAP model SVM poly are surfactant, co-surfactant, and oil. All features used significantly influence the prediction of the microemulsion phase. This is shown in the SVM poly model, where reducing the number of features in microemulsion phase prediction results in decreased model performance. |
|
Skema: Penelitian Fundamental - Reguler Kemdiktisaintek |
2025 |
2025 |
Selesai
|
| 2 |
Pengontrolan Pergerakan Robot Mobil Berbasis EEG P300 dengan Klasifikasi Metode Anfis |
- |
|
Dikti |
2015 |
2015 |
Selesai
|
| 2 |
Optimasi Sistem Kendali Drone Berbasis EEG dengan Machine Learning untuk Aplikasi Pertahanan |
Penelitian ini bertujuan untuk mengembangkan sistem kendali drone berbasis Brain-Computer Interface dengan pemanfaatan sinyal EEG dan algoritma Machine Learning guna meningkatkan efektivitas dan efisiensi operasional dalam aplikasi pertahanan. Sistem ini dirancang untuk menginterpretasikan pola aktivitas otak dan menerjemahkannya menjadi perintah kendali drone secara real-time. Tahapan penelitian mencakup pengumpulan dan pemrosesan sinyal EEG, preprocessing sinyal, ekstraksi fitur, serta pengembangan model klasifikasi berbasis Machine Learning, seperti Support Vector Machine (SVM), Random Forest, CNN, GRU, LSTM, Transformer, dan metode lainnya. Model yang telah dilatih diuji dalam lingkungan simulasi kendali drone berbasis pemograman python sebelum diterapkan pada perangkat keras. Untuk meningkatkan portabilitas dan efisiensi kendali real-time, sistem ini diintegrasikan dengan Lab Streaming Layer (LSL) dan Raspberry Pi, memungkinkan pemindahan proses komputasi dari komputer ke perangkat embedded. Evaluasi kinerja dilakukan dengan membandingkan hasil simulasi dan eksperimen nyata, guna mengukur keandalan sistem dalam kondisi operasional yang sebenarnya. Dengan penelitian ini, diharapkan pengembangan sistem kendali drone berbasis EEG dapat memberikan solusi inovatif dalam meningkatkan efektivitas operasi taktis di bidang pertahanan. Hasil penelitian juga berkontribusi dalam pengembangan teknologi BCI yang dapat diterapkan pada berbagai sektor, termasuk sistem kontrol nirkabel dan interaksi manusia-mesin. |
BRIN - Universitas Mandiri - Mabes TNI Sesko-AU |
|
2026 |
2027 |
Berjalan
|
| 2 |
IntelliMart |
IntelliMart adalah sistem Point of Sale (POS) dan Retail Management System berbasis cloud yang dirancang untuk membantu bisnis ritel, grosir, minimarket, toko kelontong, apotek, dan berbagai jenis usaha lainnya dalam mengelola operasional secara terintegrasi. IntelliMart menggabungkan proses penjualan, pengelolaan inventori, pembelian, pelanggan, pemasok, hingga analisis bisnis dalam satu platform yang mudah digunakan dan dapat diakses secara real-time. Sistem POS modern umumnya mengintegrasikan transaksi, inventaris, pelaporan, dan manajemen pelanggan untuk meningkatkan efisiensi operasional. |
|
|
2026 |
2028 |
Berjalan
|
