Article Per Year (5 Year)
p-Index From 2021 - 2026
3.51
P-Index
This Author published in this journals
All Journal Indonesian Journal of Electrical Engineering and Informatics (IJEEI) JURNAL NASIONAL TEKNIK ELEKTRO Jurnal CoreIT JURNAL TEKNIK INFORMATIKA DAN SISTEM INFORMASI Jurnal Infomedia DoubleClick : Journal of Computer and Information Technology JuSiTik : Jurnal Sistem dan Teknologi Informasi Komunikasi CSRID (Computer Science Research and Its Development Journal) JURNAL SURYA ENERGY Jutisi: Jurnal Ilmiah Teknik Informatika dan Sistem Informasi Progresif: Jurnal Ilmiah Komputer JISKa (Jurnal Informatika Sunan Kalijaga) INTI Nusa Mandiri Jurnal JTIK (Jurnal Teknologi Informasi dan Komunikasi) JURMATIS : Jurnal Manajemen Teknologi dan Teknik Industri Jurnal Generic Brilliance: Research of Artificial Intelligence Algoritme Jurnal Mahasiswa Teknik Informatika PROSISKO : Jurnal Pengembangan Riset dan observasi Rekayasa Sistem Komputer Arcitech: Journal of Computer Science and Artificial Intelligence Digital Transformation Technology (Digitech) MDP Student Conference Scientific Journal of Informatics IPSSJ
Dedy Hermanto
Universitas Multi Data Palembang
Computer Science & IT Control & Systems Engineering Decision Sciences, Operations Research & Management Economics, Econometrics & Finance Education Electrical & Electronics Engineering Engineering Industrial & Manufacturing Engineering Languange, Linguistic, Communication & Media Library & Information Science Mathematics Mechanical Engineering Public Health Social Sciences Other

Published : 38 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 38 Documents
Search

 1   2   3   4 
DETEKSI JUMLAH KENDARAAN RODA EMPAT MENGGUNAKAN YOLO Therino Elevan, Rivaldo; Hermanto, Dedy; Puji Widiyanto, Eka
Integrative Perspectives of Social and Science Journal Vol. 2 No. 2 Maret (2025): Integrative Perspectives of Social and Science Journal
Publisher : PT Wahana Global Education

Show Abstract | Download Original | Original Source | Check in Google Scholar

Abstract

Kemacetan lalu lintas merupakan salah satu permasalahan utama di kawasan perkotaan, terutama di Indonesia yang memiliki populasi dan jumlah kendaraan yang terus meningkat. Penelitian ini bertujuan untuk mengembangkan perangkat lunak yang mampu memberikan informasi jumlah kendaraan pada suatu ruas jalan secara real-time, real-time atau waktu nyata ini sangat penting dalam pengelolaan lalu lintas, karena memungkinkan otoritas dan pengguna jalan untuk mengambil keputusan yang cepat dan tepat sehingga dapat membantu mengurangi potensi kemacetan. Penelitian ini mengembangkan perangkat lunak yang mampu mendeteksi jumlah kendaraan roda empat secara real-time menggunakan metode YOLOv8 (You Only Look Once version 8). Model dilatih menggunakan dataset kendaraan dari Roboflow, dengan jumlah data latih 70% dan data uji 30%. Pelatihan model dilakukan dengan 50, 75, dan 100 epoch, menghasilkan nilai mAP sebesar 77%, 79,4%, dan 79,9%. Dalam konteks lalu lintas, nilai mAP menunjukkan seberapa baik model dapat mengidentifikasi kendaraan dengan benar, sementara akurasi deteksi kendaraan berdasarkan klasifikasi (mobil, bus, dan truk) mencapai 93%, yang berarti perangkat lunak dapat secara efektif mengklasifikasikan kendaraan yang terdeteksi. Perangkat lunak juga dapat digunakan dengan berbagai perangkat elektronik seperti laptop, komputer, dan ponsel. Penelitian ini diharapkan dapat memberikan kontribusi dalam mendukung pengelolaan lalu lintas yang lebih efektif di masa depan.
Implementasi Metode YOLOv8 Mendeteksi Komputer Aktif dengan Subjek Layar Monitor Wijaya, Frisky; Hermanto, Dedy
JISKA (Jurnal Informatika Sunan Kalijaga) Vol. 10 No. 3 (2025): September 2025
Publisher : UIN Sunan Kalijaga Yogyakarta

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.14421/jiska.2025.10.3.319-330

Abstract

Computers are one example of technological advances used in education. The use of computers that are not turned off can cause damage to computer components, and the use of electrical energy can increase. Student disobedience in turning off school laboratory computers when finished using them causes teachers to conduct manual checks by visiting each computer laboratory in the school. Deep learning is a machine learning algorithm that uses artificial neural networks. Deep learning is usually used for image recognition, voice identification, and data pattern analysis. Therefore, this study will apply the Deep Learning method, specifically YOLOv8, which aims to detect active computers based on the subject of the monitor screen and is expected to provide information about computers that are still active in the school laboratory. Based on the study's results, which detected 10 active computers, the 200-epoch model was selected with 100% accuracy at a speed of 2ms. Twenty active computers were selected, with 200 epoch models achieving 95% accuracy at a speed of 6ms per epoch. Thirty active computers were selected, with 100 epoch models achieving 96.67% accuracy at a speed of 3ms.
Penentuan Epochs Hasil Model Terbaik: Studi Kasus Algoritma YOLOv8 Jonathan, Jasen; Dedy Hermanto
Digital Transformation Technology Vol. 4 No. 2 (2024): Periode September 2024
Publisher : Information Technology and Science(ITScience)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47709/digitech.v4i2.4640

Abstract

Salah satu pengembangan machine learning yaitu deep learning merupakan salah satu metode inti dalam artificial intelligence yang sedang berkembang dengan pesat, dikarenakan kemampuannya dalam mempelajari informasi dalam jumlah besar. Salah satu cabang dari deep learning adalah computer vision, dan Convolutional Neural Network (CNN) yang merupakan metode yang paling banyak digunakan untuk melakukan pemrosesan citra. YOLOv8 merupakan salah satu algoritma yang menggunakan CNN yang telah dimodifikasi sebagai dasar, YOLOv8 merupakan algoritma open-source yang paling banyak digunakan dikarenakan menghasilkan hasil pengenalan objek yang akurat, cepat, dan mudah untuk di implementasikan. Proses pelatihan model dari YOLOv8 membutuhkan perangkat yang cukup memadai dengan jumlah epochs yang ditentukan secara manual. Penelitian ini bertujuan untuk mengetahui jumlah epoch yang dibutuhkan dalam membuat model YOLOv8 sesuai dengan kriteria yang di tentukan pada penelitian ini. Pelatihan akan dilakukan dengan menggunakan 50 epochs, 100 epochs, 150 epochs, 200 epochs, 250 epochs, dan 300 epochs. Pelatihan akan di jalankan dengan menggunakan dataset citra bibit ikan lele yang terdiri dari 753 gambar bibit ikan lele yang telah di anotasikan. Pelatihan dijalankan dengan menggunakan CPU Ryzen 5 4600H. Berdasarkan dari hasil pelatihan didapatkan bahwa 50 epochs memiliki waktu pelatihan tercepat dengan hasil yang kurang baik. Hasil terbaik terdapat pada 200-300 epochs dengan rata-rata precision sebesar 96% dengan waktu pelatihan yang cukup lama.
Deteksi Penyakit Daun Teh Berdasarkan Citra Menggunakan Deep Learning Saputra, Andreas; Hermanto, Dedy
Jurnal JTIK (Jurnal Teknologi Informasi dan Komunikasi) Vol 10 No 2 (2026): APRIL 2026
Publisher : Lembaga Otonom Lembaga Informasi dan Riset Indonesia (KITA INFO dan RISET)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35870/jtik.v10i2.5657

Abstract

Tea plant (Camellia sinensis) originates from China and is one of the most widely consumed beverages in the world. Tea plants are vulnerable to leaf diseases such as Tea Leaf Blight, Tea Red Leaf Spot, and Tea Red Scab, which can reduce the quality and productivity of the harvest. Manual disease identification is still commonly used, but this method has many limitations, such as dependence on farmers’ experience and inaccuracy in early detection. This study aims to apply the YOLOv11 algorithm as an object detection method to automatically, quickly, and accurately detect four classes of tea leaf conditions (three diseases and one healthy). The dataset used consists of 3,960 high-resolution tea leaf images that have undergone segmentation, augmentation, and normalization processes. The research was carried out through image preprocessing, YOLOv11 model training, and model performance evaluation using precision, recall, F1-score, and mean Average Precision (mAP) metrics. The results of tea leaf disease detection using YOLOv11 achieved an average precision of 97.2%, recall of 98.2%, mAP@0.5 of 98.8%, and mAP@0.5:0.95 of 95.5%. This model can be used to help farmers identify tea leaf diseases more quickly and reduce the risk of crop yield losses.
Klasifikasi Motif Kain Jumputan Palembang Menggunakan Metode CNN dengan Arsitektur Resnet-50 Mauladi, Muhammad; Hermanto, Dedy
Arcitech: Journal of Computer Science and Artificial Intelligence Vol. 5 No. 2 (2025): December 2025
Publisher : Institut Agama Islam Negeri (IAIN) Curup

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.29240/arcitech.v5i2.15310

Abstract

This study develops an automated classification system for Palembang jumputan textile motifs based on computer vision to address inter-motif pattern similarities that often challenge non-expert users and hinder the digital documentation of textile cultural heritage. Unlike traditional textile studies that typically employ generic Convolutional Neural Networks (CNNs), this research applies transfer learning using the ResNet-50 architecture on a primary dataset consisting of five motif classes: lilin, titik 7, titik 9, bunga tabur, and akoprin daun. The dataset is divided into training, validation, and testing sets, followed by preprocessing and image augmentation to enhance data variability. The model is trained with learning rate tuning, and the best configuration achieves a training accuracy of 95.57%, a validation accuracy of 87.33%, and a testing accuracy of 88%. Evaluation using a classification report and confusion matrix indicates excellent performance for the titik 9 and bunga tabur motifs, with precision and recall values approaching 1.00, while misclassifications still occur in the lilin motif due to visual similarity. These results confirm the effectiveness of ResNet-50 for jumputan motif classification and support cultural preservation through faster and more consistent motif identification.
KLASIFIKASI JENIS IKAN AIR TAWAR MENGGUNAKAN ALGORITMA CNN DAN ARSITEKTUR ALEXNET Ahmad Rizky; Dedy Hermanto
INTI Nusa Mandiri Vol. 20 No. 2 (2026): INTI Periode Februari 2026
Publisher : Lembaga Penelitian dan Pengabdian Pada Masyarakat

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33480/inti.v20i2.8025

Abstract

Freshwater fish are an important commodity in the fishing industry that requires an accurate classification system. This study aims to develop a freshwater fish classification system using the Convolutional Neural Network (CNN) algorithm with AlexNet architecture, as well as applying data augmentation techniques to improve model accuracy. The dataset used consists of 488 images of five types of freshwater fish, namely catfish, baung fish, tapah fish, juaro fish, and patin fish, which were then augmented into 68,400 images. The model was trained using the Adam optimizer, with a batch size of 16, a learning rate of 1e-5, and 200 epochs. The results of the experiment show that the model achieved a training accuracy of 71.09%, a validation accuracy of 85.00%, and a testing accuracy of 80.29%. Precision reached 0.8310, Recall 0.7909, and F1-score 0.7912, indicating the model's excellent performance in classifying freshwater fish species. This research is expected to support the development of an automatic classification system for the freshwater fisheries industry
Klasifikasi Tingkat Kematangan Buah Kelapa Sawit Menggunakan EfficientNet-B7 Davincylin, Valen Julyo Armando; Hermanto, Dedy
Jutisi : Jurnal Ilmiah Teknik Informatika dan Sistem Informasi Vol 14, No 3: Desember 2025
Publisher : STMIK Banjarbaru

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35889/jutisi.v14i3.3399

Abstract

Determining the ripeness level of oil palm fresh fruit bunches (FFB) is a crucial factor affecting oil yield and quality; however, field assessment is still largely performed manually and is prone to subjectivity and errors. This study aims to develop an image-based classification system for oil palm fruit ripeness using a Convolutional Neural Network (CNN) with the EfficientNet-B7 architecture. The proposed method applies transfer learning and fine-tuning on the public dataset “An Ordinal Dataset for Ripeness Level Classification in Oil Palm Fruit Quality Grading,” which contains 4,728 images across five ripeness classes. The methodology includes image preprocessing, normalization, and data augmentation techniques such as rotation, flipping, and zooming. The model is trained using the Adam optimizer and evaluated using accuracy, precision, recall, and F1-score metrics. Experimental results show that the proposed model achieves an accuracy of 93.23% with stable performance across all classes. These findings indicate that EfficientNet-B7 is effective for oil palm fruit ripeness classification and has strong potential to be implemented as a decision-support system for more objective and consistent harvest timing.Keywords: EfficientNet-B7; Convolutional Neural Network; Ripeness classification AbstrakPenentuan tingkat kematangan tandan buah segar (TBS) kelapa sawit merupakan faktor penting yang memengaruhi rendemen dan kualitas minyak sawit, namun proses penilaiannya di lapangan masih dilakukan secara manual sehingga rentan terhadap subjektivitas dan kesalahan. Penelitian ini bertujuan untuk mengembangkan sistem klasifikasi tingkat kematangan buah kelapa sawit berbasis citra digital menggunakan algoritma Convolutional Neural Network (CNN) dengan arsitektur EfficientNet-B7. Metode yang digunakan meliputi transfer learning dan fine-tuning pada dataset publik “An Ordinal Dataset for Ripeness Level Classification in Oil Palm Fruit Quality Grading” yang terdiri dari 4.728 citra dalam lima kelas kematangan. Tahapan penelitian mencakup preprocessing citra, normalisasi, serta augmentasi data berupa rotasi, flip, dan zoom. Model dilatih menggunakan optimizer Adam dan dievaluasi menggunakan metrik accuracy, precision, recall, dan F1-score. Hasil pengujian menunjukkan bahwa model mencapai accuracy sebesar 93,23% dengan performa klasifikasi yang stabil pada seluruh kelas. Berdasarkan hasil tersebut, EfficientNet-B7 terbukti efektif untuk klasifikasi tingkat kematangan buah sawit dan berpotensi diterapkan sebagai sistem pendukung keputusan dalam penentuan waktu panen yang lebih objektif.  
Prediksi Kelayakan Kredit Nasabah Dengan Penerapan Cost-Sensitive Random Forest Lucretia, Jolyn; Hermanto, Dedy
Progresif: Jurnal Ilmiah Komputer Vol 22, No 1 (2026): Januari
Publisher : STMIK Banjarbaru

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35889/progresif.v22i1.3401

Abstract

The high risk of credit default and class imbalance in customer data pose major challenges in developing accurate credit scoring systems. This condition causes predictive models to be biased toward the majority class, thereby reducing the ability to detect high-risk borrowers. This study develops a credit scoring model for imbalanced data using the Synthetic Minority Oversampling Technique (SMOTE) and cost-sensitive Random Forest with hyperparameter optimization via GridSearchCV. The dataset consists of 32,581 customer records. Experimental results show that the best configuration with n_estimators = 200 achieves a cross-validation F1-score of 0.813750. On the test data, the model attains an accuracy of 0.927267, precision of 0.911458, recall of 0.738397, and an F1-score of 0.815851, indicating improved and more balanced detection of high-risk borrowers.Keywords: Random Forest; SMOTE; Cost-sensitive learning; GridSearchCV AbstrakTingginya risiko gagal bayar kredit dan ketidakseimbangan kelas pada data nasabah menjadi tantangan utama dalam pengembangan sistem credit scoring yang akurat. Kondisi ini menyebabkan model prediksi cenderung bias terhadap kelas mayoritas sehingga kemampuan deteksi debitur berisiko menjadi kurang optimal. Penelitian ini mengembangkan model credit scoring pada data tidak seimbang menggunakan Synthetic Minority Oversampling Technique (SMOTE) dan Cost-Sensitive Random Forest dengan optimasi hyperparameter GridSearchCV. Dataset yang digunakan berjumlah 32.581 data nasabah. Hasil pengujian menunjukkan konfigurasi terbaik dengan n_estimators = 200 menghasilkan F1-score validasi silang sebesar 0,813750. Pada data uji, model mencapai akurasi 0,927267, precision 0,911458, recall 0,738397, dan F1-score 0,815851, yang menunjukkan peningkatan kemampuan deteksi debitur berisiko secara lebih seimbang.Kata kunci: Random Forest; SMOTE; Cost-sensitive learning; GridSearchCV.
Co-Authors Abdul Rahman Ahmad Rizky Ahmad, Prima Anam, M. Syaiful Andreas Saputra, Andreas Ardianto, Feby Davincylin, Valen Julyo Armando Deka Putra Pamungkas Deris Stiawan Dermawan, Sepri Desy Iba Ricoida Eka Puji Widiyanto Eldo Suryadi Emi Yuliati Evan Susanto Hendriko, Viky Iis Iis Pradesan Ivan Ricardo Jiwanata Jonathan, Jasen Joseph Alberto Jovian Aristo Komah, Neilsen Nicholas Kornelia* Kornelia* Kynta, Diva Putri Laksono, Ivan Luthfi Latius Hermawan Lucretia, Jolyn Mauladi, Muhammad Meilinda, Meilinda Muhammad Fadli Muhammad Fahmi Muhammad Rizky Pribadi Negara, Edi Surya Pibriana, Desi Qois Al Qorni Ramayanti, Indri Richardo Anggara Rizqie, Muhammad Qurhanul Rusbandi rusbandi rusbandi, rusbandi Saputra, Darwin Sepri Dermawan Shinta Puspasari Suluh Arif Wibowo Syakurah, Rizma Adlia Theng, Arifin Therino Elevan, Rivaldo Thingkilia Finnatia Husin Tri Hartati Utami Valentino, Calvin Bertnas Vannes Wijaya Wijaya, Frisky Yuniarti, Erliza