Article Per Year (5 Year)
p-Index From 2021 - 2026
0.702
P-Index
This Author published in this journals
All Journal Jurnal Informatika dan Teknik Elektro Terapan Jurnal Ilmiah Betrik : Besemah Teknologi Informasi dan Komputer ILKOMEDIA: Jurnal Ilmu Komputer dan Multimedia Journal of Applied Science, Technology & Humanities
Dodik Ariyanto
IPB University
Humanities Computer Science & IT Control & Systems Engineering Economics, Econometrics & Finance Electrical & Electronics Engineering Languange, Linguistic, Communication & Media Library & Information Science Social Sciences

Published : 4 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 4 Documents
Search

 1 
MycoTrack: An Integrated Web and YOLOv5-Based Intelligent System for Monitoring and Predicting Wood Ear Mushroom Maturity Wuliddah Tamsil Barokah; Dwi Putra Kunto Anggoro; Nabil Kurnia Rozano; Ariel Mughnika Beers; Inna Novianty; Dodik Ariyanto; Lathifunnisa Fathonah
Journal of Applied Science, Technology & Humanities | JASTH Vol. 3 No. 2 (2026): March 2026
Publisher : Batrisya Education

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.62535/kfkect89

Abstract

Wood ear mushroom (Auricularia auricula-judae) cultivation requires strict environmental control and accurate harvest monitoring. To overcome the shortcomings of labor-intensive and error-prone manual inspection, this research developed MycoTrack, an intelligent system integrating rail-based robotics, YOLOv5 computer vision, and IoT sensors. MycoTrack utilizes a rail-based robot powered by a Raspberry Pi 4. The robot carries a Pi Camera for visual data acquisition and DHT-22 sensors to measure environmental temperature and humidity. This environmental data is continuously monitored and transmitted to a web-based dashboard for real-time visualization, providing instantaneous decision support to farmers. The YOLOv5 model is specifically trained to detect three critical growth phases—incubation, pinning, and fruiting—which enables the prediction of optimal harvest timing. System validation showed DHT-22 sensor accuracy of 96.4% and the YOLOv5 model achieved a mAP@50 of 0.782 with inference speeds suitable for edge devices. The rail robot demonstrated minimal positional deviation (less than 2.3 cm). MycoTrack offers an accessible, automated solution, representing an advancement in precision agriculture for mushroom cultivation. The system is modularly designed for easy adaptation to other mushroom environments and species.
Implementasi Model YOLOv8 dan OCR pada Sistem ANPR untuk Monitoring Smart Drop-off Zone Zafira A’idah Gunawan; Mauladi Fadhillah; Anargya Rabbani Aslam; Mayanda Mega Santoni; Dodik Ariyanto; Faldiena Marcelita
Jurnal Ilmu Komputer dan Multimedia Vol. 3 No. 1 (2026): ILKOMEDIA Edisi Juni 2026
Publisher : Akademi Komunitas Negeri Putra Sang Fajar Blitar

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.46510/ilkomedia.v3i1.97

Abstract

Area drop-off pada lingkungan publik sering mengalami kemacetan akibat kendaraan berhenti melebihi batas waktu yang ditentukan. Proses pengawasan yang masih dilakukan secara manual dinilai kurang efisien karena membutuhkan tenaga tambahan dan memperlambat penindakan pelanggaran. Penelitian ini bertujuan mengimplementasikan sistem Automatic Number Plate Recognition (ANPR) berbasis YOLOv8 dan Optical Character Recognition (OCR) untuk monitoring Smart Drop-off Zone secara otomatis. Dataset penelitian diperoleh dari Roboflow Universe sebanyak 1.870 gambar kendaraan dan ditingkatkan menjadi 3.275 gambar melalui proses augmentasi seperti perubahan pencahayaan, rotasi, blur, dan penambahan noise. Proses augmentasi hanya diterapkan pada data training, sedangkan data validation dan testing tetap menggunakan data asli untuk menjaga objektivitas evaluasi. Model YOLOv8 Nano dilatih menggunakan 100 epoch dengan dua konfigurasi ukuran input, yaitu 640×640 piksel dan 416×416 piksel, untuk membandingkan performa model pada kondisi implementasi nyata. Hasil pengujian menunjukkan bahwa Eksperimen 1 dengan input 640×640 piksel memperoleh nilai mAP50 sebesar 0.9214, precision sebesar 0.9347, dan recall sebesar 0.8795, sedangkan Eksperimen 2 dengan input 416×416 piksel memperoleh mAP50 sebesar 0.8084. Meskipun Eksperimen 1 menghasilkan nilai evaluasi yang lebih tinggi, model Eksperimen 2 dipilih untuk implementasi karena menunjukkan stabilitas deteksi yang lebih baik pada berbagai kondisi lingkungan seperti pencahayaan rendah dan sudut kamera dari atas. Setelah proses deteksi plat nomor dilakukan, sistem OCR menggunakan PaddleOCR berhasil membaca karakter pada plat kendaraan secara otomatis. Sistem kemudian diintegrasikan ke dalam aplikasi berbasis web yang mampu melakukan pencatatan durasi kendaraan, deteksi pelanggaran, dan pengiriman notifikasi otomatis kepada pengguna. Hasil implementasi menunjukkan bahwa sistem ANPR yang dikembangkan mampu mendukung monitoring kendaraan pada area Smart Drop-off Zone secara lebih efektif dan efisien pada berbagai kondisi lingkungan.
Komparasi Arsitektur Densenet121, MobilenetV2, dan Resnet50 Untuk Klasifikasi Awan Menggunakan Transfer Learning Yashin Al Fauzy Sabara; Carissa Fathinah Balti; Ahmad Hasan Faqih Aulia; Dodik Ariyanto; Faldiena Marcelita; Mayanda Mega Santoni
BETRIK Vol. 17 No. 02 (2026): Jurnal Ilmiah BETRIK : Besemah Teknologi Informasi dan Komputer
Publisher : PPPM Institut Teknologi Pagar Alam

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.36050/6hwf0s96

Abstract

Cloud image classification plays an important role in supporting atmospheric analysis and the development of artificial intelligence-based weather prediction systems. However, visual similarities among cloud types and variations in lighting conditions remain major challenges in automatic classification tasks. This study aims to compare the performance of three Convolutional Neural Network (CNN) architectures, namely DenseNet121, MobileNetV2, and ResNet50, for cloud image classification using a transfer learning approach. The dataset used in this study was the TJNU Ground-based Cloud Dataset, consisting of seven cloud classes with a total of 19.000 images; after removing the Mixed Clouds category with the total of 18.047 images; after removing. The research stages included pre-processing, data augmentation using the Color Jitter technique, model training through feature extraction and fine-tuning, and evaluation using Accuracy, Precision, Recall, and F1-Score metrics. Experimental results showed that ResNet50 achieved the best performance with an accuracy of 90,58% and an F1-score of 91,38%, followed by DenseNet121 with 89,53% accuracy and MobileNetV2 with 85,10% accuracy. In addition to obtaining the highest classification performance, ResNet50 also demonstrated good computational efficiency with the fastest training time of 20.9 minutes. These findings indicate that architectures based on residual learning are more effective in capturing the visual characteristics of cloud images compared to the other architectures evaluated in this study.
PENGEMBANGAN SISTEM GERBANG OTOMATIS MENGGUNAKAN PENGENALAN PLAT NOMOR KENDARAAN BERBASIS COMPUTER VISION DI IPB UNIVERSITY Puti Aisyah Lailatulrahmi; Muhammad Faiq Fawwaz; Ghaniyy Fattah Ramadhan; Dodik Ariyanto; Faldiena Marcelita
Jurnal Informatika dan Teknik Elektro Terapan Vol. 14 No. 3 (2026)
Publisher : Universitas Lampung

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.23960/jitet.v14i3.9838

Abstract

Sistem pencatatan kendaraan manual di gerbang IPB Univesity berpotensi menyebabkan antrean dan kesalahan data. Penelitian ini mengembangkan sistem gerbang otomatis berbasis Automatic License Plate Recognition (ALPR) menggunakan YOLOv11 untuk deteksi plat nomor dan OCR berbasis Compact Convolutional Transformer (CCT) untuk pengenalan karakter. Dataset yang digunakan sebanyak 349 citra plat nomor kendaraan Indonesia dengan rasio 80:20. Fine-tuning model OCR selama 30 epoch berhasil meningkatkan plate accuracy dari 75,71% menjadi 82,86% dengan character accuracy mencapai 98,14%. Metode Consecutive Result juga meningkatkan kestabilan pembacaan OCR pada video real-time. Sistem berhasil diintegrasikan dengan kamera, database kendaraan, dan frontend monitoring untuk identifikasi kendaraan otomatis.
Co-Authors Ahmad Hasan Faqih Aulia Anargya Rabbani Aslam Ariel Mughnika Beers Carissa Fathinah Balti Dwi Putra Kunto Anggoro Fathonah, Lathifunnisa Ghaniyy Fattah Ramadhan Inna Novianty Marcelita, Faldiena Mauladi Fadhillah Mayanda Mega Santoni Muhammad Faiq Fawwaz Nabil Kurnia Rozano Puti Aisyah Lailatulrahmi Santoni, Mayanda Mega Wuliddah Tamsil Barokah Yashin Al Fauzy Sabara Zafira A’idah Gunawan