Claim Missing Document
Check
Articles

Found 26 Documents
Search

TRAFFIC FLOW DETECTION USING YOLOV4 AND DEEPSORT ON NVIDIA JETSON NANO Taufiq, Reny Medikawati; Syahril, Syahril; Rafdi, Faris Abi; Firdaus, Rahmad; Sunanto, Sunanto; Muarif, Putri Fadhilla
JURTEKSI (jurnal Teknologi dan Sistem Informasi) Vol. 11 No. 3 (2025): Juni 2025
Publisher : Lembaga Penelitian dan Pengabdian Kepada Masyarakat (LPPM) STMIK Royal Kisaran

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33330/jurteksi.v11i3.3871

Abstract

Abstract: This study aims to develop a Deep Learning-based Traffic Flow Detector to automatically and accurately observe traffic flow. Conventional traffic observation is often conducted manually or via CCTV, but it is prone to human error and difficult to use for real-time trend analysis. In this study, the YOLOv4 method is used to detect four types of vehicles (cars, motorcycles, buses, trucks). To continuously track vehicle movement and address occlusion issues, the Deep SORT algorithm is implemented. The YOLOv4 model used is a pre-trained model and was tested on seven CCTV video recordings obtained from the official website of the Pekanbaru City Transportation Department. The system was implemented on a limited device, the Nvidia Jetson Nano, as a simulation of direct CCTV integration. Test results showed a highest precision of 98%, but the maximum accuracy achieved was only 26%. This low accuracy is influenced by several factors, including video resolution, detection model quality, and lighting conditions. Nevertheless, the system demonstrates potential to support future traffic management and engineering decisions but still requires further optimization, including improving video resolution and quality, retraining the model with a more representative local dataset, using lighter and more accurate detection models, and optimizing the tracking algorithm. Keywords: deep learning; deepsort; NVIDIA Jetson NANO; traffic flow; YOLOv4  Abstrak: Penelitian ini bertujuan mengembangkan Traffic Flow Detector berbasis Deep Learning untuk mengobservasi arus lalu lintas secara otomatis dan akurat. Observasi lalu lintas konvensional sering dilakukan secara manual atau melalui CCTV, namun rentan terhadap human error dan sulit digunakan untuk menganalisis tren secara real-time. Pada penelitian ini digunakan metode YOLOv4 untuk mendeteksi empat jenis kendaraan (mobil, motor, bus, truk). Untuk melacak pergerakan kendaraan secara berkelanjutan dan mengatasi masalah occlusion, digunakan algoritma Deep SORT. Model YOLOv4 yang digunakan merupakan pre-trained model dan diujikan pada tujuh rekaman video CCTV yang diambil dari situs resmi Dinas Perhubungan Kota Pekanbaru. Sistem ini diimplementasikan pada perangkat terbatas Nvidia Jetson Nano sebagai simulasi penerapan langsung pada CCTV. Hasil pengujian menunjukkan presisi tertinggi mencapai 98%, namun akurasi tertingginya hanya sebesar 26%. Rendahnya akurasi dipengaruhi oleh beberapa faktor seperti resolusi video, kualitas model deteksi, serta kondisi pencahayaan. Meski demikian, sistem ini menunjukkan potensi untuk membantu pengambilan keputusan dalam manajemen dan rekayasa lalu lintas di masa depan, namun masih membutuhkan optimasi lebih lanjut, seperti  peningkatan kualitas video input, pelatihan ulang model dengan dataset lokal, penggunaan model deteksi yang lebih ringan dan akurat serta pengoptimalan algoritma pelacakan. Kata kunci: deep learning deepsort; Nvidia Jetson Nano; traffic flow; YOLOv4
Klasifikasi Citra Penyakit Daun Tomat Menggunakan Metode Convolutional Neural Network (CNN) Dengan Arsitektur VGG-19 Fitri Handayani; Baidarus, Baidarus; Sunanto, Sunanto; Putra, Bayu Anugerah; Anggraini, Chelina; Taufiq, Reny Medikawati
Computer Science and Information Technology Vol 6 No 3 (2025): Jurnal Computer Science and Information Technology (CoSciTech)
Publisher : Universitas Muhammadiyah Riau

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.37859/coscitech.v6i3.10699

Abstract

Tomatoes, known as Solanum lycopersicum in Latin, are a type of horticultural commodity with high economic value in Indonesia.Tomato production can decrease due to leaf diseases that are hard to identify manually because the symptoms of different diseases often appear similar. The purpose of this study is to apply a deep learning-based tomato leaf disease classification system using the Convolutional Neural Network (CNN) VGG-19 architecture. The dataset was obtained from Kaggle and contains 6,600 images of tomato leaves divided into six disease classes and one healthy leaf class. The research stages include preprocessing (resizing, normalization), data augmentation, dataset division (80% training, 20% testing), model training with transfer learning, and fine-tuning for optimization. The evaluation using the confusion matrix and classification report includes accuracy, precision, recall, and F1-score. Test results show that the VGG-19 model achieved 97% accuracy on the test data, with an average precision, recall, and F1-score of 0.97. These findings show that VGG-19 effectively identifies tomato leaf diseases and could be applied in web- or mobile-based detection systems to help farmers with early diagnosis and proper treatment.
A Systematic Review of AIoT in Agricultural Environmental Monitoring: A Comparative Analysis of Machine Learning Approaches Kharisma, Derry Artha; Sunanto, Sunanto
ILKOMNIKA Vol 8 No 1 (2026): Volume 8, Number 1, April 2026
Publisher : Lembaga Penelitian dan Pengabdian Masyarakat

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.28926/ilkomnika.v8i1.865

Abstract

Agricultural water management increasingly deploys AIoT systems that integrate IoT sensing with machine learning, yet deployable autonomous irrigation control remains largely unrealized despite widely reported accuracy above 90%. This systematic literature review of 25 peer-reviewed studies (2020–2026), conducted across five databases following PRISMA 2020 guidelines, diagnoses why predictive performance fails to translate into operational autonomy. The analysis identifies six interdependent structural gaps: open-loop prediction architectures, informationally narrow sensing, correlated co-sensor packaging, static non-adaptive models, accuracy–deployability decoupling, and metric inconsistency. These gaps form a dependency chain across data, inference, and actuation layers in which closed-loop integration depends on resolving data adequacy. A parallel finding reveals systematic methodological divergence between national and international research contexts, driven by infrastructure and deployment constraints rather than research quality, with reinforcement learning, hybrid multi-modal architectures, and continual learning largely absent in national studies. This study contributes a reframing of AIoT system maturity by demonstrating that within-study accuracy is misaligned with operational validity. It further establishes environmental generalizability as a more appropriate evaluation criterion, shows that the six structural gaps form a sequential dependency structure that prevents single-gap solutions from producing deployable improvement, and provides directional evidence that reported accuracy and validation scope are inversely related across the corpus, suggesting that current performance claims systematically overstate operational readiness.
EFEKTIVITAS PENERAPAN INTERNET MARKETING 4.0 TERHADAP PEMESANAN TIKET TRAVEL PEKANBARU DURI DUMAI Fauzan Azim; Ikhbal Akhmad; Hadi Purwanto; Khairul Anshari; Sunanto Sunanto
Rabit : Jurnal Teknologi dan Sistem Informasi Univrab Vol 6 No 1 (2021): Januari
Publisher : LPPM Universitas Abdurrab

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.36341/rabit.v6i1.1531

Abstract

This research was conducted at PT. Jasa Mulya Trans Gemilang, a company engaged in transportation services by routes of Pekanbaru-Duri-Dumai. PT. Jasa Mulya Trans Gemilang has two websites that are quite good on the search engines of Google: Bing and Yahoo. It also carries out Advertising Promotions through Social Media and Google Ads which have been running for a long time and until the time this research was conducted, there was no measurement of how well the effectiveness of Internet Marketing that has been running at PT. Jasa Mulya Trans Gemilang. Measuring the Effectiveness of an Internet Marketing Program is indispensable for consideration and promotional strategies that can be carried out for the future by the company. One of the ways to measure the effectiveness of Internet Marketing Utilization was the EPIC Model. In the EPIC Model, there were four variables which were investigated to measure effectiveness, namely Empathy, Persuasion, Impact and Communication. Respondents in this study were the people of Pekanbaru and its surroundings who ordered travel tickets with destination of Duri and Dumai Cities. After conducting this research, it was obtained that the effectiveness levels of Internet Marketing Application of travel ticket reservations for Pekanbaru-Duri-Dumai which was measured in four EPIC dimensions were as follows: the value of Emphaty was 4.09, the Persuasion was 4.11, the Impact was 3.88, and the Communication was 3.99. From the four EPIC dimensions, the average result of the EPIC Model or EPICrate was 4.02 that could be interpreted on an effective scale. This means that in this research, the implementation of Internet Marketing at PT. Jasa Mulya Trans Gemilang through the internet consisting of Website, Social Media and Google Ads was effective.
PENERAPAN ALGORITMA C4.5 UNTUK MEMBUAT MODEL PREDIKSI PASIEN YANG MENGIDAP PENYAKIT DIABETES Nanto Sunanto; Ghazi Falah
Rabit : Jurnal Teknologi dan Sistem Informasi Univrab Vol 7 No 2 (2022): Juli
Publisher : LPPM Universitas Abdurrab

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.36341/rabit.v7i2.2435

Abstract

Diabetes is a disease in which the pancreas cannot produce insulin properly. Insulin is a hormone produced from the pancreas, which is useful as a door to channel glucose from food that is absorbed to be flowed into blood cells so that the body can produce energy. Meanwhile, according to WHO diabetes is a very deadly disease which is ranked 9th in the world. People with diabetes generally die from damage to several vital organs such as the heart, kidneys and liver. The early sufferers of diabetes are not known with certainty so that diabetics who have been treated are in a worrying condition. To reduce the number of deaths due to diabetes, a system is needed that can identify early symptoms of diabetes, so that people with diabetes can be handled properly. Data mining technology can help build a system to predict diabetes using the C4.5 decision tree algorithm. In this study, diabetes prediction data was taken from the UCI Repository. Then the data is processed in stages, select data, pre-processing and split validation using rapid miner. The results of data processing using a rapid miner, in the form of rules that can be used to predict diabetes. The rules generated from the RapidMiner decision tree have an accuracy of 95.51%.
RANCANG BANGUN ALAT UKUR KADAR GULA DALAM DARAH NON INVASIVE MENGGUNAKAN ESP32 DAN BLYNK: DESIGN A NON-INVASIVE BLOOD SUGAR MEASUREMENT USING ESP32 AND BLYNK Vadel Amanika; Sunanto S.Kom,M.Kom
Rabit : Jurnal Teknologi dan Sistem Informasi Univrab Vol 9 No 2 (2024): Juli
Publisher : LPPM Universitas Abdurrab

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.36341/rabit.v9i2.4503

Abstract

The compatibility of ESP32 microcontrollers in Internet of Things (IoT) implementations covers almost all fields. Has the potential to monitor blood sugar levels non-invasively with the presence of Fotodioda Sensors and Infared LED lights and fingers as measurement objects without the need to injure patients or invasively in the aim to improve public health services, especially in an effort to prevent diabetes mellitus. ESP32 microcontroller also has compatibility in the use of Blynk which can be used to be a medium to monitor blood sugar levels and can be displayed on a computer monitor or smartphone by opening Blynk private, also with the utilization of Light intensity using fotodioda sensors and LED lights will be designed Non-Invasive Blood Sugar Measuring Instruments Using Prototyping Methods by utilizing ESP32 and Blynk and conducted experiments to see performance and comparison of accuracy with Glucometers in general and utilization in applications in the Health sector. In the prototype experiment of the Blood Glucose Measurement Device, the light absorption calculation produced by the combination of photodiode sensors and LED lamps can be utilized for measuring blood glucose levels without the need for direct blood sampling from the user. The implementation is straightforward and includes a monitoring system using Blynk, facilitating the process of blood glucose monitoring via both smartphones and users' computers. The overall accuracy result of the non-invasive blood glucose measurement device prototype reached 95% based on multiple conducted experiments.