Claim Missing Document
Check
Articles

Found 13 Documents
Search

DISIMINASI PENGGUNAAN POINT OF PRESENCE TERHADAP LAYANAN JARINGAN DISTRIBUSI PELANGGAN BERBASIS ARTIFICIAL INTELLIGENT R.A Halimatussa’diyah; Sholihin Sholihin; Ali Nurdin; Heri Liamsi; Eka Susanti; Siswandi Siswandi
Martabe : Jurnal Pengabdian Kepada Masyarakat Vol 9, No 8 (2026): MARTABE : Jurnal Pengabdian Kepada Masyarakat
Publisher : Universitas Muhammadiyah Tapanuli Selatan

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.31604/jpm.v9i8.3051-3062

Abstract

Salah satu tantangan sebagai anak bangsa adalah mengembangkan dan membuat alat untuk membantu layanan kesehatan agar dapat berkembang mengikuti dengan perkembangan teknologi, yaitu dengan membuat alat pendeteksi diabetes dengan artificial intelligence.Permasalahan pada pengguna memiliki keterbatasan dalam membedakan warna,baik karena factor pencahayaan, gangguan penglihatan, atau perbedaan persepsi warna.Teknologi image processing hadir sebagai solusi dengan memanfaatkan kamera untuk menangkap citra strip, kemudian diproses secara digital agar hasil yang ditampilkan menjadi lebih akurat dan konsisten.Pengabdian ini bertujuan menjelaskan atau mendisiminasi kepada mitra PT.Latansa Palembang Jaya untuk prototipe sistem deteksi gangguan berbasis Raspberry Pi yang terintegrasi dengan teknologi Internet of Things dan kecerdasan buatan. Sistem yang dirancang menggunakan sensor ultrasonik, sensor MQ-2, dan AI Camera untuk mendeteksi berbagai kondisi gangguan. Data hasil deteksi kemudian dikirim dan ditampilkan secara real-time melalui dashboard Node-RED serta dilengkapi buzzer sebagai alarm peringatan.Hasil pengujian menunjukkan bahwa sistem mampu mendeteksi gangguan dan menampilkan informasi pemantauan secara real-time sesuai dengan fungsi yang dirancang. Dengan demikian, prototipe ini dapat menjadi solusi pendukung dalam meningkatkan efektivitas monitoring dan keamanan pada Point of Presence.Dari hasil pengujian yang telah dilakukan, sensor ultrasonik mampu mendeteksi keberadaan objek di area pemantauan dengan cukup baik. Sensor tetap dapat memberikan pembacaan yang stabil pada berbagai jarak pengujian sehingga dapat dimanfaatkan sebagai salah satu indikator adanya aktivitas di sekitar perangkat PoP. Sementara itu, sensor MQ-2 juga menunjukkan kemampuan yang baik dalam mendeteksi keberadaan asap atau gas yang berpotensi mengganggu operasional perangkat telekomunikasi.Proses komunikasi data antara perangkat dan dashboard berlangsung dengan lancar melalui protokol MQTT
Rancang Bangung Sistem Absensi Guru Berbasis Webcam Menggunakan Haar Cascade dan Local Binary patterns Histogram untuk Pengenalan Wajah Nurhajar Anugraha; Dea Ardyanti; Ali Nurdin
Jurnal Minfo Polgan Vol. 15 No. 2 (2026): Artikel Penelitian
Publisher : Politeknik Ganesha Medan

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33395/jmp.v15i2.16419

Abstract

The teacher attendance system, which is still done manually, has many shortcomings, such as the ease with which data can be changed, errors when recording, and the slow process of calculating attendance. This research aims to design and create a teacher attendance system that uses a webcam with the Haar Cascade method and Local Binary Pattern Histogram (LBPH) to recognize faces. The Haar Cascade method is used to detect faces directly, while LBPH is used to recognize who the teacher is based on facial characteristics that have been stored in the data. This system was created using the Python programming language to process images and the Laravel framework as a web application platform. Existing features include teacher data management, facial data management, attendance using facial recognition, permit applications, absenteeism calculations, and attendance reports. The results of the system show that it can detect and recognize teachers' faces automatically, so that recording attendance is faster, more precise and efficient compared to the usual method. In addition, attendance data can be stored in one place and displayed in the form of statistics and reports that are helpful in school administration. In this way, the system created can improve the management of teacher absences by using facial recognition technology.
Perancangan Sistem Smart Home Berbasis Iot Menggunakan Komunikasi LoRa Dan Algoritma Fuzzy Logic Raihan Aldiantino; Ali Nurdin; Nurhajar Anugraha
Jurnal Minfo Polgan Vol. 15 No. 3 (2026): Artikel Penelitian
Publisher : Politeknik Ganesha Medan

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33395/jmp.v15i3.16420

Abstract

The development of smart home systems has increased due to the Internet of Things' (IoT) rapid growth. However, most existing solutions rely on Wi-Fi or Bluetooth, which have limited communication ranges and relatively high power consumption. In addition, automation is often restricted to simple threshold-based ON/OFF control. This study proposes a smart home system that integrates LoRa (Long Range) communication with a Mamdani Fuzzy Logic algorithm to provide reliable long-range communication and adaptive control. The system adopts a hybrid communication architecture consisting of an online mode using Wi-Fi connected to the Adafruit IO Cloud platform and an offline mode using LoRa for local communication through a gateway. Two ESP32 microcontrollers are employed: a transmitter node equipped with DHT22 temperature and humidity, MQ-2 gas, and ZMPT101B voltage sensors, and a receiver node acting as a gateway with relay-controlled lighting and exhaust fan actuators. The Fuzzy Logic controller uses temperature and humidity as input variables and generates a PWM duty cycle output to regulate fan speed through nine IF–THEN rules. According to experimental findings, LoRa communication was able to achieve transmission delays of 2.20–8.00 seconds between 0 and 100 meters; however, at 110 meters, the delay increased to 22.50 seconds, indicating the effective communication limit. The Fuzzy Logic algorithm achieved 100% rule-base compliance in all test scenarios. A Node-RED dashboard successfully displayed sensor data in real time. The proposed system provides a reliable, energy-efficient, and adaptive smart home solution that outperforms conventional Wi-Fi-based approaches