Claim Missing Document
Check
Articles

Found 2 Documents
Search

Perancangan Sistem Deteksi Peta Panas (Heatmap) Keramaian Pengunjung di Area Publik Selama Pandemi COVID-19 Berbasis YOLOV4-Tiny Al Barra Harahap; Nike Dwi Grevika Drantantiyas; Ismoyo Aji Sasmita
Electrician : Jurnal Rekayasa dan Teknologi Elektro Vol. 19 No. 3 (2025)
Publisher : Department of Electrical Engineering, Faculty of Engineering, Universitas Lampung

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.23960/elc.v19n3.2796

Abstract

Pandemi COVID-19 mengakibatkan masyarakat melakukan aktivitas dengan cara baru yaitu dengan penerapan protokol kesehatan. Pengawasan yang ketat sangat penting dilakukan pada kawasan wisata di area publik untuk menjaga jarak aman dan menghindari kerumuman untuk menekan angka penyebaran virus. Salah satu metode pemantauan aktivitas di ruang terbuka yang dapat digunakan adalah dengan penerapan deep learning. Pada penelitian ini dilakukan perancangan sistem deteksi objek untuk pemetaan kerumunan, kemudian dilakukan penilaian akurasi metode algoritma deteksi objek, dan dilakukan pengujian kemampuan metode algoritma deteksi objek dalam melakukan pemetaan kerumunan. Peta Panas kerumunan dibentuk dengan pendekatan prediktif menggunakan algoritma deteksi objek satu tahap You Only Look Once (YOLO) v4-Tiny berdasarkan pengurangan latar belakang (background subtraction). Pada penelitian ini berhasil dirancang algoritma deteksi objek di area keramaian dengan akurasi 95% dan model akurasi rata-rata 59,45%. Hasil visual pemetaan heatmap kerumunan pengunjung dibentuk dengan warna hitam dan kuning berdasarkan kepadatan arus pengunjung yang melalui suatu lokasi. Warna kuning pada peta di suatu lokasi menandakan pada jalur tersebut terdapat tingkat lalu-lintas manusia yang padat.
Baseline Vapour Pressure Deficit Dynamics in an Emersed Anubias Cultivation Incubator: Sensor Construct Validity and Humidity-Governed VPD Drivers Al Barra Harahap; Vera Khoirunnisa; Aprilia Siregar; Sintya Shety Simanjuntak
International Journal of Electrical Engineering, Mathematics and Computer Science Vol. 3 No. 2 (2026): June: International Journal of Electrical Engineering, Mathematics and Computer
Publisher : Asosiasi Riset Teknik Elektro dan Infomatika Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.62951/ijeemcs.v3i2.390

Abstract

Vapour pressure deficit (VPD) has gained recognition as a physiologically superior microclimate control variable in controlled-environment agriculture, yet no quantitative VPD characterisation exists for emersed cultivation of variegated Anubias barteri var. nana 'Pinto'. This study characterises the VPD microclimate of an ebb-and-flow cyber-physical incubator equipped with an MLX90614 infrared radiometric sensor, two calibrated DHT22 humidity sensors, a DS18B20 nutrient solution sensor, and an ESP32-S3 microcontroller. A 5.36-hour passive commissioning session yielded 1,884 valid data rows at a 10-second logging interval. The mean baseline  was 0.703 ± 0.022 kPa, falling within the upper portion of the reference envelope inferred from available literature on emersed Anubias acclimatisation. Indoor relative humidity was identified as the dominant VPD driver (r = −0.695, R² = 0.483), while air temperature showed negligible association (r = −0.008), supporting mist actuation as the appropriate primary control strategy. The MLX90614 recorded a consistent radiometric offset of ΔT_rad = −1.61 ± 0.10 °C relative to indoor air temperature, reflecting the composite FOV average of cooler evaporative surfaces; the resulting  was 0.327 kPa (32%) lower than the air-temperature-only estimate, demonstrating that the choice of surface temperature reference has a material effect on the computed VPD. A persistent ambient VPD gradient of +0.540 kPa was documented throughout the session, quantifying the continuous moisture gradient from the incubator interior to the ambient environment. These results establish a quantitative VPD baseline for emersed Anubias cultivation, define the construct validity boundaries of the sensor suite, and provide the empirical foundation for a companion closed-loop Fuzzy-PI VPD controller.