Shadiq, Ja'far (2024) Deteksi Orang Pada Ruang Perkuliahan Menggunakan Algoritma FOMO Pada ESP32. Other thesis, Institut Teknologi Sepuluh Nopember.
![]() |
Text
07211940000023-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 October 2026. Download (9MB) | Request a copy |
Abstract
Teknologi informasi yang ada saat ini berkembang pesat, dan dapat menawarkan peningkatan efisiensi di berbagai bidang. Dalam lingkup smart building, peningkatan efisiensi yang memungkinkan diantaranya kemudahan manajemen, penghematan biaya, peningkatan kelestarian lingkungan, dan lainnya. Penelitian ini bertujuan untuk mengetahui cara mendeteksi okupansi, dan mengembangkan alat yang dapat mendeteksi ketersediaan ruang berdasarkan okupansi menggunakan kamera. Sistem ini akan dilengkapi dengan teknologi deep learning, sehingga kamera dapat mendeteksi ada tidaknya orang dalam ruangan. Hasil dari deteksi ditampilkan dalam sebuah web yang menyajikan antarmuka keadaan ruangan. Sistem ini diharapkan dapat dipasang di gedung perkuliahan atau kantor dosen. Sistem yang dikembangkan ini memiliki potensi untuk meningkatkan optimalisasi penggunaan ruang yang efektif, penghematan biaya, peningkatan kelestarian lingkungan, dan lainnya.
============================================================
Information technology that exists today is growing rapidly and can offer increased efficiency in various fields. Within the scope of smart building, possible efficiency improvements include ease of management, cost savings, increased environmental sustainability, and others. This study aims to find out how to detect occupancy and to develop a tool that can detect space availability based on occupancy using a camera. This system will be equipped with deep learning technology so that the camera can detect whether or not there are people in the room. This system is expected to be installed in lecture buildings or lecturer offices. This system has the potential to improve the optimization of effective use of space, cost savings, increased environmental sustainability, and others. The result shows that at 192x192 resolution s estimated to use
1 MB of RAM and 94.4 KB of flash memory, with F1-score of 95.2%. The model is estimated to perform inference in 4092ms. Even though it takes long time to perform inference this is negligible, since the inference is only performed every once in a while.
Item Type: | Thesis (Other) |
---|---|
Uncontrolled Keywords: | edge cmputing, ESP32, internet of things, occupancy, smart building,okupansi TinyML |
Subjects: | T Technology > TA Engineering (General). Civil engineering (General) > TA1637 Image processing--Digital techniques T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK5105.546 Computer algorithms |
Divisions: | Faculty of Intelligent Electrical and Informatics Technology (ELECTICS) > Computer Engineering > 90243-(S1) Undergraduate Thesis |
Depositing User: | Ja'far Shadiq |
Date Deposited: | 09 Aug 2024 08:24 |
Last Modified: | 09 Aug 2024 08:24 |
URI: | http://repository.its.ac.id/id/eprint/111394 |
Actions (login required)
![]() |
View Item |