Please use this identifier to cite or link to this item: http://repository.ipb.ac.id/handle/123456789/155217
Title: Sistem Prediksi Nilai Kekeruhan Air Secara Real-Time Menggunakan ESP32-CAM dengan Platform Roboflow
Other Titles: Real-Time Water Turbidity Value Prediction System using ESP32-CAM with Roboflow Platform
Authors: Novianty, Inna
Fathurahman, Farhan
Issue Date: 2024
Publisher: IPB University
Abstract: Kualitas air dalam budi daya ikan merupakan hal penting untuk menjaga keberlangsungan hidup ikan. Kekeruhan salah satu karakter fisik air yang memegang peranan penting dalam penentuan kualitas air. Kekeruhan dalam air diukur dengan satuan Nephelometric Turbidity Units (NTU). Sensor yang dapat digunakan untuk memeriksa kekeruhan air adalah sensor turbidity/Total Suspended Solid (TSS). Akan tetapi, sensor tersebut mudah dimasuki oleh air sehingga menjadi rusak. Oleh karena itu, dibuatlah suatu alternatif sensor tersebut, yaitu dengan ESP32-CAM serta algoritme computer vision. ESP32-CAM bertindak sebagai kamera yang menangkap gambar permukaan air yang ada di kolam atau akuarium. Setelah itu, ESP32-CAM akan mengirim gambar menuju server untuk diproses dengan algoritme computer vision yang dibuat dengan platform Roboflow berupa model multi-label classification. Akurasi model multi-label classification didapatkan sebesar 98,9%. Hasil keluaran model tersebut adalah nilai kisaran kekeruhan air dengan satuan NTU.
Water quality in fish farming is important to maintain the survival of fish. Turbidity is one of the physical characteristics of water that plays an important role in determining water quality. Turbidity in water is measured in Nephelometric Turbidity Units (NTU). The sensor that can be used to check water turbidity is the turbidity sensor/ Total Suspended Solid (TSS). However, water can easily penetrate the sensor and become damaged. Therefore, an alternative sensor was created, namely the ESP32-CAM and the computer vision algorithm. The ESP32-CAM acts as a camera that captures images of the water surface in the pond or aquarium. After that, the ESP32-CAM will send the image to the server for processing with the computer vision algorithm created with the Roboflow platform in the form of a multi-label classification model. Accuracy of the multi-label classification model is 98.9%. Output of the model is the value of the water turbidity range in NTU units.
URI: http://repository.ipb.ac.id/handle/123456789/155217
Appears in Collections:UT - Computer Engineering Tehcnology

Files in This Item:
File Description SizeFormat 
cover_J0304201033_54e35a585f624b2aa0d3364db2ddd7af.pdfCover5.64 MBAdobe PDFView/Open
fulltext_J0304201033_d0df20dd2c95416083ac604bdbb244e5.pdf
  Restricted Access
Fulltext5.46 MBAdobe PDFView/Open
lampiran_J0304201033_8e28ea44629541ba891d7a67b74e24b2.pdf
  Restricted Access
Lampiran496.89 kBAdobe PDFView/Open


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.