Prioko, Kentani Langgalih (2020) Convolutional Neural Network Untuk Klasifikasi Citra Asap Pada Gambar Satelit. Other thesis, Institut Teknologi Sepuluh Nopember.
Preview |
Text
07211540000045-Undergraduate_Thesis.pdf Download (5MB) | Preview |
Abstract
Pendeteksian kebakaran dapat dilakukan sedini mungkin dengan bantuan teknologi agar titik api dapat diketahui dengan cepat dan dapat segera dilakukan tindakan lebih lanjut. Pada penelitian ini akan dikembangkan sebuah sistem yang diharapkan dapat mengklasifikasi citra asap pada gambar satelit menggunakan Convolutional Neural Network dengan menggunakan dataset USTC SmokeRS yang berjumlah sebanyak 6225 gambar dan terpisah menjadi 6 kelas. Proses klasifikasi citra akan menggunakan empat macam pre-trained model Convolutional Neural Network yaitu MobileNet V2, Inception V3, InceptionResNet V2, dan NASNet. Hasil pengujian menunjukkan bahwa pre-trained model MobileNet V2, Inception V3, InceptionResNet V2, dan NASNet masing - masing memiliki nilai akurasi sebesar 83,30%, 83,30%, 88,33%, dan 86,67%.
============================================================Fire detection can be done as early as possible with the help of technology so that hotspots can be known quickly and further action can be taken immediately. In this research, a system which is expected to be able to classify smoke images based on satellite images will be developed using Convolutional Neural Network using the USTC SmokeRS dataset, totaling 6225 images and separated into 6 classes. The image classification process will use four types of pre-trained model Convolutional Neural Network which are MobileNet V2, Inception V3, InceptionResNet V2, and NASNet . The results of the testing show that pre-trained model MobileNet V2, Inception V3, InceptionResNet V2, and NASNet respectively - has an accuracy value of 83.30%, 83.30%, 88.33%, and 86.67%.
Item Type: | Thesis (Other) |
---|---|
Uncontrolled Keywords: | fire, smoke, image classification, convolutional neural network, kebakaran, asap, klasifikasi citra, convolutional neural network. |
Subjects: | T Technology > TA Engineering (General). Civil engineering (General) > TA1637 Image processing--Digital techniques |
Divisions: | Faculty of Intelligent Electrical and Informatics Technology (ELECTICS) > Computer Engineering > 90243-(S1) Undergraduate Thesis |
Depositing User: | Kentani Langgalih Prioko |
Date Deposited: | 21 Aug 2020 03:13 |
Last Modified: | 19 Jul 2023 13:26 |
URI: | http://repository.its.ac.id/id/eprint/78957 |
Actions (login required)
View Item |