A PHP Error was encountered
Severity: Warning
Message: fopen(/home/rama/public_html/application/cache/sessions/rama_sessiontq5b76ol3g6lkbe0e9kk4ls7v3oj3cnq): failed to open stream: No space left on device
Filename: drivers/Session_files_driver.php
Line Number: 174
Backtrace:
File: /home/rama/public_html/application/core/MY_Controller.php
Line: 8
Function: __construct
File: /home/rama/public_html/application/controllers/Document.php
Line: 6
Function: __construct
File: /home/rama/public_html/index.php
Line: 296
Function: require_once
@thesis{thesis,
author={Susilawati Susilawati},
title ={Mengidentifikasi Material Batu dan Ranting Menggunakan Convolutional Neural Network},
year={2022},
url={http://repository.uki.ac.id/9262/},
abstract={Pendeteksian objek material arang yang tercampur berguna bagi masyarakat yang ingin membeli arang, sehingga membantu pembeli dalam pemilihan secara modern lewat pendeteksian. Penggunaan metode Convolutional neural networks sangat cepat berkembang dalam pendeteksian atau pengenalan objek, CCN memiliki beberapa banyak arsitektur, salah satunya yaitu Faster R-CNN yang berkembang dari arsitektur R-CNN dan Fast R-CNN. Sehingga pada pendeteksian ini menggunakan arsitektur Fast R-CNN. Pendeteksian menggunakan data input berupa gambar dengan data dibagi menjadi data train 80% dan data test 20%, akurasi yang didapatkan sangat baik yaitu 99%, dimana menggunakan beberapa kali percobaan yaitu adanya perubahan parameter dan penambahan data input, sehingga dibutuhkan parameter yang membuat loss function terus semakin berkurang saat proses training./ The object detector of sing charcoal can be used by people who want charcoal, so that it can help buyers in selecting modern kanthi through detection. The use of the Convolutional neural networks method is growing very fast in object detection or recognition, CCN has several architectures, one of which is Faster R-CNN which developed from the R-CNN and Fast R-CNN architecture. In order to detect this using Fast R-CNN architecture. This Panaliten hopes to take charcoal buyers and make it easy to find material objects and twigs, the detector uses input data, such as data that is divided into 80% train data and 20% test data, the accuracy is very high, namely 99%, sings using rice cakes Tried times, namely, I tried to use hyperparameters and added input data, so that I needed to use the loss function parameters, so I have checked the training process.}
}