PENGGUNAAN CONVOLUTIONAL NEURAL NETWORK DALAM IDENTIFIKASI BAHAN KULIT SAPI DAN BABI DENGAN TENSORFLOW

• Main Author: Ellvina Reksi Hardyanti, (Author)
• Format: Book
• Published: 2020-07-07.
• Subjects: Thesis; NonPeerReviewed
• Online Access: Link Metadata

Summary

Cowhide material is one type of material that is much in demand by the wider community. The increasing industrial demand for leather is also less able for a buyer to identify leather products on the market. Given these problems, a solution is needed to help buyers identify leather. This research will use Convolutional Neural Network (CNN) which is part of Deep Learning with the help of TensorFlow to conduct the learning process so that it can detect images of cow and pig skin material. The dataset used was 190 images of cowhide and 158 images of pigskin. Convolution is the main process that exists in the CNN architecture network, so that each image can be extracted better and simplify the classification process. The model used in the study is the best model chosen from 6 experiments by researchers and the division of training and testing data is 75% and 25%. The best results are obtained with accuracy rates as high as 100% and loss 0.000012393 with epoch 100, learning rate 0.001, and batch size 32.

Item Description

http://repository.upnvj.ac.id/6747/1/ABSTRAK.pdf
http://repository.upnvj.ac.id/6747/2/AWAL.pdf
http://repository.upnvj.ac.id/6747/3/BAB%201.pdf
http://repository.upnvj.ac.id/6747/4/BAB%202.pdf
http://repository.upnvj.ac.id/6747/5/BAB%203.pdf
http://repository.upnvj.ac.id/6747/6/BAB%204.pdf
http://repository.upnvj.ac.id/6747/7/BAB%205.pdf
http://repository.upnvj.ac.id/6747/8/DAFTAR%20PUSTAKA.pdf
http://repository.upnvj.ac.id/6747/9/RIWAYAT%20HIDUP.pdf
http://repository.upnvj.ac.id/6747/10/LAMPIRAN.pdf
http://repository.upnvj.ac.id/6747/11/ARTIKEL%20KI.pdf