Perancangan Dan Implementasi Self-Checkout System Pada Toko Ritel Menggunakan Convolutional Neural Network (Cnn)

Pencarian berdasarkan :

SEMUA Pengarang Subjek ISBN/ISSN Pencarian Spesifik

Pencarian terakhir:

Image of Perancangan Dan Implementasi Self-Checkout System Pada Toko Ritel Menggunakan Convolutional Neural Network (Cnn)

CD Skripsi

Perancangan Dan Implementasi Self-Checkout System Pada Toko Ritel Menggunakan Convolutional Neural Network (Cnn)

Rahma Adi Putra / 1807124728 - Nama Orang;

ABSTRACT
The payment system at retail stores today generally still uses the traditional cashier system. The cashier must scan the barcodes one by one on each grocery product, so other consumers will have to wait quite a long time if there is only one cashier available. Currently, the self-checkout system method has begun to be developed. Self-checkout system is a facility that allows consumers to make payments and pack their own groceries without the help of a cashier. The system is more efficient than traditional cashiers since it is not operated by workers. The design of this self-checkout system uses the convolutional neural network (CNN) method as an image data processor. The MobileNetV2 model architecture from CNN was chosen because it has a high accuracy value and a low number of computations. Hamming loss method is used to evaluate and to determine the performance of the model made in the Multi-label classification type. The results of the model that has been trained will be implemented in the GUI as a user interface. 247 images with image resolution of 224 x 224 pixels from the products: Teh Botol, Indomie, and Chitato is used as the dataset. The results of the training model in this study obtained an accuracy value of 88.8%. The hamming loss value gets a loss of 0.12%. Processing time on the GUI system to detect grocery products on average takes 1 second. The placement position of each product must be spaced, because it greatly affects the detection results.
Keywords : Self-Checkout System, Convolutional neural network, MobileNetV2

Ketersediaan

Perpustakaan Universitas Riau 1807124728

1807124728

Tersedia

Informasi Detail

Judul Seri: -
No. Panggil: 1807124728
Penerbit: Pekanbaru : Universitas Riau – Fakultas Teknik – Teknik Elektro., 2023
Deskripsi Fisik: -
Bahasa: Indonesia
ISBN/ISSN: -
Klasifikasi: 1807124728
Tipe Isi: -
Tipe Media: -
Tipe Pembawa: -
Edisi: -
Subjek: TEKNIK ELEKTRO
Info Detail Spesifik: -
Pernyataan Tanggungjawab: DAUS

Perancangan Dan Implementasi Self-Checkout System Pada Toko Ritel Menggunakan Convolutional Neural Network (Cnn)

Ketersediaan

Informasi Detail

Versi lain/terkait

Lampiran Berkas

Komentar