KLASIFIKASI OBJEK MINUMAN KALENG PADA MINIMARKET DENGAN METODE CNN-SINGLE SHOT MULTIBOX DETECTOR (SSD)

SAPUTRA, PRAMUDYA (2023) KLASIFIKASI OBJEK MINUMAN KALENG PADA MINIMARKET DENGAN METODE CNN-SINGLE SHOT MULTIBOX DETECTOR (SSD). S1 thesis, Universitas Mercu Buana Jakarta-Menteng.

[img]
Preview
Text (Cover)
41519110177-PRAMUDYA SAPUTRA-01 Cover - Pramudya Saputra.pdf

Download (364kB) | Preview
[img]
Preview
Text (Abstrak)
41519110177-PRAMUDYA SAPUTRA-02 Abstrak - Pramudya Saputra.pdf

Download (116kB) | Preview
[img] Text (Bab 1)
41519110177-PRAMUDYA SAPUTRA-03 Bab 1 - Pramudya Saputra.pdf
Restricted to Registered users only

Download (143kB)
[img] Text (Bab 2)
41519110177-PRAMUDYA SAPUTRA-04 Bab 2 - Pramudya Saputra.pdf
Restricted to Registered users only

Download (364kB)
[img] Text (Bab 3)
41519110177-PRAMUDYA SAPUTRA-05 Bab 3 - Pramudya Saputra.pdf
Restricted to Registered users only

Download (145kB)
[img] Text (Bab 4)
41519110177-PRAMUDYA SAPUTRA-06 Bab 4 - Pramudya Saputra.pdf
Restricted to Registered users only

Download (712kB)
[img] Text (Bab 5)
41519110177-PRAMUDYA SAPUTRA-07 Bab 5 - Pramudya Saputra.pdf
Restricted to Registered users only

Download (91kB)
[img] Text (Daftar Pustaka)
41519110177-PRAMUDYA SAPUTRA-09 Daftar Pustaka - Pramudya Saputra.pdf
Restricted to Registered users only

Download (102kB)
[img] Text (Lampiran)
41519110177-PRAMUDYA SAPUTRA-10 Lampiran - Pramudya Saputra.pdf
Restricted to Registered users only

Download (2MB)
[img] Text (Formulir pernyataan keaslian dan persetujuan unggah dan publikasi tugas akhir)
41519110177-PRAMUDYA SAPUTRA- Form Pernyataan Keaslian dan Persetujuan Unggah dan Publikasi Tugas Akhir - Pramudya Saputra.pdf
Restricted to Registered users only

Download (221kB)

Abstract

Teknologi deteksi objek, dalam pemrosesan gambar dan visi komputer, telah menjadi perhatian di Indonesia, terutama di sektor Barang Konsumsi Cepat (FMCG). Namun, tantangan muncul ketika produk meniru desain pemimpin pasar, mengakibatkan kesalahan pemilihan oleh konsumen yang kurang berhatihati. Teknologi deteksi objek dapat membantu mengatasi tantangan ini dengan mengenali varian produk dengan akurat dan efisien, memastikan kemasan yang sesuai dengan jenis minuman yang benar.Penelitian ini bertujuan untuk mengimplementasikan model deteksi objek menggunakan algoritma CNNSingle Shot Multibox Detector (SSD) khususnya untuk deteksi objek minuman kaleng berdasarkan klasifikasi jeni kopi pada brand Nescafe.Penelitian ini melibatkan tahapan pemrosesan gambar menggunakan algoritma CNN-SSD, ekstraksi fitur, pelabelan data, dan evaluasi kinerja model. Faktor-faktor yang mempengaruhi dalam mendeteksi objek minuman kaleng termasuk kualitas data pelatihan, arsitektur model, dan parameter pelatihan. Hasil penelitian menunjukkan bahwa implementasi model dengan algoritma CNN-SSD dapat menghasilkan deteksi dan klasifikasi objek yang akurat untuk minuman kaleng. Evaluasi kinerja model mengungkap tingkat presisi, recall, dan F1-score yang baik untuk setiap kelas minuman. Kata Kunci: Single Shot Multibox Detector, deteksi objek, minuman kaleng, klasifikasi objek. Object detection technology, in image processing and computer vision, has been gaining attention in Indonesia, especially in the Fast Consumer Goods (FMCG) sector. However, challenges arise when products mimic the designs of market leaders, resulting in selection errors by less cautious consumers. Object detection technology can help overcome these challenges by accurately and efficiently recognizing product variants, ensuring packaging that corresponds to the correct beverage type. This study aims to implement an object detection model using the CNN-Single Shot Multibox Detector (SSD) algorithm specifically for canned beverage object detection based on coffee type classification in the Nescafe brand. This study involves the stages of image processing using the CNN-SSD algorithm, feature extraction, data labeling, and model performance evaluation. The influencing factors in detecting canned beverage objects include training data quality, model architecture, and training parameters. The results show that the implementation of the model with CNN-SSD algorithm can produce accurate object detection and classification for canned beverages. Evaluation of the model performance reveals good precision, recall, and F1-score levels for each beverage class. Keywords: Single Shot Multibox Detector, object detection, canned beverages, object classification.

Item Type: Thesis (S1)
NIM/NIDN Creators: 41519110177
Uncontrolled Keywords: Kata Kunci: Single Shot Multibox Detector, deteksi objek, minuman kaleng, klasifikasi objek. Keywords: Single Shot Multibox Detector, object detection, canned beverages, object classification.
Subjects: 000 Computer Science, Information and General Works/Ilmu Komputer, Informasi, dan Karya Umum > 000. Computer Science, Information and General Works/Ilmu Komputer, Informasi, dan Karya Umum > 004 Data Processing, Computer Science/Pemrosesan Data, Ilmu Komputer, Teknik Informatika
Divisions: Fakultas Ilmu Komputer > Informatika
Depositing User: SITI NOVI NUR CAHYANI
Date Deposited: 14 Sep 2023 07:14
Last Modified: 14 Sep 2023 07:14
URI: http://repository.mercubuana.ac.id/id/eprint/80847

Actions (login required)

View Item View Item