A SCALABLE RETINANET-CNN ARCHITECTURE FOR AUTOMATED OIL PALM BUNCH CLASSIFICATION AND DETECTION

ALFARIZ, MUHAMMAD ALKAM (2025) A SCALABLE RETINANET-CNN ARCHITECTURE FOR AUTOMATED OIL PALM BUNCH CLASSIFICATION AND DETECTION. S1 thesis, Universitas Mercu Buana Jakarta.

Preview	Text (HAL COVER) 01 COVER.pdf Download (319kB) | Preview
	Text (BAB I) 02 BAB 1.pdf Restricted to Registered users only Download (40kB)
	Text (BAB II) 03 BAB 2.pdf Restricted to Registered users only Download (296kB)
	Text (BAB III) 04 BAB 3.pdf Restricted to Registered users only Download (239kB)
	Text (BAB IV) 05 BAB 4.pdf Restricted to Registered users only Download (254kB)
	Text (BAB V) 06 BAB 5.pdf Restricted to Registered users only Download (26kB)
	Text (DAFTAR PUSTAKA) 07 DAFTAR PUSTAKA.pdf Restricted to Registered users only Download (150kB)
	Text (LAMPIRAN) 08 LAMPIRAN.pdf Restricted to Repository staff only Download (764kB)

Abstract

The classification of palm oil ripeness is vital in maximizing yield and quality in the palm oil industry. This study introduces a scalable framework employing a modified RetinaNet-CNN architecture for automated oil palm bunch classification and detection. The framework emphasizes the use of deep learning techniques to achieve accurate classification, addressing the traditional reliance on manual assessments that are often subjective and labor-intensive. By assembling a comprehensive dataset of high-resolution images of oil palm fruit at various ripeness stages, this research ensures that the training process is wellinformed and applicable to real-world scenarios. The proposed model demonstrates impressive performance, achieving a mean Average Precision (mAP) of 83.6% and a high F1-score of 98.3%. Notably, the model exhibits a robust training process with a significant reduction in training loss, indicating effective learning capabilities. Additionally, the application of RetinaNet significantly reduces labor costs associated with manual grading while maintaining high classification accuracy across different ripeness stages. The implications of this study indicate that leveraging deep learning and automated classification systems can substantially enhance the efficiency of harvesting operations in the palm oil sector. By integrating these techniques, this research contributes to advancing precision agriculture, ultimately leading to improved sustainability in palm oil production. Keywords: RetinaNet, Convolutional Neural Networks, Palm oil, Deep learning, Automated classification.

Item Type:	Thesis (S1)
Call Number CD:	FIK/INFO. 25 167
NIM/NIDN Creators:	41521010092
Uncontrolled Keywords:	RetinaNet, Convolutional Neural Networks, Palm oil, Deep learning, Automated 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 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 > 006 Special Computer Methods/Metode Komputer Tertentu > 006.3 Artificial Intelligence/Kecerdasan Buatan > 006.32 Neural Nets (Neural Network)/Jaringan Saraf Buatan 100 Philosophy and Psychology/Filsafat dan Psikologi > 150 Psychology/Psikologi > 153 Conscious Mental Process and Intelligence/Intelegensia, Kecerdasan Proses Intelektual dan Mental > 153.1 Memory and Learning/Memori dan Pembelajaran > 153.15 Learning/Pembelajaran
Divisions:	Fakultas Ilmu Komputer > Informatika
Depositing User:	khalimah
Date Deposited:	29 Aug 2025 08:28
Last Modified:	29 Aug 2025 08:28
URI:	http://repository.mercubuana.ac.id/id/eprint/97269

Actions (login required)

View Item