Desain Sistem Pengenalan Varietas Bibit Tanaman Kelapa Sawit dengan Pendekatan Design Science Research Methodology (DSRM)

Penulis

Erick Fernando, Surjandy Surjandy, Meyliana Meyliana, Padapotan Siagian

Abstrak

Penelitian ini menjelaskan pengembangan sistem proses klasisfikasi varietas bibit tanaman kelapa sawit berdasarkan metodologi perkembangan baru, yang disebut design science research methodology (DSRM). Metodologi penelitian ini diadopsi untuk mencakup enam kegiatan: identifikasi masalah dan motivasi, definisi tujuan solusi, desain dan pengembangan, demonstrasi, evaluasi, dan komunikasi. Berdasarkan metode pengembangan DSRM ini, sistem ini berhasil dikembangkan dan dapat digunakan dengan baik untuk dapat mengklasifikasi varietas kelapa sawit. Dimana permasalah yang terjadi adalah sulitnya menentukan varietas tersebut sehingga dapat menimbulkan kesalahan dalam membeli untuk ditanam oleh petani. Penelitian ini menggunakan Metode PNN (probabilistic neural network) yang diterapkan didalam sebuah aplikasi digunakan untuk mendeteksi daun dari bibit kelapa sawit. Dimana menggunakan proses pelatihan (training) supervised terlebih dahulu untuk proses pembelajaran data. Penelitian ini berkontribusi metodologi pengembangan baru dari bidang Sistem Informasi (IS) sebagai model referensi untuk pengembangan aplikasi dimasa depan, bersama dengan integrasi metode PNN sebagai klasifikasi varietas bibit kelapa sawit. Penelitian ini juga menghasilkan sebuah aplikasi yang dapat dengan baik dalam mendeteksi bibit kelapa sawit dengan deskripsi daun sebagai objek dengan menghasilkan tingkat akurasi bisa mencapai 78.1% dan presisi bisa mencapai 78.8% apabila nilai Laplacian/spread mendekati 1. Jumlah data akan mempengaruhi dari nilai akurasi dan presisi untuk proses klasifikasi varietasnya.

 

Abstract

This study explains the development of a system of classification processes for oil palm seedling varieties based on a new development methodology, called the design of science research methodology (DSRM). The methodology of this research was adopted to cover six activities: identification of problems and motivations, definition of objective solutions, design and development, demonstration, evaluation, and communication. Based on the method of developing this DSRM, this system was successfully developed and can be used well to be able to classify oil palm varieties. Where the problem that occurs is the difficulty of determining the variety so that it can cause errors in buying to be planted by farmers. This study uses the PNN Method (probabilistic neural network) that is applied in an application used to detect leaves from oil palm seedlings. Where to use the training process (training) supervised first for the data learning process. This research contributes to the new development methodology of the Information Systems (IS) field as a reference model for future application development, along with the integration of the PNN method as a classification of oil palm seed varieties. This research also produces an application that can be good at detecting oil palm seeds with leaf descriptions as objects by producing an accuracy rate of 78.1% and precision can reach 78.8% if the Laplacian / spread value approaches 1. The amount of data will affect the accuracy value and precision for the classification process of the varieties.

Teks Lengkap:

PDF

Referensi


ANSHORI MOCH, D. M. (2009) Biologi untuk SMA Kelas X. Jakarta Pusat: Departemen Pendididkan Nasional.

Ekspor Minyak Sawit Indonesia Turun 6 persen di Semester Pertama 2018 - The Palm Scribe (no date). Available at: https://thepalmscribe.id/id/ekspor-minyak-sawit-indonesia-turun-6-persen-di-semester-pertama-2018/ (Accessed: 14 December 2018).

FERNANDO, E. et al. (2014) ‘Analysis of security and performance service in service oriented architecture (SOA) and data integration’, International Conference on Electrical Engineering, Computer Science and Informatics (EECSI), 1(August), pp. 270–274. doi: 10.11591/eecsi.1.339.

FERNANDO, E., ASSEGAFF, S. AND ROHAYANI, A. H. H. (2016) ‘Trends information technology in E-agriculture: A systematic literature review’, in 2016 3rd International Conference on Information Technology, Computer, and Electrical Engineering (ICITACEE). IEEE, pp. 351–355. doi: 10.1109/ICITACEE.2016.7892470.

GRIMBLE, M. J. (1980) ‘A new finite-time linear smoothing filter’, International Journal of Systems Science, 11(10), pp. 1189–1212. doi: 10.1080/00207728008967083.

GÜRELLI, M. I. AND ONURAL, L. (1996) ‘A class of adaptive directional image smoothing filters’, Pattern Recognition, 29(12), pp. 1995–2004. doi: 10.1016/S0031-3203(96)00046-5.

HAYKIN, S. (2008) Neural Networks and Learning Machines: A Comprehensive Foundation, arXiv preprint. doi: 978-0131471399.

KARTIKA FIRDAUSY, A. B. (2005) Teknik Pengolahan Citra Digital menggunakan Delphi. Andi Yogyakarta.

KOWALSKI, P. A. AND KUSY, M. (2018) ‘Determining significance of input neurons for probabilistic neural network by sensitivity analysis procedure’, Computational Intelligence, 34(3), pp. 895–916. doi: 10.1111/coin.12149.

LIGA, W. AND FERNANDO, E. (2017) ‘Penyandang Tunarungu Berbasis Android’, 12(1), pp. 926–937.

MARYANDI, FERNANDO, E. AND B, M. R. P. (2014) ‘Ilmu Pengetahuan Sosial Berbasis Android ( Studi Kasus Sman 4 Jambi )’, Jurnal Ilmiah Media SISFO, 8(2), pp. 114–127.

NASAMSIR AND INDRAYADI, M. (2016) ‘Karakteristik Fisik dan Produksi Kelapa Sawit ( Elaeis guineensis Jacq .) pada Tiga Agroekologi Lahan Program Studi Agroteknologi’, 1(2), pp. 55–61.

PARANDEKAR, A. B., DHANDE, S. S. AND VHYAWHARE, H. R. (2014) ‘A Review on Changing Image from Grayscale to Color’, International Journal of Advanced Research in Computer Engineering & Technology, 3(1), pp. 143–146.

PARHUSIP, J. et al. (no date) ‘DESIGN OF THE MAPPING OF ORANG UTAN WITH SPATIAL DATA’, 12(2), pp. 38–46.

PEFFERS, K. et al. (2007) ‘A Design Science Research Methodology for Information Systems Research’, Journal of Management Information Systems, 24(3), pp. 45–77. doi: 10.2753/MIS0742-1222240302.

PRASETYO, E. (2011) Pengolahan Citra Digital dan Aplikasinya menggunakan Matlab. ANDI Yogyakarta.

PUTRA, D. (2010) Pengolahan Citra Digital. Yogyakarta: ANDI Yogyakarta.

RACHMAN ANDI, R., BENY AND FERNANDO, E. (2017) ‘Perancangan E-Commerce Berbasis Website Pada Toko Dunia Palembang’, Jurnal Ilmiah Processor, 12(2), pp. 1102–1117. Available at: https://scholar.google.com/scholar?hl=id&as_sdt=0%2C5&q=Perancangan+E-Commerce+Berbasis+Website+Pada+Toko+Mirabella+Batik+Jambi+Andi&btnG=.

SARAVANAN, C. (2010) ‘Color image to grayscale image conversion’, 2010 2nd International Conference on Computer Engineering and Applications, ICCEA 2010, 2, pp. 196–199. doi: 10.1109/ICCEA.2010.192.

Sekarang, Indonesia Punya Harga Acuan CPO Sendiri - Market Bisnis.com (no date). Available at: https://market.bisnis.com/read/20180412/94/783841/sekarang-indonesia-punya-harga-acuan-cpo-sendiri (Accessed: 14 December 2018).

SIMON HAYKIN (McMaster University, Hamilton, Ontario, C. (2005) ‘Neural Networks - A Comprehensive Foundation - Simon Haykin.pdf’, p. 823.

SUN, Y. et al. (2018) ‘Indoor Sound Source Localization with Probabilistic Neural Network’, IEEE Transactions on Industrial Electronics, 65(8), pp. 6403–6413. doi: 10.1109/TIE.2017.2786219.

WU, S. G. et al. (2007) ‘2007 IEEE International Symposium on Signal Processing and Information Technology A Leaf Recognition Algorithm for Plant Classification Using Probabilistic Neural Network’. doi: 10.1109/ISSPIT.2007.4458016.

YAN FAUZI, YUSTINA E WIDYASTUTI, IMAN SATYAWIBAWA, R. H. P. (2002) Seri Agribisnis Kelapa Sawit. Depok: Penebar Swadaya. Penebar Swadaya Grup.




DOI: http://dx.doi.org/10.25126/jtiik.2020721456