Implementasi Metode Feature Extraction pada Klasifikasi Kualitas Daun Tembakau Madura

Kunto Aji Wibisono, Achmad Fiqhi Ibadillah

Abstract

Madura merupakan salah satu daerah penghasil tembakau di Indonseia. Tembakau Madura  merupakan jenis komoditi perkebunan yang memiliki nilai ekonomi tinggi. Sebagian besar tembakau madura diserap oleh pabrik rokok sebagai bahan baku utama rokok maupun sebagai racikan atau campuran kretek. Secara umum tembakau Madura sendiri dibagi menjadi tiga bagian: tembakau gunung, tembakau tegal, dan tembakau sawah. Jenis tembakau gunung adalah yang paling diburu oleh pabrik rokok, meski produktivitasnya terbilang sangat rendah dibanding tembakau sawah. Terdapat banyak jenis  varietas tembakau gunung yang ditanam petani di Madura, namun  yang memiliki karakteristik khas adalah tembaku Prancak – 95. Hal ini disebabkan  Aroma tembakau Prancak-95 Madura tidak bisa ditiru oleh jenis varietas tembaku lain di Indonesia. Hal lain yang membedakan yaitu terjadi karena kontur atau struktur tanah Madura yang memang khas, yang merupakan kelebihan dari tembakau Madura.Pada penelitian ini didesain sebuah sistem gradding untuk mendeteksi kualitas tembakau Prancak – 95 madura. Deteksi kualitas daun tembakau ini didasarkan pada dua ekstraksi fitur yaitu tekstur dan aromatik. Berdasarkan kedua fitur tersebut nantinya akan diklasifikasikan dengan menggunakan standard kualifikasi SNI. Sehingga  level akurasi deteksi kualitas daun tembakau Madura menjadi lebih optimal

Kata Kunci: Image extraction, Sensor Gas, Tembakau Madura.

Implementation of Feature Image Extraction on Quality Classification of Maduraness Tobacco

ABSTRACT

Madura is one of the tobacco producing areas in Indonesian. Madura tobacco is a type of plantation commodity that has high economic value. Most tobacco Madura is absorbed by cigarette manufacturers as the main raw material of cigarettes as well as as a concoction or clove mixture. In general Madura tobacco itself is divided into three parts: mountain tobacco, tobacco tegal, and tobacco sawah. Types of mountain tobacco are the most hunted by cigarette manufacturers, although the productivity is very low compared to tobacco. There are many types of varieties of mountain tobacco grown by farmers in Madura, but which has a distinctive characteristic is the Prancak-95 tobacco. This is because the Prancak-95 Madura tobacco aroma cannot be imitated by other types of copious varieties in Indonesia. Another thing that distinguishes that occurs due to the contour or structure of Madura land that is typical, which is the advantage of Madura tobacco. In this study designed a grading system to detect the quality of Prancak tobacco - 95 madura. The tobacco leaf quality detection is based on two feature extractions, namely texture and aromatics. Based on these two features will be classified using SNI qualification standards. So that the accuracy level of Madura tobacco leaf quality detection becomes more optimal

Keywords: Image extraction, Gas Sensor, Maduraness Tobacco 

Keywords

Image extraction; Sensor Gas; Tembakau Madura

Full Text:

PDF

References

Badri, M. H. S. A.; Anthana, M.; Hardika, K.. Standart Operasional Kultur Teknis Tembakau. Surakarta; 1994.

Sari, Y., dkk. 2015, Optimasi Conjugate Gradient Pada Backpropagation Neural Network untuk Deteksi Kualitas Daun Tembakau , Konfresnsi Nasional Sistem & Informatika, STIKOM Bali

Zhang, X., dkk. 2008, Images Features Extraction of Tobacco Leaves, Congress on Image and Signal Processing, China

Anilkumar.Muthevi., dkk. 2017, Leaf Classification Using Completed Local Binary Pattern Of Textures, IEEE 7th International Advance Computing Conference, India

Zhang F, Zhang X. Classification and quality evaluation of tobacco leaves based on image processing and fuzzy comprehensive evaluation. Sensors (Basel, Switzerland). 2011; 11: p. 2369–84.

Figaro, (2005), General Information for TGS Sensors.

Dinas Perkebunan Jawa Barat, Buku Pedoman Standarisasi Mutu Tembakau: 2009

Luay Fraiwan, Khaldon Lweesy, Aya Bani-Salma, Nour Mani, “A Wireless Home Safety Gas Leakage Detection System”, IEEE ,pp. 11- 14, 2014.

DOI

https://doi.org/10.21107/rekayasa.v10i2.3607

Metrics

Refbacks

  • There are currently no refbacks.


Copyright (c) 2018 Kunto Aji Wibisono, Achmad Fiqhi Ibadillah

Creative Commons License
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.