Pengembangan Modul Klasifikasi Apel Envy dan Pasific Rose Menggunakan Jaringan Saraf Tiruan (JST)
Main Article Content
Keywords
Abstract
The variety of apple has similarity with each other, therefore it makes human visual perception difficult to differentiate some of them. The conventional way is often carried out human vision subjective perception, this way makes classification result becomes less. Apple classification using information technology is one step to classify apple more accurate and to help human in apple plantations processing. The goal of this research is developing classification stage on envy and pasific rose apple. The major goal is to classify envy and pacific rose apple using Artificial Neural Network (ANN). Doing feature extraction of training and testing apple images aims that images can be recognized as envy or pasific rose. The way of getting feature extraction is using statistical approach of RGB color histogram from an image. The feature extraction values of RGB color histogram of apple images are intensity, standard deviation, skewness, energy, entropy, and smoothness. Then, those values as an input of classification step using ANN feed-forward backpropagation. The results of this research are consists of three scenario of experiment, first skenario is using 50 training data 10 testing data, it has accuracy value 80%. Second scenario is using 70 training data 10 testing data, it also has accuracy value 80%. Third scenario is using 90 training data 10 testing data, it has accuracy value 90%. This experiment result shows that the highest accuracy value is in third scenario.
Keywords : Apple, Feature extraction, Artificial neural network, Backpropagation
Abstrak — Jenis apel yang beragam dan beberapa diantaranya memiliki kemiripan membuat persepsi pandang manusia sulit membedakan jenis-jenis apel tertentu. Cara konvensional yang sering dilakukan adalah penilaian subyektif dari persepsi pandang manusia. Penilaian subyektif tersebut bisa menghasilkan pengelompokan jenis apel yang kurang tepat. Teknik pengenalan jenis apel berbasis teknologi informasi menjadi salah satu langkah untuk membantu pengelompokan jenis apel supaya lebih tepat dan akurat serta lebih meringankan tugas manusia pada bidang pengolahan hasil perkebunan apel. Penelitian ini bertujuan untuk mengembangkan tahapan klasifikasi apel envy dan pasific rose. Tujuan utamanya adalah mengklasifikasi buah apel jenis envy dan pasific rose dengan menggunakan Jaringan Saraf Tiruan (JST) atau sering disebut dengan Artificial Neural Network (ANN). Citra apel terlebih dahulu dilakukan ekstraksi fitur yang unik dari citra latih dan citra uji, agar citra tersebut dapat dikenali. Salah satu cara ekstraksi fitur adalah dengan menggunakan pendekatan statistik dari histogram warna RGB sebuah citra. Nilai-nilai yang bisa diambil dari histogram warna RGB citra apel adalah nilai rerata intensitas, standar deviasi, skewness, energi, entropi, dan smoothnes. Kemudian nilai-nilai tersebut sebagai nilai masukan untuk tahapan klasifikasi menggunakan JST feed-forward backpropagation. Hasil pengujian dibagi menjadi tiga skenario, pertama dengan data latih 50 data uji 10 memiliki nilai akurasi sebesar 80%. Skenario pengujian kedua dengan data latih 70 data uji 10 memiliki nilai akurasi sebesar 80%. Skenario pengujian ketiga dengan data latih 90 data uji 10 memiliki akurasi sebesar 90%. Hasil pengujian tersebut menunjukan bahwa nilai akurasi tertinggi terletak pada pengujian ketiga.
Kata Kunci : Apel, Ekstraksi fitur, Jaringan saraf tiruan, Backpropagation
References
[2] Austeclino, M.B.J., Angelo, A.D., Manoel, B.N., Artificial Neural Network and Bayesian Network As Supporting Tools for Diagnosis of Asymptomatic Malaria, 978-1-4244-6376-3/10/$26.00, IEEE, 2010.
[3] Calvin Hung, Zhe Xu, Salah Sukkarieh, Feature Learning Based Approach for Weed Calssification Using High Resolution Aerial Images from a Digital Camera Mounted on a UAV, Remote Sensing, 2014.
[4] Fajar Budiman, Muhammad A., Muhammad R., & Suwito, Pengenalan Suara Burung Menggunakan Mel Frequency Cepstrum Coefficient dan Jaringan Syaraf Tiruan pada Sistem Pengusir Hama Burung, Jurnal Nasional Teknik Elektro (JNTE) Vol 5 No 1, (2016).
[5] Farah Z.R., Identifikasi Parasit Malaria pada Think Blood Film Menggunakan Artificial Neural Network (ANN), Jurusan Teknik Elektro, Institut Teknologi Sepuluh Nopember (ITS) Surabaya (2013).
[6] Ion A. H., Identifikasi Jenis Buah Apel Menggunakan Algoritma K-Nearest Neighbor (KNN) dengan Ekstraksi Fitur Histogram, Jurusan Teknik Informatika, Universitas Dian Nuswantoro Semarang (2015).
[7] Mauridhi, H.P., Agus, K., Supervised Neural Network dan Aplikasinya, Graha Ilmu Yogyakarta, (2006).
[8] Robert E. Uhrig, Introduction to Artificial Neural Network, Proceedings of the 1995 IEEE IECON 21st Vol. 1, (1995).
[9] Wei J., Xiangli M., Yun T., Bo X., Dean Z., Fast Segmentation of Colour Apple Image under All-weather Natural Conditions for Vision Recognition of Picking Robots. International Journal of Advanced Robotic Systems, (2016).
Article Sidebar
Article Details
Authors who publish with this journal agree to the following terms:
- Authors retain copyright and grant the journal right of first publication with the work simultaneously licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License that allows others to share the work with an acknowledgement of the work's authorship and initial publication in this journal.
- Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgement of its initial publication in this journal.
- Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work (See The Effect of Open Access).
The authors declare that:
1. This paper has not been published in the same form elsewhere.
2. It will not be submitted anywhere else for publication prior to acceptance/rejection by this Journal.
3. A copyright permission is obtained for materials published elsewhere and which require this permission for reproduction.