Image Classification of Ribbed Smoked Sheet using Learning Vector Quantization

International Conference on Computing and Applied Informatics 2016

物理学;计算机科学

Rahmat, R.F.^1 ; Pulungan, A.F.^1 ; Faza, S.^1 ; Budiarto, R.^2

Department of Information Technology, Faculty of Computer Science and Information Technology, University of Sumatera Utara, Medan, Indonesia^1

Department of Information System, College of Computer Science and Information Technology, Albaha University, Saudi Arabia^2

关键词: Economic development;    Image processing technique;    International standards;    Learning Vector Quantization;    Ribbed smoked sheets;    Rubber material;    Sorting process;    Training dataset;   

Others  :  https://iopscience.iop.org/article/10.1088/1742-6596/801/1/012050/pdf
DOI  :  10.1088/1742-6596/801/1/012050

学科分类：计算机科学（综合）

来源: IOP

【 摘 要 】

Natural rubber is an important export commodity in Indonesia, which can be a major contributor to national economic development. One type of rubber used as rubber material exports is Ribbed Smoked Sheet (RSS). The quantity of RSS exports depends on the quality of RSS. RSS rubber quality has been assigned in SNI 06-001-1987 and the International Standards of Quality and Packing for Natural Rubber Grades (The Green Book). The determination of RSS quality is also known as the sorting process. In the rubber factones, the sorting process is still done manually by looking and detecting at the levels of air bubbles on the surface of the rubber sheet by naked eyes so that the result is subjective and not so good. Therefore, a method is required to classify RSS rubber automatically and precisely. We propose some image processing techniques for the pre-processing, zoning method for feature extraction and Learning Vector Quantization (LVQ) method for classifying RSS rubber into two grades, namely RSS1 and RSS3. We used 120 RSS images as training dataset and 60 RSS images as testing dataset. The result shows that our proposed method can give 89% of accuracy and the best perform epoch is in the fifteenth epoch.