Genetic algorithm optimization for image classification of coconut wood-based on GLCM

• Published in: AIP conference proceedings Vol. 2578; no. 1
• Main Authors: Marleny, Finki Dona, Nurhaeni, Nugraha, Bayu
• Format: Journal Article Conference Proceeding
• Language: English
• Published: Melville American Institute of Physics 03.11.2022
• Subjects: Artificial neural networks; Digital imaging; Feature extraction; Genetic algorithms; Image classification; Image processing; Optimization
• ISSN: 0094-243X; 1551-7616
• DOI: 10.1063/5.0107332

Coconut wood is widely used as a building material in South Kalimantan. Coconut wood is processed into planks and used as walls for some of the Banjar people's houses. In this case, the quality of coconut wood needs to be considered. Image processing with pixel grouping techniques can be used to get an image of an object represented by class or material features such as coconut wood images. The method proposed in this research is Artificial Neural Networks with genetic algorithm optimization that applies feature extraction of Gray Level Co-Occurrence Matrix (GLCM). The first data used is a transverse digital image of coconut wood pieces which are then grouped based on their features using the GLCM method and classified using the ANN method optimized using a Genetic Algorithm. The accuracy value of using the proposed method is 75%.