Classification of drying methods for macadamia nuts based on the glcm texture parameters

  • Simon N. Njuguna Jomo Kenyatta University of Agriculture and Technology
  • Stephen Ondimu Jomo Kenyatta University of Agriculture and Technology
  • Glaston M. Kenji Jomo Kenyatta University of Agriculture and Technology

Abstract

Texture is an important feature in the definition of the external appearance of a material. In this study, Gray-level co-occurrence matrix (GLCM) was used in the analysis of image texture without the destruction of the two varieties of macadamia nuts , KRG-15 and MRG-20, that are dried using seven different methods. Using the Imagej software, four GLCM features were computed in order to define the effect on drying methods on quality of the two varieties. These GLCM features were contrast, Angular Second Moment (ASM), homogeneity and correlation. The aim of this study was to investigate the influence of drying methods on the quality of macadamia nuts for the two varieties: - KRG-15 and MRG-20 with reference to texture quality. The highest classification accuracy of 77% was attained for correctly identifying influence of drying method at direction at angle of 0˚ and displacement, d=1. Solar tent drying method produces superior quality macadamia nuts for both varieties comparable to drying methods; with GLCM feature values (contrast, Homogeneity, Correlation and ASM) of 8,179, 0.048, 0.798 and 0.009 respectively for KRG-15 and 10.238, 0.789, 0.007 and 0.047 respectively for MRG-20. Combination drying of solar tent and Microwave drying produce the least quality,
Keywords—Image texture, GLCM, Angular Second Moment, Homogeneity  
Published
Jul 2, 2018
How to Cite
NJUGUNA, Simon N.; ONDIMU, Stephen; KENJI, Glaston M.. Classification of drying methods for macadamia nuts based on the glcm texture parameters. Proceedings of Sustainable Research and Innovation Conference, [S.l.], p. 19-24, july 2018. ISSN 2079-6226. Available at: <http://sri.jkuat.ac.ke/ojs/index.php/proceedings/article/view/692>. Date accessed: 13 nov. 2019.