Identifying Arabica Raw Coffee Bean Varieties through Feature Extraction GLCM and Circularity

  • Abstract
  • Keywords
  • References
  • PDF
  • Abstract

    The morphology of raw coffee bean which has colour, texture, size and circularity features are used as standardization to calculate the price and the quality of the raw coffee bean. Meanwhile, coffee farmers have difficulties to distinguish the coffee varieties based on the features of raw coffee bean. Generally, the way of the distinguish the varieties of the coffee is through their own visual perception in form of the tree, leaves, and raw coffee bean. They find it difficult to distinguish the coffee varieties due to the similarity of the varieties forms and colours. This research proposes to solve the problem through the image processing method as the second opinion to help the coffee farmers in identifying the coffee varieties. The research is conducted in three steps processes: The first is pre-processing by cropping the image of raw coffee bean. The second is extracting the image feature of raw coffee bean with Gray Level Co-occurrence Matrices (GLCM) and circularity feature. The last is classifying the feature with multilayer perceptron. The results of the image processing method indicate that the accuracy is 90% with sensitivity is 90%, and 90% specific.


  • Keywords

    raw coffee bean, image processing, classification, GLCM

  • References

      [1] Radi, Muhammad Rivai, and Mauridhi Hery Purnomo, “Combination of first and second order statistical features of bulk grain image for quality grade estimation of green coffee bean,” ARPN J. Eng. Appl. Sci., vol. 10, no. 18, Oct. 2015.

      [2] R. G. Apaza, C. E. Portugal-Zambrano, J. C. Gutiérrez-Cáceres, and C. A. Beltrán-Castañón, “An approach for improve the recognition of defects in coffee beans using retinex algorithms,” in Computing Conference (CLEI), 2014 XL Latin American, 2014, pp. 1–9.

      [3] R. H. M. Condori, J. H. C. Humari, C. E. Portugal-Zambrano, J. C. Gutiérrez-Cáceres, and C. A. Beltrán-Castañón, “Automatic classification of physical defects in green coffee beans using CGLCM and SVM,” in Computing Conference (CLEI), 2014 XL Latin American, 2014, pp. 1–9.

      [4] E. M. de Oliveira, D. S. Leme, B. H. G. Barbosa, M. P. Rodarte, and R. G. F. A. Pereira, “A computer vision system for coffee beans classification based on computational intelligence techniques,” J. Food Eng., vol. 171, pp. 22–27, 2016.

      [5] G. Liu, R. Wang, Y. Deng, R. Chen, Y. Shao, and Z. Yuan, “A new quality map for 2-D phase unwrapping based on gray level co-occurrence matrix,” IEEE Geosci. Remote Sens. Lett., vol. 11, no. 2, pp. 444–448, 2014.

      [6] M. M. Sebatubun, C. Haryawan, and B. Windarta, “Classification of ground glass opacity lesion characteristic based on texture feature using lung CT image,” J. Exp. Theor. Artif. Intell., vol. 30, no. 2, pp. 203–215, 2018.

      [7] Z. Fu and Y. Han, “A circle detection algorithm based on mathematical morphology and chain code,” in Computing, Measurement, Control and Sensor Network (CMCSN), 2012 International Conference on, 2012, pp. 253–256.

      [8] Z. Fu and Y. Han, “A Circle Detection Algorithm Based on Mathematical Morphology and Chain Code,” in 2012 International Conference on Computing, Measurement, Control and Sensor Network, 2012, pp. 253–256.

      [9] H. Ghaderi and P. Kabiri, “Fourier transform and correlation-based feature selection for fault detection of automobile engines,” in Artificial Intelligence and Signal Processing (AISP), 2012 16th CSI International Symposium on, 2012, pp. 514–519.

      [10] R. Wald, T. M. Khoshgoftaar, and A. Napolitano, “Using correlation-based feature selection for a diverse collection of bioinformatics datasets,” in Bioinformatics and Bioengineering (BIBE), 2014 IEEE International Conference on, 2014, pp. 156–162.

      [11] L. Noriega, “Multilayer perceptron tutorial,” Sch. Comput. Staffs. Univ., 2005.




Article ID: 26964
DOI: 10.14419/ijet.v7i4.44.26964

Copyright © 2012-2015 Science Publishing Corporation Inc. All rights reserved.