APPLICATION OF DIGITAL IMAGE PROCESSING METHOD FOR ROASTED COFFEE BEAN QUALITY IDENTIFICATION: A SYSTEMATIC LITERATURE REVIEW

I Santoso, Ea Yuanita and Rs Karomah

African Journal of Food, Agriculture, Nutrition and Development (AJFAND), 2024, vol. 24, issue 01

Abstract: In coffee processing, there are several important stages, one of which is roasting. The roasting process is an important determinant of coffee quality. Determination of coffee quality can be done using digital image processing methods to produce parameters and quality classifications precisely, make images of better quality so that photos and moving images can be easily understood. This analysis uses a Systematic Literature Review (SLR) for the identification, evaluation, and interpretation of all available research results on the topics discussed. The purpose of this study was to identify and analyze the main quality parameters and the best digital image processing methods used in classifying the quality of roasted coffee beans. From the results of the analysis of 31 journals, it is known that the parameters for evaluating the quality of roasted coffee are color parameters, texture parameters, and shape parameters. The color parameters consist of Red Green Blue (RGB), Grayscale, Hue Saturation Intensity (HSI), and L*a*b* features. The texture parameters consist of energy, entropy, homogeneity, and contrast. As for the feature shape parameters, they are area, circumference, diameter, and percentage of roundness. Results of the analysis show that the main parameter that plays an important role in assessing the quality of roasting coffee is the color parameter. This can be seen from the function of the color parameter in quality identification based on the image of the roasted coffee beans. The quality parameters used are image capture, image resolution, training data, testing data, iterations, and accuracy values. In addition, the resulting image processing methods used for quality classification include Backpropagation (BP), Learning Vector Quantization (LVQ), and K-Nearest Neighbor (KNN). Based on results of the analysis, the best method for classifying the quality of roasting results is Backpropagation, and it is known that the accuracy value of this method has a high range of values.

Keywords: Crop Production/Industries; Research and Development/Tech Change/Emerging Technologies (search for similar items in EconPapers)
Date: 2024
References: View complete reference list from CitEc
Citations:

Downloads: (external link)
https://ageconsearch.umn.edu/record/340616/files/Santoso.pdf (application/pdf)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:ags:ajfand:340616

DOI: 10.22004/ag.econ.340616

Access Statistics for this article

More articles in African Journal of Food, Agriculture, Nutrition and Development (AJFAND) from African Journal of Food, Agriculture, Nutrition and Development (AJFAND)
Bibliographic data for series maintained by AgEcon Search ().