Abstract:
Coffee is a beverage obtained from cherry, the fruit of coffee plant. Grading
serves as a process for controlling the quality of an agricultural commodity
so that buyer and seller can do business without personally examining every
lot sold. This study attempts to apply image processing techniques towards
sample coffee raw quality value grading. A total of 145 image datasets and
10,000 coffee beans were used from different grade of coffee plant from
ECX Jimma center. The experimental research design was employed. ImageJ
tool and Matlab programming language were used. For image preprocessing
Gaussian filter to remove noise, contrast enhancement method to enhance
the quality of coffee bean image, normalization and binarization bythresholding 8-bit images algorithm to separate image into region in image
segmentation process were used. Techniques and algorithms such as ANN,
SVM and KNN were used in this study. For the purpose of computing the
grading accuracy of datasets, 80% of total dataset were used for training the
model and the remaining 20% of dataset were used for testing. The major
challenges during conducting this study were keeping the best quality control
environment when acquiring images, extracting best features of HSB color
feature and the homogeneity of coffee plant bean color features. Hence,
appropriate selection of image processing and classification modules paves
the way for higher accuracy in the higher-level process for decision making
