Beef Quality Classification based on Texture and Color Features using SVM Classifier

Rani Farinda, Zul Rijan Firmansyah, Chaerus Sulton, I Gede Pasek Suta WIJAYA, Fitri Bimantoro

Abstract


Beef quality can be examined visually by observing the beef color or texture using human eyes.  This manual method is very simple yet very subjective because of differences in knowledge about fresh or defective beef characteristics and differences in accuracy. Therefore, a system that can automatically classify beef quality whether it is still fresh or already defective is needed. In this research, we developed a system that can classify beef quality based on its color and texture features using Support Vector Machines classifier. Statistical approach and Gray Level Co-Occurrence Matrix (GLCM) methods were used for the feature extraction process. The total of data used in this research was 480 images, divided into training and testing datasets. The highest accuracy was 97% for cold beef when the system was tested using color features of HSI color space.


Full Text: PDF

Refbacks

  • There are currently no refbacks.


View JTI Stats