Abstract
Native and Joper chickens are types of chickens whose meat is difficult to distinguish in terms of texture and color. The aim of this study is to develop an information system capable of detecting the type of chicken meat (native or Joper) based on image analysis using the Gray Level Co-Occurrence Matrix (GLCM) method combined with the K-Nearest Neighbour (K-NN) algorithm. In this research, 200 training data samples were used to extract color and texture features and perform calculations using five GLCM parameters (energy, entropy, homogeneity, contrast, and correlation) with four texture distribution directions: 0°, 45°, 90°, and 135°. Classification was then conducted to determine the type of chicken meat using the K-NN algorithm. The results of this study include a system capable of identifying chicken types based on meat, specifically distinguishing between Joper chicken meat and native chicken meat. The system consists of two main processes: calculating gray-level co-occurrence values and determining proximity using the K-Nearest Neighbor algorithm. Based on testing results, the system can perform detection using the GLCM and K-NN methods with an accuracy rate of 80%, as evaluated by 8 out of 10 respondents in this study.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call