Classification of Coffee Bean Defects Using Gray-Level Co-Occurrence Matrix and K-Nearest Neighbor

Defects in coffee beans can significantly affect the quality of production so that defects cause a decreasing level production. The purpose this study is to implement GLCM (gray-level co-occurrence matrix) and K-NN (k-nearest neighbor) method on web-based program provided website detect bean defects. This uses algorithm extract features images classify defect beans. system development was built using Unified Modeling Language. utilized programming structure PHP, HTML, CSS, Javascript, Mozilla Firefox as browser for MySql database management systems. results show provide output form classification images. Then, accuracy assessment achieved by 90%. Finally, concluded proposed could help farmers determine input.