Dried Water Lily Stem Validation for Weaving Fiber Production using K-Nearest Neighbors (K-NN) Classification Algorithm

One of the problems associated with the drying process of water lily stems for the weaving industry is the quality of water lily fiber. After the drying process, the dried water lily stems need to be visually inspected, validating their color, to make sure that it is extremely dry. This manual process is normally tedious and time-consuming. To classify the dried water lily stems, an automated validation system was presented. The dried water lily stem was captured by Raspberry Pi Camera. The captured or raw image underwent image processing such as image enhancement, resizing, gray conversion, and color masking. The processed image is fitted to the K-NN Classification Algorithm model to predict the target class. The system yielded 95.45% accuracy for identifying "dry" and "less dry" classifications of dried water lily stems.