Coconuts delivered from farmers generally have different variations in maturity. This study aims to identify the maturity of coconut fruit based on acoustic feature analysis using the Principal Component Analysis (PCA) method. This method combines data acquisition of coconut tapping sounds with multivariate statistical analysis to acoustically recognise the type of coconut maturity. The research experiment involved 40 coconut samples and the recording of coconut knocking sound was done by knocking the coconut three times from each coconut sample using a coconut knocking device in an open space. Time and frequency domain acoustic features were extracted from the resulting audio signals. Subsequently, PCA analysis was used to reduce the dimensionality of the acoustic features and identify patterns that represent the ripeness level of the coconut. The results of PCA visualisation obtained differences in acoustic features from young and old coconut maturity levels can be identified. From the results of analysis using PCA, the first two principal components explained about 40.28% and the second principal component explained about 30.07% of the data variation. Visualisation of the data using a scree plot shows that the young coconut group is clearly separated from the old coconut group.
Copyrights © 2025