Ripeness Classification of Cacao Using Cepstral-Based Statistical Features and Support Vector Machine

JAM Galindo, JEC Rosal… - 2022 IEEE International …, 2022 - ieeexplore.ieee.org
JAM Galindo, JEC Rosal, JF Villaverde
2022 IEEE International Conference on Artificial Intelligence in …, 2022ieeexplore.ieee.org
High quality fine cocoa beans come from farms on which the cacao is grown, harvested and
processed properly. Harvesting under-ripe cacao pods would result in poor-quality yield
because it won't have developed all its flavors and aromas and it will not continue to ripen
after harvest and not suitable for fermentation. On the other hand, over-ripe cacao pods may
germinate and possibly will lead to fungal-infective, defective and generate poor quality
yields. The best way of harvesting cacao pods is when it is about to ripe. The conventional …
High quality fine cocoa beans come from farms on which the cacao is grown, harvested and processed properly. Harvesting under-ripe cacao pods would result in poor-quality yield because it won't have developed all its flavors and aromas and it will not continue to ripen after harvest and not suitable for fermentation. On the other hand, over-ripe cacao pods may germinate and possibly will lead to fungal-infective, defective and generate poor quality yields. The best way of harvesting cacao pods is when it is about to ripe. The conventional way of determining the ripeness of a cacao pod is by manual tapping to get a hollow sound. Hollow sound will indicate ripeness because the cacao beans are no longer tightly packed inside the pod. This paper recommends an approach to classify the ripeness of cacao using the acoustic sensing technology. The device will generate an acoustic signal from cacao pods which was recorded for acoustic signal analysis. Cepstral-based technique was used for feature extraction and Support Vector Machine (SVM) algorithm model and tested using nested cross validation resulted in 94% mean test score.
ieeexplore.ieee.org
以上显示的是最相近的搜索结果。 查看全部搜索结果