<![CDATA[The implementation of post-harvest treatments and good horticultural shed management enhances the yield and quality of products, supporting horticultural competitiveness. However, determining the maturity of pineapple fruits still relies on traditional methods, lacks consistency, and has the potential to cause financial losses. Selecting and classifying pineapples according to standard maturity indices can reduce yield losses and maintain quality, meeting the needs of local and international markets. Research on pineapple maturity classification focuses on deep learning, utilizing data from smallholder farms, and specifically targeting the pineapple body without the crown in 4 maturity classes. The MobileNetV2 classification model with geometric and photometric augmentation achieves 93% accuracy, compared to AlexNet and VGG16. The research aims to improve agricultural efficiency, assist farmers, and provide insights into pineapple maturity detection and classification for local and international markets. Evaluation experiments provide an understanding of the model's performance in detecting and classifying pineapple fruit maturity.]]>
Copyrights © 2024
