Treffer: Java preanger coffee quality control with convolutional neural networks.

The coffee producing industry in Indonesia has the challenge of upholding stringent quality standards, which have conventionally relied on labor-intensive techniques of sorting. Convolutional Neural Networks (CNNs) have made substantial advancements in automating the identification of defects in coffee beans. Nevertheless, these models often encounter constraints in terms of their precision and flexibility when confronted with intricate real-world scenarios. The objective of this study is to address the current deficiency by presenting an enhanced Convolutional Neural Network (CNN) technique that incorporates advanced preprocessing and data augmentation methods to enhance the dependability and accuracy of the model. The suggested model attained an exceptional overall accuracy of 94.5% by using a primary dataset from our Java Preanger coffee beans collection of 2.000 images that were labeled as either 'good' or 'defective'. The results demonstrated outstanding performance metrics, such as a precision of 93% in recognizing damaged beans, a recall rate of 96%, and well-balanced F1 scores of 0.945. The findings illustrate the model's efficacy in precisely identifying defects in coffee beans. [ABSTRACT FROM AUTHOR]

Copyright of AIP Conference Proceedings is the property of American Institute of Physics and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)