<![CDATA[The quality of edible bird’s nests is a crucial factor in determining their market value, thus requiring an accurate and automated classification system. This study aims to develop a quality classification model for edible bird’s nests using a Convolutional Neural Network (CNN) algorithm with a transfer learning approach based on the MobileNetV2 architecture. The dataset consists of 3,406 bird’s nest images collected directly from farmers, which were processed through background removal and aggressive augmentation to highlight the main object. The data were evenly split into 2,723 training images and 683 validation images, covering three quality classes: high, medium, and low. The model was trained in two stages: initial training with a frozen base and subsequent fine-tuning. Evaluation results showed an improvement in accuracy from 93% to 97% after fine-tuning, with average precision, recall, and F1-score values of 0.97. The confusion matrix indicated high classification accuracy. This study contributes to the development of an image-based classification model with high accuracy, offering potential for efficient and objective industrial sorting.]]>
Copyrights © 2026
