<![CDATA[Waste is a persistent issue in Indonesia and worldwide, with increasing waste volumes driven by population growth and consumption. One effective solution is waste sorting, but public awareness and limited infrastructure remain significant challenges. This study explores the use of image processing technology based on Deep Learning, specifically Convolutional Neural Networks (CNN), for automated waste sorting. The data consists of 3300 waste images categorized into six classes: cardboard, glass, metal, paper, plastic, and organic. The study focuses on the use of the DenseNet169 architecture to determine the model's effectiveness in classifying types of waste. Based on previous research, the DenseNet169 architecture has shown advantages in terms of computation and accuracy. The results of this study indicate that DenseNet169 achieves a classification accuracy of 92% for waste, with an F1 Score of 0.92 or 92% in both macro and weighted average metrics.]]>
Copyrights © 2024
