<![CDATA[Data from the Ministry of Environment and Forestry shows that the amount of organic and non-organic waste in 2023 has started to decline compared to the previous year. However, waste management in the central landfill is still not optimal. This is a problem for the community and the environment because it can cause pollution and disrupt public health around the disposal site. The reason for the difficulty of waste management at the landfill is that people still dispose of waste without separating it first. In addition, it is also due to a lack of public awareness and knowledge. One of the things that can be done to help overcome the problem of waste and its management is to develop an application that can help people understand the importance of waste selection and facilitate socialization in the community. For that, a model is needed that can classify waste based on its type with accurate accuracy. In this study, we propose a deep learning model, CNN with mobilenetV2 architecture, to classify organic and non-organic waste. This model uses a dataset consisting of 4380 images of organic and non-organic waste. Then 3 preprocessing stages were carried out, namely resize, normalization, and augmentation. From this process, data training was carried out and researchers obtained model evaluation results with 98.47% accuracy, 97% precision, 97% recall, and 97% F1 Score evaluation results. These results show that the proposed model is categorized as excellent.]]>
Copyrights © 2024
