Immanuela Puspasari Saputro, Junaidy Budi Sanger, Angelia Melani Adrian, Gilberth Mokorisa
Nowadays, plastic waste has become a global problem. Plastic waste can be found in the land, oceans, rivers, and even soil sediments. This problem has motivated various countries to overcome environmental pollution, especially plastic waste primarily through recycling programs. Plastic waste that is still good can be recycled into handicrafts or sold at the waste bank according to the current selling price. However, the waste management officer encountered difficulties classifying plastic waste since there are many forms of waste. Recently, a deep learning-based method, namely The Convolutional Neural Network (CNN), has been widely used for image processing tasks. In this study, we developed a mobile-based application by employing the CNN Algorithm to identify and classify plastic waste. The application will recognize and classify recyclable plastic waste such as plastic bags, plastic bottles, shampoo bottles, etc. There were 156 images collected manually by a smartphone camera and Google Image. All images are converted into jpg format with the size of 300 × 300 pixels. The dataset was divided into two parts, 106 images for data training and 50 images for data testing. From the experimental result, the model obtained an accuracy of 86%. Moreover, there are features to view price lists and tutorials for handicrafts from plastic waste. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023.
Computer Science Department, BINUS Online Learning, Bina Nusantara University, Jakarta, 11480, Indonesia; Informatics Engineering, Faculty of Engineering, Universitas Katolik De La Salle Manado, Manado, 95000, Indonesia
