Waste Classification and Its Analysis Using RCNN Algorithm

The paper includes a classification model based on RCNN algorithm, using a dataset of waste images with 9 different categories: Trash, plastic, metal objects, cardboard, papers, glasses. The RCNN model accurately detects and classifies waste in images, outperforming other state-of-the-art methods. The system provides insights into waste composition and characteristics, revealing paper and cardboard as the most common types of waste, followed by plastic and trash. Plastic waste has a high variance in shape and colour, making it challenging to classify accurately. Overall, the proposed system is a promising approach to effective waste management and waste analysis the RCNN algorithm is a promising approach to effectively manage waste and provide insights into waste characteristics. The model used here is using a training dataset using the VGG16 pretrained transfer learning model which was trained till 28 epochs.