Classification of Waste Materials using CNN Based on Transfer Learning

Abstract

Waste Management is important for humans as well as nature for healthy life and a clean environment. The major step for effective waste management is the segregation of waste according to its types. The advancement of technology such as hardware and artificial intelligence is used for the segregation of waste. There are several machine learning and deep learning algorithms available for image classification. Among them, Convolutional Neural Network is the most used one. The main objective of this work is to classify images of waste materials using CNN into seven categories (cardboard, glass, metal, organic, paper, plastic, and trash). Then, cardboard, organic, and paper class images are considered biodegradable waste, and other classes are considered non-biodegradable waste. The pre-trained CNN model such as InceptionV3, InceptionResNetV2, Xception, VGG19, MobileNet, ResNet50 and DenseNet201 have been trained and performed fine-tuning on the waste dataset. Among these models, the VGG19 model performed with less accuracy, whereas the InceptionV3 model performed with high learning accuracy. Overall, the obtained result is promising.