A Novel Method for Waste Classification in Organic and Recyclable images using Deep Convolutional Neural Network

Ali Noshad¹ Reza Karami² Ahmadreza Khonaksar³

1) Student of Computer Engineering,Iran University of Salman Farsi Kazerun Email:
2) Student of Electrical Engineering,Iran University of Salman Farsi Kazerun Email:
3) Student of Computer Engineering,Iran Zand Institute of Higher Education Email:

Publication : 2nd. International Congress on science & Engineering - paris(parisconf.com)
Abstract :
Waste disposal has a direct or indirect impact on human lives and the environment, waste can be produced during the extraction or production process of raw materials or as a result of the use of chemical products and human activities. Waste can be categorized as industrial waste, medical waste (hospital), and household waste. Improper waste disposal can pose risks to all living species by polluting the air, water, and soil. The main method of waste disposal is to bury them in the ground, which is inefficient and expensive and pollutes the environment. To prevent the spread of disease, the first attempt was to recycle sensible materials such as scrap iron, paper, and wood. recycling is very important from an economic and environmental point of view, so efficient recycling is very important for a bright future. For more efficient and safe recycling, it is necessary to use intelligent systems. In order to provide the most efficient approach, we proposed a new deep convolutional neural network architecture as a deep learning approach to detect organic and recyclable objects. RGB organic and recyclable images were used to train and test the model. The proposed model presented in this study, in the classification of organic and recyclable object, achieved an accuracy of 92.9% and according to the evaluation results, the proposed model, based on accuracy, precision, sensitivity and F1-Score, compared to existing architectures in deep learning has better performance in classifying organic and recyclable objects. The results obtained from this study indicates that the proposed CNN architecture is promising example of usage of artificial intelligence more specifically deep learning on behalf of ecological awareness.
Keywords : Image Processing Deep Learning Convolutional Neural Network waste classification