A cloud based smart recycling bin for waste classification Nikolaos Baras, Dimitris Ziouzios, Minas Dasygenis, Constantinos Tsanaktsidis University of Western Macedonia Kozani 50131, Greece nbaras@uowm.gr, dziouzios@uowm.gr, mdasyg@ieee.org, ktsanaktsidis@uowm.gr Abstract—Due to the Earth’s population rapid growth along with the modern lifestyle the urban waste constantly increases. People consume more and the products are designed to have shorter lifespans. Recycling is the only way to make a sustainable environment. The process of recycling requires the separation of waste materials, which is a time consuming procedure. However, most of the proposed research works found in literature are neither budget-friendly nor effective to be practical in real world applications. In this paper, we propose a solution: a low-cost and effective Smart Recycling Bin that utilizes the power of cloud to assist with waste classification. A centralized Information System (IS) collects measurements from smart bins that are deployed all around the city and classifies the waste of each bin using Artificial Intelligence and neural networks. Our implementation is capable of classifying different types of waste with an accuracy of 93.4% while keeping deployment cost and power consumption very low. Index Terms—cloud, smart bin, classification, computation offloading, recycle, recycling bin, IoT I. I NTRODUCTION Due to the Earth’s population rapid growth, billions of tons of garbage are generated annually across the globe. In European countries, 6 metric tons of garbage being generated by each person per year [1]. The growth of the population, however, is not the only reason for the increased waste gener- ated. Each person individually consumes an increased amount of products and therefore produces more waste. Disposable products have also been very popular lately due to their ease of use and low cost. The majority of these waste is dumped in landfills and water bodies. If this trend continues to grown, the Globe will soon be a giant garbage bin. This can cause a major impact on the environment. The sorting of waste should be performed at the earliest stage possible, in order to maximize the amount of recycled materials and reduce the possibility of being contaminated by other waste materials. This will help to minimize health and environmental problems, such as greenhouse gas emissions, water and air pollution. This is why recycling plays a very important role in preserving the environment. It is extremely important to preserve valuable resources, reduce non-reusable waste and minimize pollution. Reusing materials and prolonging their lifetime is the only way to make a sustainable environment [2]. This is why that over the last few years, the EU has established many laws and policies regarding recycling [3] (Fig. 1 [4]). Fig. 1. Average recycling rate of municipal waste in EU of total waste generated Because of the nature of the recycling process, the sep- aration of the waste materials is necessary. Many organiza- tions believe that by learning people’s behavior they can im- prove the recycling process. Some researchers investigate pro- environmental behaviors such as Theory of Planned Behavior [5], Norm-Activation-Theory [6], and Value Belief Theory [7] to predict how people behave independently. Gathering suffi- cient information on recycled materials can prompt improved recycling conduct [8]. The optimal way to learn people’s behavior, however, is not by questionnaires and studies but by analyzing the disposed materials in real time and predicting recycle patterns. As the problem of recycling has worsened over the last years, it has also become a political issue. In a typical modern city, multiple waste bins are being used for waste disposal; usually glass, plastic, aluminium and paper. Because of this, citizens play a very important role as they are the ones who collect and assort waste by material. The assortment process is one of the reasons why citizens don’t get involved to recycle. We are confident that we can offer a solution to this issue by designing a system to automate the waste classification process and lift some weight off of citizens. In this paper, we propose an extension of our previous research work [9]. A modern smart-bin; an IoT device that uses Computer Vision (CV), Artificial Intelligence (AI) techniques and computational offloading to the cloud to detect and assort waste that is being disposed into by material (such as paper, plastic etc) and type (such as bottles). The structure of the paper is as follows. In Section II, the background of this paper is presented along with related