期刊名称：International Journal of Advances in Engineering and Management
电子版ISSN：2395-5252
出版年度：2020
卷号：2
期号：11
页码：185-188
DOI：10.35629/5252-02116067
语种：English
出版社：IJAEM JOURNAL
摘要：The generation of waste India is becoming a great concern, and it has affected our environment and may even affect the life of people living near these dump sites. The recent study figures show that India generates nearly 26,000 MT of plastic waste on a daily basis and 94 lakh tonnes trash every year. In 2017-18, the Capital generated 6,800 tonnes of municipal waste daily, out of which 690 tonnes of trash was plastic - the highest in the country. This project focuses on building a Deep learning classifier to classify a given waste product into one of the recycling categories such as plastic, glass so that it is easier for people to recycle waste and less waste reaches the wrong place like illegal dumpsites which can lead to not only health issues but also cause a big damage to our surrounding environment. The model is based on the InceptionV3 architecture and uses the concept of transfer learning for training. The system has been tested on the dataset collected from various resources ranging from the TACO dataset to the TrashNet dataset the overall accuracy of the model was around 94.6%.
