Garbage Classification with Deep Learning Techniques
Aakash Sharma, Aarambh Keshri, Aayush Kumar, Rajesh Kumar Yadav · 2023
With growing populations across the world and high consumption of goods and services, many problems arise with respect to waste management and dumps. The waste, if burned emits harmful fumes, and if not sorted properly, can lead to avoidable pollution due to loss of recyclable material. There is no active mechanism geared towards sorting and sifting through the waste generated, proper treatment and recycling of which can be beneficial to both the environment and governments. In order to provide a viable solution to the problem of manual classification and its costs, both human and monetary, we propose an automation system for detecting, sorting and classifying recyclable waste. If scaled for practical use, this can have a long-lasting impact on our relationship with the waste we generate. In this paper, we try to identify waste objects and classify them into categories by making use of Deep Learning techniques, Convolution Neural Networks, ResNet50, MobileNetV2, DenseNet, and other viable techniques and comparing the final results thus achieved.