An Analysis on Intelligent Garbage Detection using CNN Architecture

An effective image classifier built on a convolutional neural network (CNN) is created, and it can analyse training features to detect and identify any object that is considered to be garbage. Additionally, it has the ability to count the quantity of identified items and give each one a price. Any person who brings rubbish to the ATD will have no trouble having it recognised by the image classifier, and they may take advantage of the recycle value that has been assigned to the object. As a result, it is feasible to exchange waste directly for its comparable price, which will encourage people to use the smart dustbin we suggest. By recycling the trash in it, you can offset the cost of installation and reduce the cost of operation and maintenance. The data set was created using photographs of various types of waste without paper cups in addition to images of garbage with paper cups, which is among the most polluting garbage. Convolutional neural networks that have already been trained on Squenzenet, VGG-19, and GoogLeNetwere used to identify paper cups in these photos. The SquenzeNet, Google Net, and Vgg-19 networks utilised in the study were found to have performance rates of 98.79%, 97.46%, and 95.86%, respectively.