Tools
Tools
Satvilkar, Mandar (2018) Image Based Trash Classification using Machine Learning Algorithms for Recyclability Status. Masters thesis, Dublin, National College of Ireland.
Preview |
PDF (Master of Science)
Download (1MB) | Preview |
Abstract
This research paper talks about Trash classification which comes across as a social and unique cause where Image processing through Analytics can help with better management of garbage and waste materials. A standout amongst the most proficient approach to process trash as indicated by its recyclability is to comprehend what class that waste has a place. Analytics can make this herculean task achievable through image classification into its proper categorizations. The goal of this task is to take pictures with single items, segregate them and group those pictures into five particular classifications of cardboard, glass, metal, paper and plastic. We will utilize a dataset of around 400-500 pictures from every class indicated previously. This dataset was made accessible freely and the creators of the dataset give full access to it through their GitHub store. The model utilized will be a Convolutional Neural Network (CNN) because of its impressive and efficient capabilities in the world of machine learning, particularly in image classification. Further, an Extreme Gradient Boosting will be implemented and checked if it can perform better than a CNN. Multiple methods were juxtaposed in order to zero in on the best method or combination of methods which can prove beneficial and efficient for image classification.
| Item Type: | Thesis (Masters) |
|---|---|
| Subjects: | Q Science > QA Mathematics > Electronic computers. Computer science T Technology > T Technology (General) > Information Technology > Electronic computers. Computer science G Geography. Anthropology. Recreation > GE Environmental Sciences > Environment |
| Divisions: | School of Computing > Master of Science in Data Analytics |
| Depositing User: | Caoimhe Ní Mhaicín |
| Date Deposited: | 05 Nov 2018 10:12 |
| Last Modified: | 05 Nov 2018 10:12 |
| URI: | https://norma.ncirl.ie/id/eprint/3422 |
Actions (login required)
 |
View Item |