Machine learning based classification of recyclable materials

• Main Author: Huang, Danyi
• Other Authors: Wang Dan Wei
• Format: Final Year Project
• Language: English
• Published: 2017
• Subjects: DRNTU::Engineering::Electrical and electronic engineering
• Online Access: http://hdl.handle.net/10356/72957
• Institution: Nanyang Technological University

With the development of technology, Artificial Intelligence (AI) becomes popular and people make use of it to do jobs. But for recyclable materials selection, most of the classification jobs are still done manually. Therefore, this project is aimed to developed a system for classifying materials by using Machine Learning. This paper introduces TensorFlow which is an open source for Machine Learning. By using it, single object is able to be recognized but not for multiple objects in one image. Because of this limitation on TensorFlow, the idea on the combination of Machine Learning and Open Source Computer Vision Library (OpenCV) image processing is also illustrated in this paper. As a result, most of the materials can be recognized and highlighted in an image.