Publication Type : Conference Paper
Publisher : Computer and Communication Technologies (ICECCT)
Source : 2019 IEEE International Conference on Electrical, Computer and Communication Technologies (ICECCT), IEEE, Coimbatore, India, India (2019)
Url : https://ieeexplore.ieee.org/document/8869191
Keywords : bottles, Classification algorithms, classifier, color based segmentation, color based segmentation algorithm, color detection, Decision Tree, environmental science computing, Feature extraction, image classification, Image color analysis, image colour analysis, Image segmentation, image sensors, kNN, KNN classifier, logistic regression, Machine learning algorithms, nearest neighbour methods, Object Detection, physical features, Physical properties, plastic bottle classification, plastic material classification, plastic object, Plastic products, Plastics, recycled plastics, Recycling, Regression analysis, Support vector machines, SVM, tactile sensors, tactile touch sensor, visual features, visual properties, Waste management
Campus : Coimbatore
School : School of Engineering
Department : Computer Science
Year : 2019
Abstract : This paper presents an approach to identify the type of plastics depending upon it's material, it can be concluded whether the plastic can be recycled or not. In this approach, visual and physical properties were used to classify the plastic materials. It makes use of the fact that recycled plastics are having some similar features like weight, pressure and color. Given an image of a plastic object, these features will form a dataset to train different classifiers which will classify the given plastic into recycled or non-recycled. A color based segmentation algorithm is used to detect color and KNN classifier is used to predict the color of plastic. A tactile touch sensor is used to calculate the pressure that can be applied on the plastic object. Different types of plastics are having inconsistent set of features. Therefore, perceptively we are using four different classifiers for the classification namely SVM, KNN, Decision tree and Logistic Regression.
Cite this Research Publication : L. R. Kambam and Aarthi, R., “Classification of plastic bottles based on visual and physical features for waste management”, in 2019 IEEE International Conference on Electrical, Computer and Communication Technologies (ICECCT), Coimbatore, India, India, 2019.