Waste Collection & Segregation using Computer Vision and Convolutional Neural Network for Vessels

The global crisis of pollution has influenced our lives adversely. Preparing the most salutary in reducing the catastrophe has its dominant necessity. Due to the expanding population, plastics in our neighborhood water bodies have increased for the past ten years, progressing the obligation to clean it up. This work intends towards the analysis of an automatic garbage collection system for a vessel/boat with improved performance. The focus of this paper is the development of a sorting and classification mechanism/model for the collected garbage by the vessel/boat incorporating the application of convolutional neural network (CNN) and computer vision. Using CNN & computer vision, the garbage features can be extracted and can be classified further into biodegradable and non-biodegradable with a prediction accuracy of more than 90%, which can be further increased by increasing the data set quantity for the model.