Bio degradable and non-bio degradable home Waste segregation and recycling

The project focuses on the segregation and recycling of home waste into biodegradable and non-biodegradable categories using a deep learning approach. The process begins with downloading a dataset of Labeled images representing different types of waste materials. These images are pre-processed by resizing them to a uniform size to ensure consistency for the classification model. A Convolutional Neural Network (CNN) is then utilized to classify the images into biodegradable or non-biodegradable categories based on predefined labels. The CNN is trained using the Labeled dataset, allowing it to effectively recognize and distinguish between the two types of waste. The classification results are then shared as text data to a hardware kit for further processing or automation. To evaluate the performance of the system, a set of 100 test images is used, and the model's accuracy in classifying the waste is measured. This approach not only streamlines the waste segregation process but also contributes to the efficient management and recycling of home waste, promoting a cleaner and more sustainable environment. The use of MATLAB as the primary tool for model development ensures that the implementation is efficient and scalable, while the integration with a hardware kit facilitates real-world applications of the system for waste management in households. The project demonstrates the potential of artificial intelligence in environmental conservation and sustainable waste management practices.

Keywords: Dataset, Pre-Processing, Convolutional Neural Networks, Deep learning, Classification, Accuracy.