Classification of E-Waste Types Using Machine Learning and Digital Image Processing

This paper explores the application of deep learning and computer vision techniques for automated classification and detection of electronic waste (e-waste). A system based on convolutional neural networks (CNN) and faster R-CNN is developed for analyzing e-waste images and extracting information about equipment type and dimensions. The experiment is conducted on a dataset of 500 real-world images of three key e-waste categories - refrigerators, kitchen stoves and TVs. Results demonstrate high classification accuracy of 92% using CNN and 91% detection accuracy with R-CNN. The obtained data enables more precise waste collection planning. The main conclusion is that deep learning holds great potential for improving e-waste management systems.