Effective management of construction and demolition waste (CDW) remains a critical concern for environmental sustainability and economic efficiency. Inadequate sorting mechanisms result in suboptimal resource use and diminished recycling opportunities. This paper presents a sophisticated machine-learning approach that utilizes multichannel convolutional neural networks (CNNs) to improve CDW sorting and recycling. Our methodology incorporates a two-phase deep learning model, beginning with a U-Net architecture for detailed, pixel-level segmentation of waste materials. This phase achieves an Intersection over Union (IoU) of 0.902, ensuring precise material localization. Following this, a CNN based on EfficientNet is applied, achieving a classification accuracy of 99%, demonstrating significant improvement over traditional methods. Our comparative analysis highlights the practical advantages of this approach, including labor savings and superior classification accuracy compared to manual sorting methods. By focusing on integrating real-time processing capabilities, this research provides actionable insights for advancing automated CDW management systems. Additionally, providing codes, training and testing datasets, and supplementary resources facilitates further research and implementation. These findings underline the potential of machine learning to revolutionize CDW sorting, aligning with the principles of environmental sustainability and economic efficiency.