@Article{jai.2026.076674,
AUTHOR = {Suman Kunwar},
TITLE = {DWaste: Greener AI for Waste Sorting Using Mobile and Edge Devices},
JOURNAL = {Journal on Artificial Intelligence},
VOLUME = {8},
YEAR = {2026},
NUMBER = {1},
PAGES = {39--49},
URL = {http://www.techscience.com/jai/v8n1/65602},
ISSN = {2579-003X},
ABSTRACT = {The rise in convenience packaging has led to generation of enormous waste, making efficient waste sorting crucial for sustainable waste management. To address this, we developed DWaste, a computer vision-powered platform designed for real-time waste sorting on resource-constrained smartphones and edge devices, including offline functionality. We benchmarked various image classification models (EfficientNetV2S/M, ResNet50/101, MobileNet) and object detection (YOLOv8n, YOLOv11n) including our purposed YOLOv8n-CBAM model using our annotated dataset designed for recycling. We found a clear trade-off between accuracy and resource consumption: the best classifier, EfficientNetV2S, achieved high accuracy (<mml:math id="mml-ieqn-1"><mml:mo>≈</mml:mo><mml:mspace width="negativethinmathspace"/><mml:mn>96</mml:mn><mml:mi mathvariant="normal">%</mml:mi></mml:math>) but suffered from high latency (<mml:math id="mml-ieqn-2"><mml:mo>≈</mml:mo><mml:mspace width="negativethinmathspace"/><mml:mn>0.22</mml:mn></mml:math> s) and elevated carbon emissions. In contrast, lightweight object detection models delivered strong performance (up to <mml:math id="mml-ieqn-3"><mml:mn>80</mml:mn><mml:mi mathvariant="normal">%</mml:mi></mml:math> mAP) with ultra-fast inference (<mml:math id="mml-ieqn-4"><mml:mo>≈</mml:mo><mml:mspace width="negativethinmathspace"/><mml:mn>0.03</mml:mn></mml:math> s) and significantly smaller model sizes (<mml:math id="mml-ieqn-5"><mml:mo>&lt;</mml:mo><mml:mspace width="negativethinmathspace"/><mml:mn>7</mml:mn></mml:math> MB), making them ideal for real-time, low-power use. Model quantization further maximized efficiency, substantially reducing model size and VRAM usage by up to <mml:math id="mml-ieqn-6"><mml:mn>75</mml:mn><mml:mi mathvariant="normal">%</mml:mi></mml:math>. Our work demonstrates the successful implementation of “Greener AI” models to support real-time, sustainable waste sorting on edge devices.},
DOI = {10.32604/jai.2026.076674}
}