| dc.description.abstract | This thesis centers on the imperative task of detecting and segmenting batteries within the
recycling industry, addressing the need for an efficient and accurate solution. The primary
goal is to conduct a comprehensive comparison of state-of-the-art methods to discern the
most suitable approach for the specified task. The comparative analysis extends to both
manual and auto annotations, where manual annotations employ Roboflow, feeding data
into Ultralytics’ YOLOv5 and YOLOv8 models, as well as Azure Custom AI. Auto annotations
leverage Grounded SAM and GroundingDINO, with Grounded SAM data integrated
into YOLOv8 and YOLOv5. Remarkably, YOLOv8, combined with manual annotations
for the custom battery dataset, demonstrates significant success. The thesis concludes by
selecting the most effective method and enhancing a health dashboard based on the chosen
model, providing a comprehensive and practical solution for the recycling industry’s battery
detection and segmentation challenges. | |
