Accurate electric meter reading is crucial for efficient energy management and billing processes. This paper introduces a comprehensive approach to automatic meter reading, leveraging the YOLOv8 model for meter display extraction and evaluating various YOLO architectures for digit recognition. Through extensive experimentation, we demonstrate that YOLOv9 achieves the highest accuracy and efficiency, with a precision of 0.917, recall of 0.899, and mean Average Precision (mAP) of 0.919. Our comparative analysis explores the strengths and weaknesses of each YOLO variant, considering factors such as detection speed, accuracy, and robustness. The results indicate that YOLOv9 outperforms other variants, offering promising potential for applications in electric meter reading systems. These results contribute significantly to the advancement of automatic meter reading technologies, providing valuable insights for researchers in computer vision and energy management fields. The research not only improves the accuracy and efficiency of electric meter reading processes but also lays the groundwork for the development of more effective systems. With implications for smart city infrastructure and energy conservation initiatives, this study marks a significant step toward a more sustainable and technologically advanced energy ecosystem
Dataset Description:
Dataset sample images:
Class distribution :
image:
images after detection:
final output in form of text:
validation set:
