This project implements a two-step process to detect and extract odometer numbers from images using YOLOv8 Nano for object detection and EasyOCR for text recognition.
This solution was designed to automate the insurance process. . It utilizes YOLOv8 Nano for its efficiency in environments with limited computational resources, and EasyOCR for its high accuracy in general OCR tasks.
Annotated Odometer Dataset: Car Odometers
- Object Detection: Uses YOLOv8 Nano to locate the odometer in an image.
- Text Recognition: Applies EasyOCR to extract numbers from the detected odometer region.
- Optimized for Limited Resources: Suitable for platforms with restricted computational capabilities like Google Colab.
- Simple Setup: Minimal configuration is needed to get the project up and running.
- Object Detection with YOLOv8 Nano: The model identifies the region containing the odometer based on the learned features from the training dataset.
- Text Recognition with EasyOCR: Numeric information is extracted from the identified ROI using OCR technology.
Contributions to this project are welcome! Please fork the repository and submit a pull request with your features or fixes.
This project is licensed under the MIT License - see the LICENSE file for details.
- Thanks to the MakeSense.ai team for their intuitive labeling tool.
- The YOLO and EasyOCR communities for their outstanding open-source contributions.