Naira Currency Detection Dataset Labeling
Worked on a computer vision project focused on detecting and classifying Nigerian naira notes. I was responsible for collecting and labeling a diverse image dataset containing multiple denominations under varying real-world conditions such as different lighting, backgrounds, angles, and partial occlusions. Using bounding box annotation, I accurately labeled each currency note to train an object detection model. The project involved annotating hundreds to thousands of images, ensuring consistency in labeling standards across all classes. I implemented quality control measures such as cross-checking annotations, maintaining clear labeling guidelines, and refining edge cases (e.g., folded or partially visible notes). My work directly contributed to improving the model’s accuracy and robustness in real-world scenarios.