Automated Blood Cell Classification Using RoboFlow and YOLOv8
- Tech Stack:Computer Vision, Data Science
- URL: Project Link
Abstract (ABSTRACT):
1.Idea, Feature
Accurately and efficiently identifying blood cells plays a crucial role in diagnosing and monitoring various diseases. In this study, I utilize RoboFlow, an advanced computer vision platform, in conjunction with the state-of-the-art object detection algorithm, YOLOv8, to predict and classify three significant types of blood cells: white blood cells (WBC), red blood cells (RBC), and platelets. The dataset used in this research consists of 1922 labelled images across the three classes.
The methodology involves training the YOLOv8 model on the blood cell dataset, leveraging the robust object detection capabilities of the algorithm to localize and classify blood cells within images accurately. By employing RoboFlow, I streamline the annotation process, enabling efficient labelling of the extensive dataset. Additionally, RoboFlow offers comprehensive data augmentation techniques, enhancing the model’s generalization ability and reducing overfitting risks.
2. Method
Data Collection and Labeling:
- Upload the dataset to RoboFlow, an advanced computer vision platform.
- Utilize the labelling tools provided by RoboFlow to label each blood cell in the images accurately.
3. Result
