This computer vision project utilizes the YOLOv8 model to detect and analyze grape bunches, bruises, stems, and leaves. The model was trained on a custom dataset I curated, which included annotated images of grape-related objects. Through iterative training, the model's weights were fine-tuned to improve its accuracy. Future work may involve expanding the dataset and fine-tuning the model architecture for enhanced performance.
- testCanli.py - This script uses a webcam to perform real-time detection of objects. It visualizes the detection results on each frame captured from the webcam.
- testKoordinat.py - Focuses on detecting specific classes (in this case, grape bunches) and prints their coordinates on the screen.
- testMerkezNokta.py - Draws a rectangle and a center point on a static image to illustrate object detection results.
- Python 3.x
- OpenCV library
- Ultralytics YOLO library
Install the required Python libraries using pip:
bash pip install opencv-python-headless ultralytics
Example
![Örnek](https://private-user-images.githubusercontent.com/92215497/327262388-3df85b15-08d0-4832-a7a4-d246d709a2c7.jpg?jwt=eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9.eyJpc3MiOiJnaXRodWIuY29tIiwiYXVkIjoicmF3LmdpdGh1YnVzZXJjb250ZW50LmNvbSIsImtleSI6ImtleTUiLCJleHAiOjE3MjAzOTMwMDMsIm5iZiI6MTcyMDM5MjcwMywicGF0aCI6Ii85MjIxNTQ5Ny8zMjcyNjIzODgtM2RmODViMTUtMDhkMC00ODMyLWE3YTQtZDI0NmQ3MDlhMmM3LmpwZz9YLUFtei1BbGdvcml0aG09QVdTNC1ITUFDLVNIQTI1NiZYLUFtei1DcmVkZW50aWFsPUFLSUFWQ09EWUxTQTUzUFFLNFpBJTJGMjAyNDA3MDclMkZ1cy1lYXN0LTElMkZzMyUyRmF3czRfcmVxdWVzdCZYLUFtei1EYXRlPTIwMjQwNzA3VDIyNTE0M1omWC1BbXotRXhwaXJlcz0zMDAmWC1BbXotU2lnbmF0dXJlPTgwYjAxODEwNTc2ZjVlZWVjYjRmZWJhOGRkMjlkMzY4NDQ0OTA4Zjg2Yzg4NTk0Mzk2MTlmNWYwYWZmZTI2YzYmWC1BbXotU2lnbmVkSGVhZGVycz1ob3N0JmFjdG9yX2lkPTAma2V5X2lkPTAmcmVwb19pZD0wIn0.fZ9tNsj9QY0J4b8084UlJr73ss9dz7DtW2VSJxX0UB4)