This code implements object detection and tracking using the YOLO (You Only Look Once) algorithm. It utilizes the ultralytics library for YOLO object detection, cv2 for image processing, cvzone for graphical overlays, and the sort library for object tracking.

1. Clone the repository:

   git clone https://github.com/your-username/your-repo.git

2. Install the required packages:

pip install -r requirements.txt

3. Download the YOLO pre-trained weights file yolov8n.pt and place it in the Yolo-Weights directory.

4. Prepare the input data:

• WebCam: Uncomment the following lines to use the webcam as the input source:

cap = cv2.VideoCapture(0)
cap.set(3, 1280) # width
cap.set(4, 720) # height

• Video: Provide the path to your video file by setting VIDEO_PATH:

cap = cv2.VideoCapture(r"D:\open-cv\yolo-projects\videos\highway-curve.mp4")

1. Run the script python carCounter.py

2. The script will perform object detection and tracking on the input video or webcam feed.

3. Press Esc to exit the program.

• Output Width and Height: You can customize the output frame dimensions by modifying the output_width and output_height variables.

• Mask Overlay: To add a mask overlay to the detected objects, provide the path to the mask image by setting imgMask:

imgMask = cv2.imread(r"D:\open-cv\yolo-projects\images\mask1.png")

• Graphics Overlay: To add a graphical overlay to the detected objects, provide the path to the overlay image by setting imgGraphics:

imgGraphics = cv2.imread(r"D:\open-cv\yolo-projects\images\car-graphics1.png", cv2.IMREAD_UNCHANGED)

• Detection Confidence Threshold: You can adjust the minimum confidence threshold for object detection by modifying the confidence value in the code:

if currentClass in allowedObjects and confidence > 0.3:

• Object Classes: The classNames list contains the names of the object classes that the model can detect. You can modify this list to include or exclude specific classes based on your requirements.

• Tracking Parameters: The Sort tracker parameters can be adjusted by modifying the max_age, min_hits, and iou_threshold values in the tracker = Sort(...) line.

Pull requests are welcome. For major changes, please open an issue first to discuss what you would like to change.

MIT