Object Detection with YOLOv8: A Guide to Using Pretrained Models on Images

We will explore how to use the YOLOv8 model for object detection on images. YOLO (You Only Look Once) is one of the most popular object detection algorithms, and YOLOv8 is the latest version that offers improved performance and accuracy. We will utilize the ultralytics library to run object detection on images and process the detection results.

Prerequisites

Make sure you have:

• Python
• The ultralytics library

If you haven’t installed ultralytics, you can do so with pip:

pip install ultralytics

Object Detection Code

Here is an example of Python code that uses YOLOv8 to detect objects in an image:

from ultralytics import YOLO

# Load a model
model = YOLO("yolov8n.pt")  # pretrained YOLOv8n model

# Run batched inference on a list of images
results = model(["image/car.jpg"])  # return a list of Results objects

# Process results list
for result in results:
    boxes = result.boxes  # Boxes object for bounding box outputs
    masks = result.masks  # Masks object for segmentation masks outputs
    keypoints = result.keypoints  # Keypoints object for pose outputs
    probs = result.probs  # Probs object for classification outputs
    obb = result.obb  # Oriented boxes object for OBB outputs
    result.show()  # display to screen
    result.save(filename="result.jpg")  # save to disk

Code Explanation

1. Loading the Model:

   • model = YOLO("yolov8n.pt") loads the trained YOLOv8 model. Ensure the yolov8n.pt model file is available in your working directory or download it from the YOLO repository.

2. Running Inference:

   • results = model(["image/car.jpg"]) runs object detection on the image. You can replace "image/car.jpg" with the path to another image you want to process. The model will return a list of detection results.

3. Processing Detection Results:

   • boxes = result.boxes accesses the bounding boxes from the detection results.
   • masks = result.masks retrieves segmentation masks if your model supports segmentation.
   • keypoints = result.keypoints accesses keypoints if you’re performing pose detection.
   • probs = result.probs stores the classification probabilities of the objects.
   • obb = result.obb accesses oriented boxes if used.

4. Displaying and Saving Results:

   • result.show() displays the image annotated with object detections on the screen.
   • result.save(filename="result.jpg") saves the detection results image to disk with the filename result.jpg.

Conclusion

By using YOLOv8 and the ultralytics library, you can easily perform object detection on images and process the results. YOLOv8 offers high accuracy and fast detection speed, making it ideal for applications requiring real-time or batch object detection. This code can be further developed to handle various input types and customize the detection results to suit your needs.