• Install requirements and download pretrained weights in Terminal

$ pip3 install -r ./docs/requirements.txt
$ wget https://pjreddie.com/media/files/yolov3.weights

• numpy>=1.16.0
• pillow==6.2.0
• scipy==1.1.0
• wget==3.2
• seaborn==0.9.0
• easydict==1.9
• grpcio>=1.24.3
• tensorflow==2.0.0.

• The images should be stored in the /docs directory.
• Change the image path before running imagedetection
• Open image_demo.py and change the image path using image_path = "./docs/DSC03040.jpg"
• Due to copyright issues, only sample images were included.

In this part, we will use pretrained weights to make predictions on seabird images.

• The neural network for the YOLOv3 model is implemented in yolov3.py

$ python image_demo.py

• The sample output shows the predicted bounding boxes.
• Further statistical analysis would provide deeper insights into the classification accuracy of object detection.