Leaf Image Segmentation

Objective

• Implement a technique that can segment(extract) leaf from an image containing leaf

Resources

• Inspired by the paper: Automatic Leaf Extraction from Outdoor Images

Usage

• python3 segment.py

  Command structure

  
  usage: segment [-h] [-m MARKER_INTENSITY] [-f {no,flood,threshold,morph}] [-s]
             [-d DESTINATION]
             image_source

positional arguments: image_source A path of image filename or folder containing images

optional arguments: -h, --help show this help message and exit -m MARKER_INTENSITY, --marker_intensity MARKER_INTENSITY Output image will be as black background and foreground with integer value specified here -f {no,flood,threshold,morph}, --fill {no,flood,threshold,morph} Change hole filling technique for holes appearing in segmented output -s, --smooth Output image with smooth edges -d DESTINATION, --destination DESTINATION Destination directory for output image. If not specified destination directory will be input image directory

### Examples:

__Command used__: `python3 segment.py 'some file or folder'`

__Input Images__

![alt Healthy Apple Leaf](testing_files/apple_healthy.JPG) ![alt Apple Leaf with Black Rot](testing_files/apple_black_rot.JPG)

__Output Images__

![alt Segmented Healthy Apple Leaf](testing_files/apple_healthy_marked.JPG) ![alt Segmented Apple Leaf with Black Rot](testing_files/apple_black_rot_marked.JPG)