/leaf-disease-detector

Leaf Disease Detector for vine leaves by using Python and a Support Vector Machine classifier.

Primary LanguagePythonMIT LicenseMIT

Leaf Disease Detector

SVM Model:

Classifies leaves in 2 classes (healthy and rotten).

Followed Steps:

  • Image is transformed to HSV (improve the RGB model).
  • Image segmentation to identify rotten parts of the leaf.
  • Feature extraction is performed for every sample that is available for training.
  • Training of the SVM classifier performs, generating two classes (healthy leaves 1 class, rotten leaves 2 class).

Datasets

  • Datasets consists of apple laf images.
  • Training Images: 2000 images (1000 healthy leaf images) and (1000 rotten leaf images).
  • Testing Images: 40 images for testing (20 healthy leaf images) and (20 rotten leaf images).

License

This project is licensed under the MIT License - see the LICENSE file for details.