Plant Disease Recognition Using Siamese Network

•Developed a deep learning-based system to classify plant types and diagnose diseases using a

Siamese Network for high accuracy in feature similarity analysis.

•Designed a mobile application with Flutter, providing an intuitive interface for users to upload

plant images and receive instant classification and diagnosis results.

•Applied segmentation techniques to isolate plants from complex backgrounds, enhancing the

quality of input data for the model.

•Integrated GradCAM to generate heatmaps, offering visual explanations of model predictions for

improved interpretability and user trust.

•Deployed the model on a mobile application backend for real-time accessibility and performance

optimization.