Cucumber_Disease_working.1.mp4
This repository contains the Cucumber Disease Diagnostic Lab, a deep learning-powered platform designed to identify common diseases in cucumber leaves. By utilizing state-of-the-art Convolutional Neural Networks (CNNs) and Hybrid architectures, the system provides real-time diagnostic reports and actionable treatment plans to assist in crop management.
- Disease Classification: Identifying conditions such as Anthracnose, Bacterial Wilt, Downy Mildew, and Gummy Stem Blight.
- Hybrid Architectures: Combining features from multiple backbones (e.g., ResNet50 + InceptionV3) for enhanced diagnostic accuracy.
- Real-time Inference: Fast image processing and probability estimation using PyTorch.
- Actionable Insights: Providing specific treatment recommendations based on the detected disease.
The application provides a comprehensive diagnostic workflow:
- Architecture Selection: Users can choose from standard models (ResNet50, InceptionV3, EfficientNetB0, MobileNet) or custom Hybrid models.
- Diagnostic Report: Generates a high-confidence prediction, a probability distribution chart across all categories, and a guided "Action Plan" for recovery.
Ensure you have Python 3.11+ installed. The system requires a models/ directory containing .pth weight files for the chosen architectures.
Clone the repository and install the required dependencies:
pip install -r requirements.txt