horshe/agrisentry-wheat-disease-classifier

AgriSentry Wheat Disease Classifier

A CNN-based disease classification model for wheat crops, part of the AgriSentry diagnostic system.

Model Details

Model Description

This is a TensorFlow CNN model trained to classify diseases in wheat plants. It identifies three common wheat diseases and healthy crop status.

• Model type: Convolutional Neural Network (CNN)
• Framework: TensorFlow/Keras
• Task: Image Classification
• Input: RGB images (224x224 pixels)
• Output: Disease classification with confidence scores

Supported Classes

• Healthy: No disease detected
• Brown Rust: Early stage brown rust infection
• Yellow Rust: Yellow rust disease
• Septoria: Septoria tritici blotch

Model Features

• Input resolution: 224x224 RGB images
• Optimized for agricultural field conditions
• Fast inference suitable for mobile/edge deployment
• Confidence scores for each prediction

Usage

Via Hugging Face Inference API

from huggingface_hub import InferenceClient

client = InferenceClient(model="your-username/agrisentry-wheat-disease-classifier")

# Load image and send for inference
with open("wheat_leaf.jpg", "rb") as f:
    result = client.image_classification(f)

print(result)

Via Python (TensorFlow)

import tensorflow as tf
from PIL import Image
import numpy as np

model = tf.keras.models.load_model('wheat_cnn.h5')
img = Image.open('wheat_leaf.jpg').resize((224, 224))
img_array = np.array(img) / 255.0
prediction = model.predict(np.expand_dims(img_array, axis=0))

Training Data

• Collected from agricultural regions
• Augmented with rotation, zoom, and brightness variations
• Balanced across disease classes

Performance

• Accuracy: ~95% on validation set
• Optimized for real-world field images

Limitations

• Best performance with well-lit, close-up leaf images
• May require model fine-tuning for new disease variants
• Designed specifically for wheat crops

License

MIT License - See LICENSE file for details

Citation

If you use this model, please cite:

@model{agrisentry_wheat_2026,
  title={AgriSentry Wheat Disease Classifier},
  author={AgriSentry Team},
  year={2026}
}

Disclaimer

This model is a diagnostic tool and should not replace professional agricultural consultation. Always verify results with domain experts.