Leaf Disease Dataset for Smart Agriculture Applications

🌿 Wheat Leaf Disease Dataset for Deep Learning Applications

This dataset provides a curated collection of labeled wheat leaf images designed for the development and evaluation of machine learning and deep learning models for plant disease classification. It captures real-world agricultural variability and supports robust model training under diverse environmental conditions.

📦 Dataset Overview

The dataset consists of high-quality images of wheat leaves categorized into the following classes:

BlackPoint

FusariumFootRot

HealthyLeaf

LeafBlight

WheatBlast

Each class represents distinct visual symptoms observed in crop pathology, enabling effective supervised learning for multi-class classification tasks.

🧠 Data Characteristics

Collected from real-world agricultural environments

Includes variations in lighting, background, and leaf orientation

Contains natural noise and occlusions, reflecting practical deployment scenarios

Manually curated and verified for label consistency and quality

⚙️ Data Preparation

Images were cleaned and filtered to remove low-quality samples

Dataset structured into training, validation, and testing subsets

Preprocessing steps include:

Image resizing

Normalization

Class balancing (where applicable)

🚀 Applications

This dataset is suitable for a wide range of applications, including:

AI-based crop disease detection systems

Precision agriculture and smart farming solutions

Automated plant health monitoring

Decision-support systems for farmers and agronomists

🧪 Benchmark & Usage

The dataset supports development and evaluation of:

Convolutional Neural Networks (CNNs)

Vision Transformer (ViT) architectures

Hybrid CNN–Transformer models

It can be used for:

Multi-class image classification

Model benchmarking and comparison

Academic research and experimentation

Real-world deployment pipelines

🔗 Associated Resources

Kaggle Dataset:https://www.kaggle.com/datasets/khanaamer/wheat-leaf-disease-dataset

GitHub Repository (Model + Implementation):https://github.com/Aamer-Gituser/Leaf-Disease-Detection

📈 Research Context

This dataset has been utilized in a deep learning pipeline integrating CNN and Vision Transformer architectures, along with interpretability techniques such as Grad-CAM and deployment via a web-based interface.

📜 License and Usage

This dataset is released for academic and research purposes.Users are encouraged to cite the dataset appropriately when used in publications or projects.