PlantDoc Classification Dataset for Plant Disease Detection

The PlantDoc Classification Dataset is designed for computer vision research focused on plant disease identification. Unlike datasets collected primarily under controlled laboratory conditions, PlantDoc includes real-world plant images with variations in lighting, backgrounds, image quality, and capture conditions.

The classification version contains approximately 2,572 images across 28 classes and covers 13 plant species, including apple, bell pepper, blueberry, cherry, corn, grape, peach, potato, raspberry, soybean, squash, strawberry, and tomato.

As a result, the dataset can help developers build models that handle more realistic agricultural image conditions.

Key Features

• 2,572 plant leaf images

• 28 plant disease and healthy-leaf classes

• 13 plant species

• Real-world plant images

• JPG image format

• Image classification use case

• Suitable for computer vision research

• Useful for machine learning and deep learning

• Supports agricultural AI applications

Plant Species Covered

The dataset includes images from a range of plant species, such as:

• Apple

• Bell Pepper

• Blueberry

• Cherry

• Corn

• Grape

• Peach

• Potato

• Raspberry

• Soybean

• Squash

• Strawberry

• Tomato

These different plant categories allow researchers to develop and evaluate models across multiple agricultural crops.

Disease Classification

PlantDoc contains classes representing both healthy leaves and different plant disease conditions. Examples include Apple Scab Leaf, Apple Rust Leaf, Bell Pepper Leaf Spot, Corn Leaf Blight, Corn Rust Leaf, Potato Early Blight, Potato Late Blight, Tomato Early Blight, Tomato Septoria Leaf Spot, Tomato Late Blight, and Tomato Leaf Bacterial Spot.

Therefore, the dataset can support multi-class image classification models that distinguish between plant species and visible disease conditions.

Applications of the PlantDoc Dataset

The dataset can support several agricultural computer vision applications, including:

• Plant disease classification

• Automated crop disease detection

• Leaf image classification

• Agricultural computer vision

• Precision agriculture research

• Crop monitoring systems

• Deep learning model development

• Image classification benchmarking

• Smart farming applications

• Agricultural AI research

Furthermore, its real-world image conditions make it useful for testing whether computer vision models can generalize beyond controlled environments.

Why Use the PlantDoc Classification Dataset?

Plant disease identification often becomes difficult when images contain different backgrounds, lighting conditions, and image qualities. PlantDoc provides images collected from real-world conditions, making it useful for developing models that need to handle such variation.

Moreover, the dataset covers multiple plant species and disease classes. Therefore, researchers can use it for multi-class classification experiments and agricultural AI projects.

Who Can Use This Dataset?

The PlantDoc Classification Dataset can benefit:

• Computer vision engineers

• Machine learning developers

• Data scientists

• Agricultural AI researchers

• Deep learning practitioners

• Students working on image classification

• Researchers developing plant disease detection systems

• Precision agriculture technology teams

Conclusion

The PlantDoc Classification Dataset provides diverse plant leaf images for developing computer vision and agricultural AI solutions. With multiple plant species and disease classes, it offers useful data for plant disease classification and machine learning research.

Its real-world image conditions also make it valuable for testing models beyond controlled agricultural environments. Therefore, researchers and developers can use PlantDoc for plant disease detection, image classification, computer vision, and precision agriculture applications.

This dataset is sourced form Kaggle