VrikshAI: 87-Class Dataset for Crop Pathology and Anomaly Detection

87 classes of images for detecting crop diseases and anomalies. The dataset is hosted on Kaggle and focuses on adversarial anomaly detection in agricultural contexts. The author, organization, and specific collection details are not provided.

Use Cases

• Train image classification models for crop disease identification based on the 87-class structure.
• Develop anomaly detection systems for agricultural monitoring based on the described adversarial anomaly detection focus.
• Benchmark model robustness against adversarial examples in plant pathology imagery.
• Create educational tools for farmers or agronomists to recognize plant diseases.

Strengths

• Covers 87 distinct classes for crop pathology.
• Designed for adversarial anomaly detection, a specific machine learning task.

Limitations

• Description metadata is limited; actual data quality requires manual inspection after download.
• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.

Provenance

Source

Kaggle