Tea Leaf Visual Degradation Dataset
Description
The Tea Leaf Visual Degradation Dataset (TLVDD) is a curated image dataset developed for research on continuous visual degradation assessment of tea leaves. The dataset contains 160 RGB images (224 × 224 pixels) representing three naturally occurring degradation stages: Fresh (53 images), Weathered (56 images), and Dry (51 images). Images were collected from tea-growing and tea-processing environments in Assam, India, under controlled conditions using a plain white background. The dataset preserves natural variations in leaf color, texture, shape, size, and orientation. It provides categorical labels only (Fresh, Weathered, and Dry) and is intended for research in computer vision, deep learning, agricultural quality assessment, and continuous degradation estimation.
Files
Institutions
- Indian Institute of Technology GuwahatiAssam, Guwahati