Time-Series Image Dataset of Pomegranate Fruits with Corresponding Environmental Data
Description
This dataset presents a longitudinal, multi-modal, time-series collection of pomegranate fruit disease progression under natural orchard conditions. It also integrates weekly high-resolution images with synchronized microclimatic data. This dataset collection is performed in Goudwadi, Taluka Sangola, District Solapur, Maharashtra, India. Here, ten trees (Tree 01–Tree 10) were selected, and individual fruits were tracked weekly from healthy to diseased stages. All images were captured using a Vivo V25 Pro (64 MP) smartphone at a 1:1 ratio under natural daylight to ensure realistic texture and illumination conditions. Environmental readings were continuously recorded via an on-site weather station positioned at canopy height, providing accurate microclimate measurements (temperature, humidity, dew point, irradiance, wind speed, precipitation, and pressure). Each record (image) is time-synchronised and also its respective weather data is mentioned with image name. This dataset supports studies in plant disease prediction, environmental–disease correlation, and precision agriculture modelling.
Files
Steps to reproduce
1. Have a good camera with minimum 64 MP 2. Find orchard having more that 10 trees of pomegranate. Locate 2-3 orchards for best results. 3. On a fixed day in every week, visit each orchard and capture image of selected pomegranate fruit of each tree. Calculate and store temperature, relative humidity, dew point, solar radiation (GHI, DHI, DNI), wind speed, pressure, precipitation, visibility these parameters too. 4. Label and store them in given structure.
Institutions
- Vishwakarma Institute of Technology