Feature-rich webapp for semi-automated image segmentation and analysis of pictures of plants, specifically designed for cranberries but extensible to other crops. Derived from https://doi.org/10.1371/journal.pone.0160439.
- Access pictures from cloud-based datastore services such as Google Drive, Box.com, Dropbox, and Cyverse.
- Upload pictures to server for server-based storage.
- Interactive tuning of classical segmentation parameters for immediate feedback of segmentation quality.
- Interactive generation of neural network images for advanced segmentation of images.
- Attach metadata to sets of images, such as dates (year/month), growth degree days, genotypes, location of samples, camera settings, color corrected tags, and lens distortion corrected tags.
- Dashboard for viewing statistics of dataset pools, with the ability to filter based on metadata tags.
This web application is best installed on a Linux or other *Nix-based system with ample resources to adequately service the expected amount of network traffic.