Plant Pathology Imaging and Analysis System
Granted · Intellectual property
A 360° leaf imaging bed with anti-glare lighting for deep-learning pathology.
The system pairs a rotating examination bed with a high-resolution camera on a stepper mount, capturing multi-angle leaf imagery under anti-glare lighting. A Raspberry Pi and custom PCB drive the scan.
The related ViTaL paper proposes this imaging arrangement alongside a transformer-based disease-identification pipeline. Capturing a useful image and classifying it are separate engineering tasks: lighting, viewpoint, and repeatability influence what a model has to work with.
The design is recorded as 411009-001. The accompanying hardware renders come from the paper; its numerical results are dataset classification results, not measured performance of an integrated field scanner.
How does a model read a leaf?
Illustrated input
Patch values + where they belong
Extracted features → CNN classifier
Begin with the leaf.
A leaf image contains colour, texture, veins, and other visual patterns. This leaf is an original illustration, not a PlantVillage sample or a diagnosis.
Explore at your own pace.
Conceptual illustration inspired by Vision Transformer and the ViTaL paper. Patch count, colours, connections, and feature bars are simplified. No model runs here; no disease prediction is made.
Record 411009-001 · Status as recorded in the portfolio. Look up the current record at IP India ↗
Inventor’s project and patent descriptions ↗- Design patent
- Imaging
- Agriculture