The corpus behind the world's #1 crop-diagnosis app — and one of the most-referenced resources in digital-agriculture research.
Since 2016, farmers have sent Plantix more than 150M+ real field images — labeled, geotagged, and validated at scale. That corpus has been referenced in hundreds of peer-reviewed studies, from crop-disease modeling to computer-vision benchmarks.
Plantix. PEAT GmbH. https://plantix.netIf your work references Plantix, cite the app and PEAT GmbH, and link to plantix.net. For dataset details in your methods section, this page is the canonical reference for corpus scale and coverage.
district-level outbreak dynamics from real diagnoses
field-condition imagery no lab dataset can substitute
adoption and advisory research at smallholder scale
Plantix and its founders in the press and research community.
Studies we've collaborated on — using the Plantix diagnosis corpus as ground truth, often with academic partners like Stanford and ZALF.
| Paper | Access | How Plantix is used |
|---|---|---|
| Mapping Crop Types in Southeast India with Smartphone Crowdsourcing and Deep Learning | Open access | The Plantix app logged ~9M geolocated farmer photos in India (2016–2019), whose expert/CNN-derived crop labels trained the crop-type classifier. |
| Mapping Sugarcane in Central India with Smartphone Crowdsourcing | Open access | Plantix-crowdsourced farmer crop photos served as ground-truth to train a supervised neural-net sugarcane classifier for the Bhima Basin. |
| Biotic Yield Losses in the Southern Amazon, Brazil: Making Use of Smartphone-Assisted Plant Disease Diagnosis Data | Open access | Joint ZALF–PEAT study using ~78,000 georeferenced Plantix pest/disease images (of >1M captured in Brazil) to map disease distribution and biotic yield losses. |
Peer-reviewed studies and preprints that reference or evaluate Plantix, independent of our team.
| Paper | Access | How Plantix is used |
|---|---|---|
| Evaluating Plant Disease Detection Mobile Applications: Quality and Limitations | Open access | Of 17 apps evaluated, Plantix scored highest overall (4.56/5) — the only one to identify plants, detect disease and suggest treatment. |
| Smartphone Applications Targeting Precision Agriculture Practices—A Systematic Review | Open access | Profiles Plantix (PEAT GmbH, Berlin) as a deep-learning image-recognition diagnosis app detecting ~400 damage types across 30 crops with treatment guidance. |
| Farmer.Chat: Scaling AI-Powered Agricultural Services for Smallholder Farmers | Open access | States that 'disease diagnostics from Plantix are integrated,' enhancing the AI advisory platform for smallholder farmers. |
| Do Digital Climate Services for Farmers Encourage Resilient Farming Practices? Pinpointing Gaps through the Responsible Research and Innovation Framework | Open access | Includes Plantix (developed by AgriTech startup PEAT) as a crop-disease-detection case study in the RRI-framework analysis of climate/advisory apps. |
| Understanding User Perceptions of Gardening Apps Supporting Sustainability | Open access | Analyzes 180,000+ app-store reviews (Plantix among the studied apps) via contextualized topic modelling; notes it gives users strong motivation. |
| A System for Automatic Rice Disease Detection from Rice Paddy Images Serviced via a Chatbot | Open access (preprint) | Introduction lists Plantix among existing apps that identify pests and plant diseases on over 30 plants. |
| Review on Convolutional Neural Network (CNN) Applied to Plant Leaf Disease Classification | Open access | Notes PEAT (Berlin) built the Android app Plantix, which supports farmers with disease identification. |
| An AI solution for Soil Fertility and Crop Friendliness Detection and Monitoring | Open access | Cites Plantix (PEAT, Berlin start-up) as a deep-learning app correlating foliage patterns with soil/nutrient deficiencies. |
This is a selected sample, not a complete list. Citations verified 2026-07-07.
Research access is collaboration-based — we don't sell the raw corpus. If your work needs crop-health ground truth at this scale, talk to us.
One data engine — three ways to plug your business into the world's largest crop-health signal.
Embed the diagnostic brain of the world's #1 crop app inside your product.
District-level demand signals from millions of live diagnoses.
Your product in front of the farmer at the exact second of diagnosis.