Livestock biometric analysis platform — non-invasive weight estimation and breed identification for cattle.
PondiFarm is a research and development project by Euro Union Consult (Portugal), targeting the European livestock industry with mobile-first biometric tools for cattle.
This repository hosts the PondiFarmApp monorepo: a React Native / Expo client paired with a FastAPI backend, together implementing the Phase 0 validation pipeline for body measurement and weight estimation from images and 3D scans.
PondiFarmApp/
├── backend/ FastAPI + scikit-learn weight prediction (Random Forest)
├── mobile/ Expo SDK 54 + React Native 0.81.5 (TypeScript)
├── .github/ CI workflows, release-please, code scanning, templates
└── … Docs, LICENSE, governance files
- See
backend/README.mdto run the API locally. - See
mobile/README.mdto run the app locally.
Sprint 1 — native LiDAR scanner + weight pipeline (v0.5.0, June 2026).
The goal is an end-to-end pipeline (LiDAR 3D scan → measurements → weight estimate) validated against ground-truth scale weights from field sessions with partner farms.
Field session #1 — Limousine herd, Beira Interior, Portugal — was conducted on 2026-06-08 with Prof. Joaquim Carvalho (IPCB).
3D capture now uses a proprietary on-device LiDAR scanner (Apple ARKit scene reconstruction) built into the app, replacing the Polycam Pro external tool used during early prototyping. This gives full control over acquisition, mesh quality, and metadata.
This project succeeds the private Tcordeir0/boviscan repository (Phase 0 demo, archived). Prior changelog entries from boviscan are not reproduced here — release tracking restarts at 0.0.0 for the public Euro Union Consult release.
Internal contributors only at this stage. See CONTRIBUTING.md for commit conventions, branch strategy, and the pull request process.
External contribution requests are evaluated on a case-by-case basis — open a Discussion or contact the maintainers before submitting a pull request.
If you discover a security issue, please follow the responsible disclosure process described in SECURITY.md. Do not open a public issue for security reports.
Proprietary — © 2026 Euro Union Consult, Lda. All rights reserved.
This source code is source-available, not open-source. It is published for transparency, peer review, and academic and regulatory inspection. No reuse, redistribution, modification, or derivative work is permitted without prior written authorization from Euro Union Consult.
See LICENSE for the full proprietary terms.
Maintained by Euro Union Consult, Lda. — Porto, Portugal.