Sting Operation AI: Bee and Wasp Detection

Coastal Alpine Tech portfolio

Part of the Kiwi Edge AI Stack | Founder OS: NZ-Start-Up | Agent policy: .github/agent-fleet/

Sovereign hybrid edge AI for NZ farms and founders - local-first + multi-model, Te Mana Raraunga aligned - collaborating with Venture Taranaki, startups.com investors and Kotahitanga Investment Fund (HITL + cultural advisory for formal approaches).

Agents inform, draft, prepare, monitor, and remind. Humans advise, sign, file, send, and pay. Anti-hallucination policy: .github/agent-fleet/anti-hallucination.md | Congruence: CAT_CONGRUENCE.md

Privacy / Security / Governance

Coastal Alpine Tech products treat operational and personal data as taonga. Defaults favour local-first operation, purpose-limited collection, and Human-in-the-Loop for high-stakes actions.

Hard commitments

Commitment	Statement
No data sales	We do not sell personal information or customer operational data to third parties for advertising, brokerage, or unrelated commercial exploitation.
NZ Privacy Act 2020	Collection, use, storage, and disclosure of personal information is designed to operate in accordance with the Privacy Act 2020 information privacy principles (including IPP awareness and IPP 3A indirect-collection notification where applicable).
Te Mana Raraunga	Where Māori data or community data interests arise, systems are designed to operate in accordance with Te Mana Raraunga principles (including Rangatiratanga, Whakapapa, Whanaungatanga, Kotahitanga, Manaakitanga, Kaitiakitanga) as a sovereignty and stewardship lens — not as a marketing slogan.
NZ AI safety	AI features follow a NZ AI safety-aligned posture: Algorithm Charter spirit (fairness, transparency, human oversight where relevant), digital.govt.nz / responsible AI guidance awareness, no silent model training on private journals without consent, and HITL for high-stakes outcomes.
Security	No silent exfiltration; owner-controlled credentials; least privilege; SecOps / dependency hygiene on the fleet cadence.
Governance	Agents inform, draft, prepare; humans advise, sign, file, send, and pay.

Pillar	Commitment
Privacy	Local-first / offline-capable where practical; Privacy Act 2020; Te Mana Raraunga spirit; third-party AI only when opt-in and labelled
Security	No silent exfil of tenant or personal data; owner-controlled keys
Governance	HITL for high-stakes; Te Mana Raraunga spirit; multi-region compliance maps in COMPLIANCE_REGIONS.md

Agents inform, draft, prepare, monitor, and remind. Humans advise, sign, file, send, and pay.

Fleet policy: fivepanelhat / Kiwi Edge AI Stack · COMPLIANCE.md · COMPLIANCE_REGIONS.md · SECURITY.md

Problems we are solving

Sting-Operation-AI is edge vision for apiculture biosecurity - wasp vs bee discrimination on-device.

1. Invasive wasp pressure - German wasps and related threats hit NZ honeybee operations hard.
2. Cloud vision latency / privacy - Apiaries need on-site inference, not a remote API round-trip.
3. Label noise - Weak datasets and class confusion destroy field trust.
4. Sovereign biosecurity data - Hive imagery and detections should stay under operator control.

Solution we have built

Built capability	What it does
YOLO multi-class detection	Honeybee / German wasp / hornet classes
Dataset hygiene tools	Cleanup and label correction paths
Edge-ready training/inference	RPi 5 + Hailo-10H target
Stack fit	Biosecurity sister to Byte Size Kai beachhead

Local (Taranaki) and national (Aotearoa) economic benefits

Lever	Benefit
Regional R&D HQ	Product design and IP stay in New Plymouth / Taranaki - not only Auckland/offshore SaaS
Primary-sector productivity	On-farm and rural tools aim to cut waste, protect consents, and support export competitiveness
Skilled employment pathways	Edge install, field support, agritech ops, software, compliance, and cultural advisory roles as pilots scale
Data sovereignty	Te Mana Raraunga-aligned local custody keeps high-value operational data onshore
HITL jobs quality	Agents inform / draft / prepare / monitor / remind; humans advise / sign / file / send / pay - augment people, do not fake full autonomy

Stage honesty (pre-seed): Impact today is founder R&D, near-term contractors, and EDA/partner leverage. Permanent multi-region payroll follows paid pilots and revenue - we do not invent headcount claims.

Coastal Alpine Tech Limited pre-seed startup, New Plymouth, Taranaki, Aotearoa New Zealand. Edge AI | Sovereign Systems | Practical Intelligence

Object detection system for protecting beehives by identifying honeybees versus invasive wasps using YOLO models and edge AI.

---

The 5 Ws: Project Context

• Who: Built by Coastal Alpine Tech Limited for New Zealand apiarists and biosecurity efforts.
• What: A YOLO-based multi-class object detection pipeline focused on accurate differentiation between honeybees and invasive wasp species.
• Where: Engineered at HQ in New Plymouth, Taranaki. Designed for on-premise and edge deployment.
• When: Active development as of June 2026.
• Why: To deliver localized data sovereignty and real-time protection for beehives without reliance on cloud services.

---

---

Key Features

• Multi-class YOLO object detection (Honeybee, German Wasp, Yellow-legged Hornet)
• Automated dataset cleanup and label correction tools
• Training and inference scripts with hardware acceleration support
• Roboflow dataset integration and validation
• Edge AI ready for Raspberry Pi 5 + Hailo-10H NPU
• Servo tracking and actuator integration potential

---

Quick Start

Prerequisites

• Python 3.10+
• Ultralytics YOLO
• Optional: Raspberry Pi 5 with Hailo-10H NPU for edge inference
• GPU recommended for training (CUDA support)

Installation & Setup

We provide separate guides for system environment setup and installation for Windows and Linux users:

• Prerequisites & System Setup Guide: Read setup.md
• Installation Guide: Read installation.md

Quick Start (Automated Setup)

The fastest way to install is running the cross-platform bootstrap script:

python bootstrap.py

ation-AI python bootstrap.py

### Manual Installation

<details open>
<summary><strong> Linux / macOS (Bash)</strong></summary>

```bash
git clone https://github.com/fivepanelhat/Sting-Operation-AI.git
cd Sting-Operation-AI

python3 -m venv venv
source venv/bin/activate

# Install shared core and dependencies
pip install git+https://github.com/fivepanelhat/coastal-alpine-core.git
pip install -r requirements.txt
pip install -r requirements-dev.txt
cp .env.example .env # If applicable

Windows (PowerShell)

git clone https://github.com/fivepanelhat/Sting-Operation-AI.git
cd Sting-Operation-AI

python -m venv venv
.\venv\Scripts\Activate.ps1

# Install shared core and dependencies
pip install git+https://github.com/fivepanelhat/coastal-alpine-core.git
pip install -r requirements.txt
pip install -r requirements-dev.txt
Copy-Item .env.example .env # If applicable

Note: If you receive an execution policy error, run Set-ExecutionPolicy -Scope CurrentUser RemoteSigned first.

Setup & Validation

# Windows automated setup
setup_project.bat

# Manual cleanup and verification
python tools/tidy_and_fix.py
python tools/verify_setup.py

Inference Example

python predict.py data/images/val/

---

Architecture Overview

Diagrams: Architecture images and Mermaid maps describe the target product architecture for this pre-seed stack. They are engineering design maps not claims of large-scale commercial fleet deployment.

Sting Operation protects hives with real-time bee vs wasp vision on the edge. YOLO detection runs on Hailo-10H; higher-level decisions use local Ollama on RPi 5 16GB no cloud video upload.

System map

%%{init: {
 "theme": "dark",
 "themeVariables": {
 "fontSize": "16px",
 "fontFamily": "Inter, ui-sans-serif, system-ui, sans-serif",
 "primaryColor": "#0ea5e9",
 "primaryTextColor": "#f8fafc",
 "primaryBorderColor": "#38bdf8",
 "lineColor": "#67e8f9",
 "secondaryColor": "#1e293b",
 "tertiaryColor": "#0f172a",
 "clusterBkg": "#0b1220cc",
 "clusterBorder": "#38bdf880",
 "titleColor": "#e2e8f0"
 },
 "flowchart": {
 "nodeSpacing": 40,
 "rankSpacing": 48,
 "padding": 20,
 "htmlLabels": true,
 "curve": "basis"
 }
}}%%
flowchart TB

 classDef sense fill:#052e16,stroke:#4ade80,stroke-width:2px,color:#f0fdf4
 classDef edge fill:#0c4a6e,stroke:#38bdf8,stroke-width:2px,color:#f0f9ff
 classDef core fill:#134e4a,stroke:#2dd4bf,stroke-width:2px,color:#f0fdfa
 classDef act fill:#422006,stroke:#fbbf24,stroke-width:2px,color:#fffbeb
 classDef store fill:#1e1b4b,stroke:#a5b4fc,stroke-width:2px,color:#eef2ff
 classDef ai fill:#3b0764,stroke:#e879f9,stroke-width:2px,color:#fdf4ff
 classDef app fill:#1e1b4b,stroke:#c4b5fd,stroke-width:2px,color:#eef2ff

 CAM["CSI / stream video"] --> YOLO["YOLO detection<br/>bee | wasp | hornet"]
 YOLO --> HEF["Hailo-10H NPU<br/>HEF / INT8 inference"]
 HEF --> MAP["Class mapping & tracks"]
 MAP --> LLM["Optional Ollama reasoning<br/>event logging"]
 MAP --> ACT["Actions<br/>alerts | servo | relays"]

 subgraph EDGE["Sovereign edge RPi 5 16GB + Hailo-10H"]
 YOLO
 HEF
 MAP
 LLM
 end

 class CAM sense
 class YOLO,MAP core
 class HEF,LLM ai
 class ACT act

Loading

Layer	Components	Role
Vision	YOLO multi-class	Bee / wasp / hornet
NPU	Hailo-10H 40 TOPS	Real-time edge FPS
Reasoning	Ollama optional	Event narrative / logs
Actuation	Alerts	servo

Full detail: ARCHITECTURE.md | docs/

Directory Structure

Sting-Operation-AI/
|-- config/ # data.yaml and configurations
|-- data/ # images, labels, raw annotations
|-- models/ # base_weights and trained_models
|-- tools/ # tidy_and_fix.py, verify_setup.py
|-- predict.py
|-- train.py
|-- setup_project.bat
|-- .github/workflows/ # CI/CD
-- README.md

---

Technology Stack

Hardware

• Raspberry Pi 5 + Hailo-10H NPU
• Camera modules and potential servo/relay actuators

Software

• Detection: Ultralytics YOLO
• Orchestration: Local scripts with optional LangGraph / Ollama
• Dataset: Roboflow integration
• Deployment: Edge-ready with systemd/Docker support

---

Real-World Examples and Implementation

• Beehive Protection in New Zealand Apiaries: Deployed at hive entrances to detect and trigger alerts or deterrents when invasive wasps approach, protecting local honeybee colonies.
• Biosecurity Monitoring: Used by regional councils or commercial beekeepers for early warning of Yellow-legged hornet incursions.
• Research and Training: Integrated into educational programs or pest management studies with custom model retraining.

Implementation Notes

• Run setup_project.bat or manual verification tools to ensure correct class mappings.
• Train or fine-tune models using train.py with your expanded dataset.
• Deploy inference via predict.py on edge hardware; integrate with camera streams and actuators per the hardware guide in docs/.
• Combine with Gemma 4 via Ollama for higher-level reasoning (e.g., logging events or deciding response actions).
• Monitor performance with validation images and iteratively improve wasp detection accuracy.

---

Performance & Benchmarks

• Inference Latency: ~12.5ms per frame processing YOLOv8 on Raspberry Pi 5 + Hailo-10H NPU.
• Energy Consumption: Peak Hailo-10H NPU draw is ~2.1W under continuous 30 FPS inference.
• Model Accuracy: German Wasp (Vespula germanica) mAP50 ~84.6%, Precision 84.2%, Recall 82.1%; Honeybee (Apis mellifera) mAP50 100%.

---

Documentation

• Edge AI & IoT Hardware Setup Guide
• ARCHITECTURE.md Detailed system design
• CHANGELOG.md Version history

---

License

This project is licensed under the Coastal Alpine Tech Limited License see the LICENSE file for details.

---

Built with focus on data sovereignty and edge intelligence. Questions or collaboration? Contact Coastal Alpine Tech Limited.

---

Last updated: June 2026.