# ⚛  L1 Principle — Automated Optical Inspection (AOI) for PCB / manufacturing defects

**ID:** `L1-117` · **Status:** ⊙ Testnet (genesis catalog)

> **🌐 Domain:** Industrial Inspection — *Vision-based industrial quality control*
> **🎯 Problem class:** linear inverse · **🧮 Solution space:** 2D defect mask
> **📡 Carrier:** photon · **🌫 Noise:** gaussian
> **⚖ Difficulty (δ):** 1 · **⛓ Block:** 41554200

---

## 🧠 1. Introduction

**Automated Optical Inspection (AOI) for PCB / manufacturing defects** is a **linear inverse problem** whose unknown lives in **2D defect mask** space, within the **Vision-based industrial quality control** sub-domain of **Industrial Inspection**.

Measurements consist of photons collected by an optical detector via a **multi angle vision** sensing mechanism.

The forward operator applies, in order: L · illuminate multi angle operator; convolution with the Airy disk of a circular aperture; D · image sensor operator; pixel-level spatial averaging on the detector.

Observations are corrupted by additive Gaussian noise. Existence of the recovered 2D defect mask is guaranteed within the declared Omega bounds. Uniqueness holds on the measurement-supported subspace; out-of-support modes are controlled by the declared priors. Stability is well-conditioned (kappa_eff ~= 5); illumination_variation dominates the stability cliff; color_shift and the remaining mismatch parameters contribute higher-order bias terms. Additive gaussian thermal/electronic noise sets the irreducible data-fidelity floor, while mild Tikhonov or analytic inversion is sufficient at the nominal Omega point.

## ⚙ 2. Forward Model

Physical chain: **x** → L · illuminate multi angle → Airy PSF convolution → D · image sensor → Spatial integration → **y** (detector).

```
y = ∫_A dA `D.image_sensor` K_Airy * `L.illuminate_multi_angle` x + n,    n ~ 𝒩(0, σ²)
```

**Measurement DAG:**

| Primitive | What it does |
|---|---|
| `L.illuminate_multi_angle` | L · illuminate multi angle operator |
| `K.psf.airy` | Convolution with the airy disk of a circular aperture |
| `D.image_sensor` | D · image sensor operator |
| `int.spatial` | Pixel-level spatial averaging on the detector |

## 🔬 3. Physics Fingerprint

| Property | Value |
|---|---|
| Domain | Industrial Inspection |
| Sub domain | Vision-based industrial quality control |
| Carrier | photon |
| Problem class | linear_inverse |
| Solution space | 2D_defect_mask |
| Noise model | gaussian |
| Integration axis | spatial |
| Difficulty delta | 1 |
| L dag | 2.5 |

## 📡 4. Measurement Model

Existence of the recovered 2D defect mask is guaranteed within the declared Omega bounds. Uniqueness holds on the measurement-supported subspace; out-of-support modes are controlled by the declared priors. Stability is well-conditioned (kappa_eff ~= 5); illumination_variation dominates the stability cliff; color_shift and the remaining mismatch parameters contribute higher-order bias terms. Additive gaussian thermal/electronic noise sets the irreducible data-fidelity floor, while mild Tikhonov or analytic inversion is sufficient at the nominal Omega point.

| Metric | Value |
|---|---|
| Metric | PSNR_dB |
| Secondary | SSIM |

## 📏 5. Operating Range (Ω)

**Center problem class:** `aoi_vision` · **Forward operator:** `aoi_vision_forward`

**Center point:**

| Parameter | Unit | Value |
|---|---|---|
| H | px | 2448 |
| W | px | 2048 |
| Snr db | dB | 35 |
| N lights | — | 4 |
| Pixel um | µm | 5 |
| Color shift | — | 0 |
| Part misalignment | — | 0 |
| Illumination variation | — | 0 |

**Allowed bounds:**

| Parameter | Unit | Range |
|---|---|---|
| H | px | 512 – 8192 |
| W | px | 512 – 8192 |
| Snr db | dB | 10.0 – 45.0 |
| N lights | — | 1 – 16 |
| Pixel um | µm | 1 – 50 |
| Occlusion | — | 0.0 – 0.2 |
| Color shift | — | 0.0 – 0.2 |
| Part misalignment | — | 0.0 – 5.0 |
| Illumination variation | — | 0.0 – 0.3 |

## 🎯 6. Tolerance (ε)

**Center tolerance:** 35.0

| Metric | Range |
|---|---|
| Psnr db | 5.0 – 40.0 |

## ⚖ 7. Hardness Function

Hardness scales as **`epsilon_fn`** on **PSNR_dB**, with κ = `100` and δ = `1`.

## 💾 8. Reference Dataset

- **primary** · weight 1.0 · IPFS _(not pinned yet)_

## 9. On-chain Registration

- **Chain hash:** `0xad7cccc3a0d7f68b125901e1e7b12bcb4a3f248b2f4202470277003eacbb601d`
- **Chain tx hash:** `0x4d189074f3d946bd9c245c86c5df21705f30a383c51fb9e3a79b1fe37b6037b5`
- **Chain block:** `41554200`

---

## File Mapping

This bundle consists of: `L1-117.md`, `L1-117.json`.

| File | Role | How to regenerate |
|------|------|-------------------|
| `L1-117.md` | Source of truth — edit this | Human or LLM |
| `L1-117.json` | Structured metadata for the registry | LLM regenerates from the sections above |

**Prompt for your LLM after editing this Markdown:**

> Read the attached Markdown. Regenerate the sibling `.json` so every field matches.
> Preserve the schema documented in the rows above.
> Output each file in its own fenced code block tagged with the filename.
> Output only the JSON object.

_This Markdown was auto-synthesized from the catalog row for `L1-117`._
_Edit it, regenerate the JSON, and submit at [/submit](/submit) to claim the artifact._