PCB Inspection · Vision AI · Publication System

PCB Visual Inspection Studio

A PCB inspection project that connects F411-board visual intake, quality gating, ROI alignment, PatchCore/FR-PatchCore, DINOv3-style features, paper figures, and public project publishing.

Paper-style figures and homepage integrated
PCB capture and quality-gate figure with separated camera, video, and ROI panels

01

Problem

The project turns camera intake, ROI alignment, anomaly heatmaps, masks, tables, paper figures, and homepage publishing into one coherent PCB inspection publication structure.

02

Architecture

USB/industrial camera or file intake → quality gate → ROI alignment → ViT/DINO/PatchCore-style features → anomaly score/heatmap/mask → reviewer gate → paper, GitHub, Notion, and homepage publishing.

03

Hardware and Software

  • Python
  • OpenCV
  • PyTorch
  • CUDA
  • PatchCore
  • FR-PatchCore
  • DINOv3
  • Jekyll
  • GitHub Pages
  • Notion

04

Data Flow

Raw frames and ROI evidence are kept separate, and each figure is reused as an independent square panel rather than a bundled contact sheet.

05

Methodology

  • Extracted layout grammar from the supplied figure bundles using a source-checked reference-intake route
  • Redrew text boxes, axis labels, arrows, color bars, and centered panels as square publication figures
  • Configured the homepage gallery around individually clickable enlarged images

06

Results

  • Separated ten PCB methodology figures into independent cards
  • Mirrored the local 8765 portal into a public GitHub Pages route
  • Verified click-to-enlarge behavior for the first methodology figure in the browser
  • Published the v9 gallery cache-bust and separated figure references in the PCB repo and public hub
01Problem

The project turns camera intake, ROI alignment, anomaly heatmaps, masks, tables, paper figures, and homepage publishing in...

02Architecture

USB/industrial camera or file intake → quality gate → ROI alignment → ViT/DINO/PatchCore-style features → anomaly score/he...

03Method

Extracted layout grammar from the supplied figure bundles using a source-checked reference-intake route

04Result

Separated ten PCB methodology figures into independent cards

A compact problem-to-result map generated from the project data used on this page.

What I worked on

  • Removed non-PCB-inspection VIScan material
  • Reorganized camera/video intake, quality gate, ROI, feature extraction, review, and publication flow
  • Built a reusable single-figure atlas for paper, GitHub, Notion, and homepage reuse

Code and results

  • v9 separated gallery and click-to-enlarge behavior in the PCB public homepage
  • Ten 900×900 paper-style square figures, fig31a through fig31j
  • Full Research Publication Workbench mirrored into GitHub Pages
  • PatchCore, FR-PatchCore, DINOv3, and source-checked figure-redraw record

Next steps

The public hub publishes the paper-style figures and project links. Quantitative claims remain limited to locally verified evidence; production-yield and manufacturing-defect claims are held until gold-label validation is approved.

This page connects the PCB inspection work to a paper-style figure atlas, result tables, GitHub, Notion, and the public homepage.

The gallery below uses separate 640x640 figure cards rather than a bundled contact sheet. Click any figure to open the full-size view.

The same Research Publication Workbench that was verified locally on 127.0.0.1:8765 is now mirrored into GitHub Pages.

Publication first-page preview