17 From theory to tools: selection, cross-platform mapping and inspection-flow design
Chapter status: Structural placeholder.
This chapter currently contains the approved heading structure but no reviewed technical content.
17.1 The industrial vision software landscape
17.1.1 Configurable tool sets versus code libraries
Content status: Placeholder.
This subsection has not yet been drafted or technically reviewed.
17.1.2 Open-source libraries: OpenCV, scikit-image, PyTorch/TF ecosystems
Content status: Placeholder.
This subsection has not yet been drafted or technically reviewed.
17.1.3 Commercial libraries and SDKs: Halcon, VisionPro, Sapera
Content status: Placeholder.
This subsection has not yet been drafted or technically reviewed.
17.1.4 Smart-sensor and no-code environments: Keyence, Zebra Aurora, In-Sight
Content status: Placeholder.
This subsection has not yet been drafted or technically reviewed.
17.1.5 Configuring versus programming: trade-offs and consequences
Content status: Placeholder.
This subsection has not yet been drafted or technically reviewed.
17.2 Cross-platform tool mapping
17.2.1 How to read the mapping tables: theory concept to platform tool
Content status: Placeholder.
This subsection has not yet been drafted or technically reviewed.
17.2.2 Acquisition and preprocessing across platforms
Content status: Placeholder.
This subsection has not yet been drafted or technically reviewed.
17.2.3 Edge, line and circle detection: OpenCV / VisionPro / Halcon / In-Sight / Keyence / Zebra Aurora
Content status: Placeholder.
This subsection has not yet been drafted or technically reviewed.
17.2.4 Pattern location and alignment across platforms
Content status: Placeholder.
This subsection has not yet been drafted or technically reviewed.
17.2.5 Measurement and gauging tools across platforms
Content status: Placeholder.
This subsection has not yet been drafted or technically reviewed.
17.2.6 Defect and surface-inspection tools across platforms
Content status: Placeholder.
This subsection has not yet been drafted or technically reviewed.
17.2.7 Identification: barcode, Data Matrix, OCR/OCV across platforms
Content status: Placeholder.
This subsection has not yet been drafted or technically reviewed.
17.2.8 Deep-learning tools across platforms
Content status: Placeholder.
This subsection has not yet been drafted or technically reviewed.
17.2.9 Limits of equivalence: where tools do not map cleanly
Content status: Placeholder.
This subsection has not yet been drafted or technically reviewed.
17.3 Selecting and prioritising tools
17.3.1 From feature and task to candidate tools
Content status: Placeholder.
This subsection has not yet been drafted or technically reviewed.
17.3.2 Prioritisation: robustness, speed, determinism, maintainability
Content status: Placeholder.
This subsection has not yet been drafted or technically reviewed.
17.3.3 Rule-based versus learned tools for a given problem
Content status: Placeholder.
This subsection has not yet been drafted or technically reviewed.
17.3.4 A tool-selection decision guide, cross-referenced to theory chapters
Content status: Placeholder.
This subsection has not yet been drafted or technically reviewed.
17.4 Designing an inspection as a sequence of steps
17.4.1 Decomposing an objective into ordered steps
Content status: Placeholder.
This subsection has not yet been drafted or technically reviewed.
17.4.2 The step pattern: objective, tools, region, output
Content status: Placeholder.
This subsection has not yet been drafted or technically reviewed.
17.4.3 Locate–mask–measure–decide as a canonical flow
Content status: Placeholder.
This subsection has not yet been drafted or technically reviewed.
17.4.4 Worked example: scratch inspection decomposed into steps
Content status: Placeholder.
This subsection has not yet been drafted or technically reviewed.
17.5 Thresholds, checks and pass/fail logic
17.5.1 Per-step checks: presence, contrast, dimension, position limits
Content status: Placeholder.
This subsection has not yet been drafted or technically reviewed.
17.5.2 Setting limits and thresholds on tool outputs
Content status: Placeholder.
This subsection has not yet been drafted or technically reviewed.
17.5.3 Building the overall pass/fail decision from step results
Content status: Placeholder.
This subsection has not yet been drafted or technically reviewed.
17.5.4 Failure attribution: knowing which step failed and why
Content status: Placeholder.
This subsection has not yet been drafted or technically reviewed.
17.6 Exposing KPIs and data-driven threshold setting
17.6.1 Choosing KPIs and measured outputs per tool and step
Content status: Placeholder.
This subsection has not yet been drafted or technically reviewed.
17.6.2 Logging tool outputs for offline analysis
Content status: Placeholder.
This subsection has not yet been drafted or technically reviewed.
17.6.3 Batch statistical analysis to set thresholds and margins
Content status: Placeholder.
This subsection has not yet been drafted or technically reviewed.
17.6.4 Separating good/bad populations and setting operating points
Content status: Placeholder.
This subsection has not yet been drafted or technically reviewed.
17.6.5 Linking threshold setting to the validation of Chapter 18
Content status: Placeholder.
This subsection has not yet been drafted or technically reviewed.
17.7 Software design practice for vision applications
17.7.1 Modularity, reuse and configuration management
Content status: Placeholder.
This subsection has not yet been drafted or technically reviewed.
17.7.2 Robustness: handling absent, occluded and out-of-spec parts
Content status: Placeholder.
This subsection has not yet been drafted or technically reviewed.
17.7.3 Defensive checks, error states and safe defaults
Content status: Placeholder.
This subsection has not yet been drafted or technically reviewed.
17.7.4 Version control, documentation and change control of vision jobs
Content status: Placeholder.
This subsection has not yet been drafted or technically reviewed.
17.7.5 Performance, cycle time and resource budgeting
Content status: Placeholder.
This subsection has not yet been drafted or technically reviewed.