SGH imaging surface inspection is an advanced optical detection technology that combines structured gradient illumination with high-resolution computational imaging to identify microscopic surface defects with exceptional accuracy. As a non-contact, non-destructive testing method, SGH imaging has become an essential quality control solution across semiconductor, automotive, precision machining, and electronics manufacturing industries where surface integrity directly impacts product performance and reliability.
Technical Overview
SGH imaging surface inspection is a state-of-the-art optical metrology technique engineered to detect, localize, and classify surface anomalies that are invisible or poorly resolved under conventional imaging methods. The core principle of the technology lies in projecting structured gradient illumination patterns onto the target surface and capturing the reflected or scattered light field with a high-sensitivity imaging sensor. Because different defect types—such as scratches, dents, protrusions, and contamination—interact with the structured light field in characteristic ways, the system can extract defect signatures through computational reconstruction of the acquired image sequence.
Unlike traditional visual inspection or single-angle bright-field imaging, SGH imaging acquires multiple images under varying illumination geometries within a fraction of a second. These image stacks are processed using gradient contrast algorithms and, in advanced configurations, phase-sensitive reconstruction techniques that enhance the visibility of low-contrast defects on reflective, curved, or textured surfaces. The result is a comprehensive surface quality map with defect localization precision in the micrometer range.
Key technical characteristics of SGH imaging surface inspection include:
- Non-contact, non-destructive measurement that preserves sample integrity
- High defect detectability on highly reflective, matte, curved, and textured surfaces
- Micrometer-level spatial resolution and sub-micrometer height sensitivity
- High-speed acquisition suitable for inline production inspection
- Automated defect classification through machine learning algorithms
- Full-field measurement capability covering large surface areas in a single pass
Testing Samples
Metallic Components
Machined metal parts, stamped panels, polished shafts, bearing surfaces, die-cast housings, and anodized aluminum components represent one of the largest application categories. SGH imaging effectively handles both mirror-finished and brushed metal surfaces, where specular reflection typically challenges conventional camera-based inspection.
Semiconductor and Electronic Substrates
Silicon wafers, photomasks, printed circuit boards, ceramic substrates, lead frames, and semiconductor packages demand extremely stringent surface quality. SGH imaging detects sub-micrometer particles, polishing marks, and pattern anomalies critical to yield management in fabrication lines.
Glass and Optical Components
Cover glass, display panels, lenses, prisms, and optical filters benefit from SGH imaging's sensitivity to micro-scratches, digs, inclusions, and coating defects that transmit or scatter light in detectable patterns.
Polymer and Coated Surfaces
Injection-molded plastic parts, painted automotive panels, film materials, and functional coatings can be inspected for flow marks, orange peel texture, coating voids, delamination indicators, and contamination.
Precision Machined and Additive Manufactured Parts
Ground and lapped components, sealing surfaces, medical implants, and 3D-printed metal parts are inspected for tool marks, burrs, porosity, and surface irregularities that affect fit, function, and fatigue life.
Testing Items
SGH imaging surface inspection covers an extensive catalog of defect categories and geometric surface characteristics, including:
- Scratches, scuffs, and polishing marks of varying depth and orientation
- Dents, dings, and indentation defects on formed or finished surfaces
- Protrusions, bumps, and material buildup such as weld spatter or residue
- Particulate contamination, dust, fibers, and foreign object debris
- Cracks, micro-fractures, and stress-induced surface fissures
- Corrosion pits, oxidation spots, and staining
- Coating defects including pinholes, voids, streaks, and uneven coverage
- Burrs and edge defects along machined boundaries
- Surface roughness deviation and texture non-uniformity
- Contamination such as fingerprints, oil films, and chemical residue
- Molding defects including flow lines, sink marks, and gate marks
- Dimensional and form-related surface anomalies on critical functional faces
Testing Methods
Structured Gradient Illumination Acquisition
The fundamental acquisition method projects a sequence of structured illumination patterns—typically gradient or fringe patterns—from multiple illumination angles onto the sample surface. A high-dynamic-range camera captures the reflected light field for each pattern. Defects that disturb the local surface gradient produce characteristic distortions in the reflected pattern, which are amplified through differential image processing.
Multi-Angle Photometric Reconstruction
By illuminating the sample from at least four to eight azimuthal directions, the system computes local surface gradient vectors for every pixel. This photometric reconstruction generates three-dimensional surface micro-topography, enabling quantitative distinction between raised defects, recessed defects, and flat contamination.
Dark-Field and Bright-Field Fusion
SGH systems combine dark-field imaging, which is highly sensitive to scattering defects such as scratches and particles, with bright-field imaging, which captures overall surface appearance and reflective uniformity. Fusing both datasets allows simultaneous detection of geometric defects and cosmetic imperfections.
Phase-Sensitive Defect Enhancement
For applications demanding ultimate sensitivity, phase-shifting techniques are applied to the structured patterns. By analyzing phase deviations between the projected and reflected patterns, the method achieves sub-micrometer height resolution, making it suitable for detecting shallow scratches on polished wafers and optical surfaces.
Computational Defect Classification
Acquired image data is processed through calibrated detection algorithms that isolate defect candidates from background texture. Machine learning classifiers trained on labeled defect libraries then assign each candidate to a defect class, estimate severity, and output pass or fail judgments according to customer-defined acceptance criteria. Results are delivered as defect maps, dimensional measurements, classification reports, and statistical quality metrics.
Applications
SGH imaging surface inspection serves diverse industries where surface quality governs product performance, aesthetics, and reliability:
- Automotive: inspection of body panels, trim components, machined engine parts, and decorative finishes prior to assembly
- Semiconductor and microelectronics: wafer surface monitoring, package inspection, and substrate quality control
- Consumer electronics: smartphone housings, display cover glass, and cosmetic surface verification
- Aerospace: turbine blade inspection, structural component verification, and fastener surface evaluation
- Medical devices: implant surface assessment, surgical instrument inspection, and catheter and tubing quality control
- Precision optics: lens, mirror, and filter defect screening in production environments
- Metal processing: rolled strip, coil, and plate surface monitoring in continuous production lines
- Injection molding and plastics: molded part appearance inspection and tooling wear diagnostics
Because SGH imaging systems can be configured as offline laboratory stations, semi-automated inspection cells, or fully integrated inline inspection modules with conveyor or robotic part handling, they adapt readily to both development-stage analysis and high-volume manufacturing quality assurance.
FAQ
What does SGH stand for in surface inspection?
SGH refers to the structured gradient holographic imaging architecture used in the inspection system, which combines structured gradient illumination with computational reconstruction of the reflected light field. This configuration delivers enhanced defect contrast compared with conventional camera inspection.
How small a defect can SGH imaging detect?
Detection capability depends on surface finish and illumination configuration, but typical systems resolve lateral defect dimensions down to a few micrometers and height variations below one micrometer on polished surfaces. Dark-field channels can detect sub-micrometer particles on semiconductor-grade samples.
Does the inspection damage or contact the sample?
No. SGH imaging is entirely non-contact and non-destructive. Illumination intensity is controlled within safe limits, making the method suitable for finished products, sterile medical components, and delicate optical parts.
Can SGH imaging inspect curved or complex-shaped surfaces?
Yes. Multi-angle illumination and computational reconstruction accommodate curved surfaces, fillets, and moderately complex geometries. Highly concave features may require specialized illumination configurations or multiple inspection stations.
How fast is an SGH inspection cycle?
Typical acquisition takes several hundred milliseconds to a few seconds depending on the number of illumination patterns and the required sensitivity. Inline systems achieve cycle times compatible with high-volume production lines, often exceeding several parts per minute with automated handling.
Is the inspection suitable for reflective metal surfaces?
Yes, and this is one of its strengths. The gradient-based illumination approach suppresses the specular saturation problems that plague conventional imaging on mirror-like metals, allowing reliable detection of faint scratches and dents on polished surfaces.
How are results reported?
Deliverables typically include annotated defect maps, defect classification and size statistics, three-dimensional micro-topography data where applicable, and pass or fail determinations against specified acceptance criteria, all exported in standard report formats for quality documentation and traceability.