AI enhanced strong-field terahertz spectral detection and imaging

Area of Science:

• Physics
• Materials Science
• Engineering

Background:

• Terahertz (THz) waves offer unique properties for nondestructive testing (NDT) like broad bandwidth and penetration.
• Current THz NDT methods face limitations including unknown material properties, signal aliasing, and slow detection.

Purpose of the Study:

• To develop an advanced THz NDT system for enhanced material characterization.
• To overcome existing challenges in THz-based nondestructive testing and imaging.

Main Methods:

• Utilized an enhanced 4-inch spintronic strong-field THz emitter.
• Integrated neural network-assisted algorithms for thickness prediction and contour detection.
• Applied the system to ultrathin materials and large-scale samples for defect identification.

Main Results:

• Achieved micrometer-scale accuracy (±8 μm) for ultrathin material thickness measurement.
• Demonstrated rapid defect identification and submillimeter-level depth resolution.
• Validated the system's scalability and flexibility for complex material analysis.

Conclusions:

• The developed THz NDT system offers high accuracy and efficiency for material inspection.
• This technology shows significant potential for applications in surface coating, component maintenance, and cultural heritage.
• Paves the way for advanced thickness measurement and internal defect detection in diverse materials.