Multi-Angle Annular Illumination Microscopy and Image Processing Strategy for Wafer Defect Inspection

The semiconductor industry is rapidly developing, and fabrication facilities demand higher standards for the surface quality inspection of patterned wafers. In this paper, a multi-angle annular illumination microscopy and an image processing strategy specifically designed for wafer defect inspection are proposed. We developed a microscopy system utilizing a variety of illumination modes, including bright field, high-angle, and low-angle dark field, based on an annular illumination diffraction model and a wafer scattering model. Additionally, we proposed an optimized fusion algorithm based on HSI (Hue, Saturation, Intensity) color space and wavelet transform for combining bright and dark field images. We conducted comparative experiments using different incident angles and various fusion algorithms. The results demonstrate that the multi-angle annular illumination significantly enhances defect contrast and reduces background noise. Overall, our image fusion algorithm and subsequent processing strategy significantly improve the efficiency and accuracy of defect inspection on wafer surfaces.