Effectiveness of specularity removal from hyperspectral images in the colour spectral measurement of wool fibres

Microscopic hyperspectral imaging technology is a potential non-destructive and non-contact method for colour measurement of micrometre-sized textile fibres. However, specularity on the fibre surface can distort accurate information affect accuracy measurement. This paper proposed specular-constrained sparse approximation (SCSA) specular-diffuse reflection separation from images wool First, specular prior map generated based lightness dissimilarity. Then SCSA model used to decompose processed image A into low-rank data L, S constrained by map, noise E, Gaussian N. non-linear logistic sigmoid function – L N are improve performance removal during iterative optimization. The experimental results show that significantly preserves diffuse reflectance texture details in highlight regions obtain actual spectral chromatic values