Investigating speckle pattern produced by biological tissue via image quality assessment

Abstract

Speckle pattern produced by biological tissue is a grainy and stochastic texture image that appeared to cover such a diffuse object when illuminated by laser radiation. While the speckle pattern has been treated as unwanted noise that distorts the measurement results in any imaging modality involved laser illumination, it holds important information concerning the observed diffuse object that can be extracted by texture analysis. In this study, texture characterization of the speckle pattern as an information carrier about the surface roughness of biological tissue was investigated through a completely blind image quality assessment. The proposed work was applied to ten bovine articular cartilage samples degenerated into different average roughness values, ranging from 0.09 μm to 2.51 μm. The obtained results provided a good relationship between the quantitative image quality assessment metric and the average surface roughness.