Fractal speckle image analysis for surface characterization of aerospace structures

• Main Authors: Haridas, Aswin, Crivoi, Alexandru, Prabhathan, P., Chan, Kelvin, Murukeshan, Vadakke Matham
• Other Authors: Asundi, Anand K.
• Format: Article
• Language: English
• Published: 2019
• Subjects: Engineering::Mechanical engineering; Fractal Dimension; Image Processing
• Online Access: https://hdl.handle.net/10356/106840; http://hdl.handle.net/10220/49667
• Institution: Nanyang Technological University

Surface characterization of the working components has always been a subject of interest among researchers and industry specialists. Especially in the aerospace industry where the aerodynamic capabilities are largely altered by the surface quality of the component of interest, there remains an extensive need for developing systems for effectively characterizing the surface quality. To realize an optical based non-contact and an in-line surface roughness measurement system, it is essential to understand the relationship between the quality of the surface and statistical parameter of the reflected speckles. The range of the measurement system being proportional to the wavelength of light used makes the analysis fundamentally important in order to understand the properties of speckles at a different wavelength. In this context, this paper examines the nature of the formed IR speckles from three different diffusers by analyzing their raw structure. Image processing algorithms that are developed study the different parameters of the 8-bit binary speckles, namely, the fractal property and number of connecting components. The paper also discusses the future work direction on relating the proposed analysis to derive the algorithm required for evaluating the surface finish parameters.