Speckle pattern optimization in laser speckle photometry for non-destructive testing

Dynamics of a speckle pattern formed on the surface of a diffusely reflecting object is a source of valuable information about the changes that occur in the object. Laser speckle photometry extracts relevant information about the object from variations in speckle intensity. Speckle dynamics is visualized by means of an activity map, which renders the 2D distribution of a statistical parameter related to intensity changes. The contrast of the map is crucial for better detection of areas of different activity. Strong fluctuations in the map input data severely degrade the contrast. The main factors affecting the contrast are the processing algorithm and the speckle intensity distribution defined by the parameters of the optical system. The aim of this work is to find the optimal speckle pattern for laser speckle photometry based on the quality assessment of a 2D activity map estimated by correlation-based algorithms with averaging in the temporal or spatial domain. The study included simulation of correlated images with symmetric/asymmetric intensity distribution or at different speckle contrasts. We checked the quality of the obtained map for the case of 8-bit encoded, binary, or JPEG-compressed speckle images. We performed sensitivity evaluation of the method for non-destructive testing of samples under tensile extension.