Performance analysis of a maximum-likelihood speckle motion estimator.

Area of Science:

• Non-destructive evaluation
• Optical metrology
• Image processing

Background:

• Speckle interferometry and related techniques are vital for non-destructive evaluation.
• Accurate estimation of speckle motion is essential for quantitative analysis in these methods.
• Existing estimators face challenges with decorrelation and motion range.

Purpose of the Study:

• To analyze the performance of a maximum likelihood estimator (MLE) for small speckle motion.
• To characterize the estimator's bias and RMS deviation concerning signal-to-noise ratio (SNR).
• To evaluate the estimator's applicability for biological tissue imaging and other decorrelating scenarios.

Main Methods:

• Performance analysis of a specific maximum likelihood estimator.
• Characterization of estimator bias and RMS deviation.
• Evaluation as a function of signal-to-noise ratio (SNR).

Main Results:

• The maximum likelihood estimator demonstrates reliable performance across a range of signal-to-noise ratios.
• Accurate estimations were achieved for instantaneous motions up to +/-0.8 pixels/record.
• The estimator's effectiveness is linked to the decorrelation of sequential speckle patterns.

Conclusions:

• The analyzed maximum likelihood estimator offers robust performance for calculating small speckle motions.
• Its accuracy is well-defined across varying SNR levels, making it suitable for challenging applications.
• The estimator provides a valuable tool for non-destructive evaluation and biological imaging.