Multiphase flow systems, crucial in fields like chemical, petroleum, and biology, often involve challenging gas-liquid flows with variable interfaces. Ultrasonic Tomography (UT) helps visualize these flows by using array sensors to create two-dimensional images.
However, issues arise in assessing the cross-section gas-holding ratio (CGR) with UT, as original images usually need filter processing and multi-dimensional evaluations can be contradictory. This study introduces a comprehensive evaluation method for UT image quality, leveraging the entropy evaluation method. It uses ultrasonic data and algorithms like
SLBP, Landweber, and Tikhonov for image reconstruction. Evaluation focuses on three aspects: Pearson correlation coefficient, image position error, and reconstruction area error, offering a thorough assessment of image accuracy. The entropy evaluation method balances contradictory indicators and aligns with most indicator distribution laws, proving effective in assessing UT image reconstruction quality.