Analysis and Comparison of Some Frequency Compounding Algorithms for the Reduction of Ultrasonic Clutter

When ultrasonically imaging sharp-edged targets en- closed by a strongly scattering medium, such as flaws in grainy metal or organ outlines in tissue, the target signal is frequently obscured by time-invariant echoes generated by the microstructure of the environment. The target visibility with respect to the clutter can be improved by compounding statistically independent echoes of each re- gion of the medium. Algorithms for accomplishing this are analyzed and compared, in which the independent signals are obtained by split- ting each ,4-line echo into nonoverlapping frequency bands. One of these algorithms, in which compounding is performed by choosing the min- imum squared signal at each range, is found to provide considerable improvement in signal to clutter ratio and mixed results in receiver operating characteristic at the expense of resolution.