Background speckle can often obscure objects of interest in an ultrasound image. The probability of detection and classification of point scatterers is highly affected by background speckle. The proposed algorithm uses a coherence-based wavelet coefficient shrinkage method. Point scatterers in the ultrasound image are enhanced by separating coherent point targets from incoherent background speckle. Results using Field II ultrasound simulations show how the algorithm retains the point scatterers and increases their conspicuity. The algorithm has potential to detect microcalcifications in breast tissue.