作者: Constantine Kotropoulos , Michael Pappas , Ioannis Pitas
DOI: 10.1016/B978-012500451-0/50005-9
关键词:
摘要: Publisher Summary This chapter reviews the class of nonlinear mean filters for noise removal and edge detection. It derives their statistical properties performance in presence impulse signal-dependent noise. Following this, it presents preserving filters. Combinations are used The optimality L2 Rayleigh multiplicative speckle Gaussian supports important applications ultrasonic image processing analysis that described chapter. Lp approximators max/min positive negative p, respectively. substitution traditional operators with means has resulted sorting networks suitable analog implementations. outlines a general filter structure encompassing based on order statistics, homomorphic filters, morphological Finally, median filtering.