Title :
Noise robust two-dimensional blind deconvolution of ultrasound images
Author :
Taxt, Torfinn ; Strand, Jarle
Author_Institution :
Sect. for Med. Image Anal. & Inf., Bergen Univ., Norway
Abstract :
This paper presents a new method for two-dimensional blind homomorphic deconvolution of medical B-scan ultrasound images. The method is based on noise robust two-dimensional phase unwrapping and a noise robust procedure to obtain the pulse in the complex cepstrum domain. Ordinary Wiener filtering is used in the subsequent deconvolution. The resulting images became much sharper, with better defined tissue structures compared to the ordinary images. The deconvolved images had a resolution gain in the order of 3 to 6, and the signal to noise ratio doubled for many of these images. The method gave stable results through several image sequences. It can be implemented in real time
Keywords :
Wiener filters; acoustic noise; biomedical ultrasonics; deconvolution; image resolution; image sequences; medical image processing; better defined tissue structures; complex cepstrum domain; medical B-scan ultrasound images; medical diagnostic imaging; noise robust procedure; noise robust two-dimensional blind deconvolution; pulse; real time implementation; resolution gain; signal to noise ratio; two-dimensional phase unwrapping; Biomedical imaging; Cepstrum; Deconvolution; Image resolution; Noise robustness; Phase noise; Signal resolution; Signal to noise ratio; Ultrasonic imaging; Wiener filter;
Conference_Titel :
Ultrasonics Symposium, 1997. Proceedings., 1997 IEEE
Conference_Location :
Toronto, Ont.
Print_ISBN :
0-7803-4153-8
DOI :
10.1109/ULTSYM.1997.661853