Title :
Noise suppression of fluorescein angiogram sequences using bilateral filter
Author :
Guo, Xiao-Xin ; Lu, Yi-Nan ; Xu, Zhi-Wen ; Liu, Zhan-Hui ; Wang, Yun-Hiao ; Pang, Yun-Jie
Author_Institution :
Coll. of Comput. Sci. & Technol., Jilin Univ., Changchun, China
Abstract :
In this paper, a relatively flexible filter called extended bilateral filter is proposed, by which some particular filters can be designed via selecting an appropriate pixel of interest (POI) and defining a kernel for the filter according to some specified requirements. Based on the filter, a fluorescein-angiogram-oriented filter, high-intensity preserving (HIP) filter is introduced to solve the problems of high-intensity preservation when denoising. The results of the experiments show that the HIP filter has the ability of noise suppression, and the ability of high-intensity preservation, which can keep the intensities of bright pixels from degrading via distributing the weights adaptively according to the location and the intensity of each pixel.
Keywords :
filtering theory; image denoising; image sequences; medical image processing; bilateral filter; fluorescein angiogram sequence; fluorescein-angiogram-oriented filter; high-intensity preserving filter; image denoising; image sequence; noise filtering; noise suppression; pixel intensity; pixel of interest; Angiography; Blood flow; Degradation; Diseases; Filters; Fluorescence; Hip; Image sequences; Noise reduction; Retina; Bilateral filter; fluorescein angiogram; image sequence, noise filtering; pixel of interest (POI);
Conference_Titel :
Machine Learning and Cybernetics, 2005. Proceedings of 2005 International Conference on
Conference_Location :
Guangzhou, China
Print_ISBN :
0-7803-9091-1
DOI :
10.1109/ICMLC.2005.1527892