Translator Disclaimer
9 March 2015 A novel method for image denoising of fluorescence molecular imaging based on fuzzy C-Means clustering
Author Affiliations +
As an important molecular imaging modality, fluorescence molecular imaging (FMI) has the advantages of high sensitivity, low cost and ease of use. By labeling the regions of interest with fluorophore, FMI can noninvasively obtain the distribution of fluorophore in-vivo. However, due to the fact that the spectrum of fluorescence is in the section of the visible light range, there are mass of autofluorescence on the surface of the bio-tissues, which is a major disturbing factor in FMI. Meanwhile, the high-level of dark current for charge-coupled device (CCD) camera and other influencing factor can also produce a lot of background noise. In this paper, a novel method for image denoising of FMI based on fuzzy C-Means clustering (FCM) is proposed, because the fluorescent signal is the major component of the fluorescence images, and the intensity of autofluorescence and other background signals is relatively lower than the fluorescence signal. First, the fluorescence image is smoothed by sliding-neighborhood operations to initially eliminate the noise. Then, the wavelet transform (WLT) is performed on the fluorescence images to obtain the major component of the fluorescent signals. After that, the FCM method is adopt to separate the major component and background of the fluorescence images. Finally, the proposed method was validated using the original data obtained by in vivo implanted fluorophore experiment, and the results show that our proposed method can effectively obtain the fluorescence signal while eliminate the background noise, which could increase the quality of fluorescence images.
© (2015) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Yu An, Jie Liu, Jinzuo Ye, Yamin Mao, Xin Yang, Shixin Jiang, Chongwei Chi, and Jie Tian "A novel method for image denoising of fluorescence molecular imaging based on fuzzy C-Means clustering", Proc. SPIE 9330, Three-Dimensional and Multidimensional Microscopy: Image Acquisition and Processing XXII, 93301E (9 March 2015);


Validated CCD camera model
Proceedings of SPIE (September 01 1991)
Research on effect of Kell coefficient on the quality of...
Proceedings of SPIE (October 22 2010)
Research of full-field deformation measurement system
Proceedings of SPIE (November 20 2009)

Back to Top