Applying oriented fourth-order partial-differential equations to fluorescence microscopic image denoising
Hong Xue · Jisuanji gongcheng yu sheji · 2012
Noise in the fluorescence microscopic images can't be avoided because of imperfect optical imaging system,photoelectric conversion,specimen tissue structure and human errors etc.during the course of optical imaging.Therefore,a new denoi-sing method is proposed based on oriented fourth-order partial-differential equations(PDEs) for fluorescence microscopic images,in which two aspects are considered.One is based on variational method,the other is based on controlling diffusion direction.Experimental results show that the proposed method not only makes the denoised images subjectively more natural and clearer,but also achieves better performance in terms of objective criterion such as peak signal to noise ratio(PSNR) and the structural similarity(SSIM) compared with the related second-order PDEs diffusion models.