Objective Image denoising is a basic pretreatment of medical image processing, which has a significant impact on the quality of subsequent analysis and processing of the image. Certain analysis and simulation of the image denoising algorithms are made in this paper. Methods Firstly, we describe the basic principles of several common image denoising algorithms. Secondly, we use different algorithms to do denoising simulation with the Lena image added with Gaussian noise, and then compare the different results of peak signal-to-noise ratio (PSNR) and mean square error (MSE). Finally, we summarize and choose the optimal image denoising algorithm in the application of functional magnetic resonance imaging (fMRI) data analysis to obtain better post-processing foundation. Results The results of PSNR and MSE by using wavelet hierarchical threshold algorithm in fMRI are better. Conclusions In fMRI denoising, wavelet hierarchical threshold algorithm can improve the utilization of image information and increase the physician’s diagnostic accuracy.