Improving signal-to-noise and, thereby, image contrast is one of the key challenges needed to expand the useful applications of mass spectrometry imaging (MSI). Both instrumental and data analysis approaches are of importance. Univariate denoising techniques have been used to improve contrast in MSI images with varying levels of success. Additionally, various multivariate analysis (MVA) methods have proven to be effective for improving image contrast. However, the distribution of important but low intensity ions can be obscured in the MVA analysis, leading to a loss of chemically specific information. In this work we propose inverse maximum signal factors (MSF) denoising as an alternative approach to both denoising and multivariate analysis for MSI imaging. This approach differs from the standard MVA techniques in that the output is denoised images for each original mass peak rather than the frequently difficult to interpret scores and loadings. Five tests have been developed to optimize and validate the resulting denoised images. The algorithm has been tested on a range of simulated data with different levels of noise, correlated noise, varying numbers of underlying components, and nonlinear effects. In the simulations, an excellent correlation between the true images and the denoised images was observed for peaks with an original signal-to-noise ratio as low as 0.1, as long as there was sufficient intensity in the sum of the selected peaks. The power of the approach was then demonstrated on two time-of-flight secondary ion mass spectrometry (ToF-SIMS) images that contained largely uncorrelated noise and a laser post-ionization matrix-assisted laser desorption/ionization mass spectrometry (MALDI-2-MS) image that contained strongly correlated noise. The improvements in signal-to-noise increased with decreasing intensity of the original peaks. A signal-to-noise improvement of as much as two orders of magnitude was achieved for very low intensity peaks. MSF denoising is a powerful addition to the suite of image processing techniques available for studying mass spectrometry images.