This paper addresses methods for removing multiplicative noise from digital images. Multiplicative noise is a complex grainy noise typically found in images obtained through specific imaging technologies such as ultrasound, radar, and laser scanning. This type of noise significantly complicates visual interpretation and processing of the image. The aim of the thesis is to investigate and compare the effectiveness of three selected methods: the spatial averaging method, the wavelet method, and the non local means method in reducing multiplicative noise while preserving important details. A description of the mathematical models and theoretical foundations of each method is also presented. All methods were implemented in the Python programming language, and their effectiveness was evaluated based on visual aspects and using two metrics: peak signal-to-noise ratio (PSNR) and structural similarity index (SSIM). The results showed that each method has its advantages and disadvantages. The spatial averaging method is simple to implement, but its drawback is that it significantly blurs the image. The wavelet method allowed for better preservation of details based on visual analysis, but provided the worst results in terms of metric improvements across all images. This method is also the most complex to implement. The non-local means method is the most computationally demanding, but consistently achieved the best results in preserving image structure. Based on the analysis conducted in this thesis, it can be concluded that the choice of the optimal noise removal method depends on the characteristics of the image, the noise, and individual needs. The thesis provides a detailed analysis and comparison of the methods, contributing to the understanding of the problem and techniques for removing multiplicative noise from digital images.