Radio astronomers are interested in measuring the intensity of the observed signal, and on that account, an accurate model of the primary beam is very crucial. However, holographic measurement of the primary beam is mostly corrupted by noise and this can impede the detection of weak emission from distant sources. The interference may be caused by antenna electronics, astrophysical sources, our cell phones or any other artificial electronics. This paper presents the Singular Value Decomposition (SVD) method to mitigate unwanted signal in perturbed primary beams. The results indicated that using 3 strongest coefficients of singular values, we can denoise and reproduce the image close to the original beam due to artefacts in the raw beam maps being reduced by a factor of at least 0.09%.