We have proposed a novel modified multichannel singular spectrum analysis (MSSA) algorithm to attenuate random noise with a new formula of low-rank reduction, which is named as the damped MSSA algorithm. Compared with the traditional truncated singular value decomposition (TSVD) formula, we introduced a damping factor to damp the singular values that correspond to the signal in order to attenuate the residual noise appearing in the traditional approach. The preservation of useful signals and the removal of random noise are compromised through the introduced damping factor. While the rank in MSSA has a big range considering the data size and data complexity, the damping factor in the damped MSSA is usually chosen as an integer that is slightly larger than 1 (such as 2, 3, or 4) to obtain a sufficient improvement. From the synthetic and field data examples, it is obvious that the proposed damped MSSA algorithm can obtain cleaner denoised image compared with the traditional MSSA algorithm.