Spatio-temporal filter with adaptive multiple outliers rejector

Digital images captured in dim light conditions can be very noisy: mixtures of different noise types may damage the image signal, e.g. impulsive and Gaussian noises. A spatio-temporal filter is proposed, which is capable to eliminate multiple adjacent defective pixels and reduce Gaussian noise in the raw data acquired by the image sensor. The proposed solution is adaptive in that it automatically adjusts its sensitivity, depending on the estimated noise.