Split Bregman method, which can converge quickly and denoise very well, is considered to be an extremely efficient method for total variation denosing model. By studying Split Bregman method, an adaptive Split Bregman method which can adjust Lagerange multiplier and penalty parameter automatically was proposed. Experimental results show that the new method can not only converge faster than the traditional Split Bregman method, but also preserve the edge information while removing noises.