A real-time recognition method for two dimension (2D) bar codes based on image process is proposed in this paper, which is capable of identifying 2D bar code images in complex imaging conditions rapidly and accurately. Unlike methods that only recognize 2D bar code images in ideal conditions, this method does well in images that have high noise, such as highlight spots, non-heterogeneous lighting and low contradistinction, and it can deal with the distorted images that have skew angles. This method includes three steps. The first step detects the code region using the Otsu algorithm and the Least Square Method (LSM). The second method searches for the cut-off rules with a scanning approach. In the third step symbol characters are segmented from the original image. Experimental results show the accuracy and performance of this method are acceptable, especially the capability of dealing with noised images.