A 2D barcode region detection method based on embedded system for augmented reality applications was developed. For the successful 2D barcode candidate region detection, the range of variance and frequency of gray level distribution in the test 2D barcodes was modeled and it was combined with the corner features to localize the final 2D barcode candidates. An automatic 2D barcode localization software was developed with the multiple features mixture method and we tested our system on real camera images of several popular 2D barcode symbologies.