2D barcode image binarization based on wavelet analysis and Otsu's method

It is extremely challenging to binarize 2D barcode image under the condition of both nonuniform illumination and cluttered background. Neither global nor local thresholding method can yield an ideal binary result. In this paper, we propose a novel 2D barcode image binarization method. The contribution of the method is that it can remove the unnormalized illumination especially in the cluttered background. This is achieved by estimating the illumination distribution by the wavelet analysis and followed by removing it from the source image according to the image formation model. The proposed method can binarize the challenging barcode images with only simple Otsu's algorithm. Experimental results reflect the robustness of this method. In addition, since the filtering implementation of wavelet transform well suits the DSP architecture, the method performs well in our DSP-based 2D barcode recognition terminal for real application.