The present invention discloses a method for realizing two-dimensional bar code preprocessing by using a threshold binarization algorithm. The method comprises a step of obtaining a two-dimensional bar code image which is subjected to gradation processing, a step of using a Gaussian function to carry out convolution processing on the two-dimensional bar code image to obtain a smooth template, a step of subtracting the pixel value of an original two-dimensional bar code image block by the smooth template to obtain a difference value, a step of using a threshold binarization algorithm to carry out iteration on the obtained difference value to obtain a maximum between-class variance value, and a step of carrying out division processing on the target part and the background part in the two-dimensional bar code image block based on the maximum between-class variance value. According to the embodiment of the present invention, the two-dimensional value image division failure caused by image blur or complex image background information can be avoided, the target part and the background part are separated, and thus the rapid positioning of the target part in a later period is facilitated.