An Improved Two-Dimensional OTSU Segmentation Method for Nvshu Character Image

For the structural characteristics of Chinese NvShu character, this paper proposes an improved two-dimensional (2-D) OTSU segmentation method based on the lateral inhibition network for Nvshu character image. A 2-D histogram with the gray level and the lateral inhibition level of the image was established and the maximum between-cluster variance was chosen as the criterion to select the optimal threshold. Experimental results show that the proposed method not only successfully reduced the effect of background noise, but also improved the accuracy of the character image segmentation, especially for NvShu character images with low contrast, uneven gray level of character strokes and uneven background.