A Generalization of Otsu's Method and Minimum Error Thresholding
This work addresses image thresholding for tasks like document binarization, offering a simple and fast method that is incremental in unifying existing techniques.
The paper tackles the problem of histogram-based image thresholding by introducing Generalized Histogram Thresholding (GHT), which unifies and interpolates between classic methods like Otsu's method and Minimum Error Thresholding, leading to improved accuracy, as demonstrated by outperforming or matching other algorithms, including deep neural networks, on a handwritten document image binarization challenge.
We present Generalized Histogram Thresholding (GHT), a simple, fast, and effective technique for histogram-based image thresholding. GHT works by performing approximate maximum a posteriori estimation of a mixture of Gaussians with appropriate priors. We demonstrate that GHT subsumes three classic thresholding techniques as special cases: Otsu's method, Minimum Error Thresholding (MET), and weighted percentile thresholding. GHT thereby enables the continuous interpolation between those three algorithms, which allows thresholding accuracy to be improved significantly. GHT also provides a clarifying interpretation of the common practice of coarsening a histogram's bin width during thresholding. We show that GHT outperforms or matches the performance of all algorithms on a recent challenge for handwritten document image binarization (including deep neural networks trained to produce per-pixel binarizations), and can be implemented in a dozen lines of code or as a trivial modification to Otsu's method or MET.