基于多阈值Otsu准则的阈值分割快速计算

Fast Computation of Threshold Based on Multi-threshold Otsu Criterion

针对传统多阈值Otsu方法在寻找最佳阈值过程中穷举计算效率低的问题,该文分析了多阈值Otsu的阈值性质,证明了使用Otsu方法找到的一组最佳阈值与分割出的各类均值之间的数学对应关系。根据多阈值Otsu的阈值性质,该文提出一个新算法用来快速计算所需最佳阈值,建立了一种新的阈值搜索模型。该算法搜寻满足Otsu多阈值与以此阈值分割出的各类均值之间关系的一组最优阈值,从而确定符合Otsu准则的最佳阈值。该算法有效减少了阈值搜索范围,并且在均值、方差等计算上引入了查找表,优化了底层运算。实验结果表明,与传统多阈值Otsu方法相比,该算法的分割速度大幅度提高,相比于其他多阈值Otsu快速算法,不仅在计算速度上有所提升,而且得到的最佳阈值克服了随机性和偶然性的缺点,是严格符合Otsu原则的。

To resolve the problem of low efficiency which traditional multi-threshold Otsu existing in searching of optimal thresholds on the brute-force method, the thresholds properties of multi-threshold Otsu are analyzed, and the mathematical correspondence is proved between a set of optimal thresholds and the means of various categories A new algorithm is proposed to calculate the optimal thresholds and a new model of searching thresholds is also built according to the properties of thresholds of multi-threshold Otsu. The algorithm searches for a set of optimal thresholds that satisfy the correspondence between the thresholds and the means of various categories segmented by them, so the optimal thresholds of Otsu can be determined. The algorithm reduces the search range effectively and optimizes the calculation of means and variances using lookup table. Experimental results show that the segmentation speed of the algorithm is greatly improved compared with the traditional multi-threshold Otsu method, and the algorithm can not only improve the computation speed, but also overcome the shortcomings of randomness and contingency of thresholds compared with other fast multi-threshold Otsu algorithm, and the results are strictly in line with the principle of multi-threshold Otsu.