中文摘要：

Otsu自适应阈值算法是图像分割的经典方法之一，在其基础上发展起来的二维Otsu阂值算法却因为计算复杂而制约了其应用。本文针对二维Otsu的耗时瓶颈问题，引入积分图像简化了二维直方图最佳阈值搜索过程的计算复杂度，从而大大缩短了算法处理时问。实验结果表明，本文提出的算法在分割效果上与原始算法完全相同，但计算时间远远小于后者，具有很好的实时性。

英文摘要：

Otsu adaptive threshold algorithm is one of the classic image segmentation methods. The application of the two-dimensional Otsu threshold algorithm based on original Otsu algorithm is restricted due to its computation complexity. To solve the time-consuming problem of the two-dimensional Otsu algorithm, in this paper, we improve the original algorithm with integral image and put forward a new fast algorithm. The segmentation result of our fast algorithm is the same as that of the original one, but the computation time is significantly reduced, which demonstrates well good real-time performance.