中文说明：
传统的图像分割方法可以分为基于阈值、边缘、区域的方法等，而传统的分割方法在复杂图像分割中效果不好。为了提高图像分割的性能，我们提出了一种基于局部自适应窗口的Otsu方法。首先，分析了传统阈值法在复杂图像分割中不能很好地形成的原因，比较了全局阈值和局部阈值对图像分割的影响。其次，提出了根据局部信息自适应地改变局部窗口大小的方法，分析了该方法的特点。最后，证明了所提出的新方法优于其他方法。实验结果表明，与其他传统方法相比，该方法能保持更多的细节，获得更满意的结果。
English Description：
Traditional image segmentation methods can be divided into the method based on threshold, edge, region and so on, and the traditional segmentation method in complex image segmentation result is bad.In order to improve the performance of image segmentation, we put forward a kind of adaptive window based on local Otsu method.First of all, the paper analyzes the traditional threshold method in complex image segmentation reason of not well, comparing the global threshold and partial threshold for image segmentation.Secondly, put forward according to the local information of local window size adaptively change method, analyzes the characteristics of this method.In the end, proved that the proposed new method is superior to other methods.The experimental results show that, compared with other traditional methods, this method can keep more details and obtain more satisfactory results.