基于改进最大类间方差法的灰度图像分割

Improved OTSU method on gray level image segmentation

对于光照不均匀的图像,用通常的图像分割方法不能取得满意的效果。为了解决这个问题,文中提出了一种实用而简便的图像分割方法。该算法采用局部最大类间方差法来确定各个子图像块的阈值。实验结果表明,用该方法能取得良好的分割效果。

To the objective-image segmentation in non-even bacckground,they cannot acquire good effect effect if using ordinary methods to segment in-rage.In order to solve this problem,offers a new way to segment images.The algorithm uses local OTSU to establish the threshold of sub-image block.The experiment proves that this way can get better result.