一种有效的水下图像分割算法

水下图像具有信噪比低、边缘模糊等特点，直接使用传统分割方法对水下图像进行处理，效果较差，存在边缘扩大、轮廓变形等缺点。为提出新的抗噪性能好、能够克服水下成像过程中非线性影响的算法，文中在模糊理论的基础上，结合用熵的概念，提出了一种能根据图像自身特点自适应选择变换参数、使图像分割效果达到最佳的算法。通过对水下图像处理实验证明，该算法对简单背景的水下图像分割是有效的。和传统分割方法相比，该算法具有更强的自适应性和抗噪性能。

An Effective Approach to Underwater Image Segment

For underwater image,its S/N is low and the edge is fuzzy.If use traditional method to dispose it directely,the result is not expected.The edge is enlarged and distorted.So must find new approach to dispose the underwater image.At the same time,this new approach must be able to overcome the impact of nonlinear in the process of photoing and also has good noise restraining performance.In this paper,propose an approach that can select transformation parameter adaptively and get the best segmentation result by combining fuzzy theory and entropy.Experiments prove that the approach is effective for simple backgrounded underwater images.Comparing with the traditional method,this approach shows better adaptive and noise restraining performance.