基于多分辨率分析和分水岭的图像分割方法

Image segmentation based on multi-resolution analysis and watershed algorithm

提出了一种基于小波多分辨率分析和分水岭算法的图像分割方法。在小波分解后的低分辨率图像上进行分水岭分割,提高了分割的速度;由低分辨率图像返回到高分辨率图像时,采用了一种基于边缘信息的合并函数,避免了边缘信息的丢失,保证了分割的准确性。此外预处理过程中,在梯度图像上基于Rayleigh分布采用阈值处理的方法,有效抑制了高斯噪声对梯度图像的影响,避免了过分割。实验结果证明,本文所提出的基于小波多分辨率分析的图像分水岭分割算法能够很好地兼顾算法的效率和分割的准确性。

To overcome the shortcomings of traditional watershed segmentation： low calculated efficiency and over-segmentation, a novel image segmentation method based on wavelet multi-resolution analysis and watershed algorithm was proposed. Watershed segmentation was completed in low resolution to reduce the burden of computer. A new function based on the edges was presented to merge regions, which could detect the high frequency information lost in low resolution image. Besides, in order to suppress noise and avoid over-segmentation, an adaptive threshold based on Rayleigh distribution was also proposed for gradient image. Experiments show that the proposed method can ensure both calculated efficiency and segmentation accuracy.