一种结合高帽变换和曲线演化模型的两阶段图像分割算法

针对图像形态学处理中高帽变换能有效提取图像中不均匀背景的特点,提出了一种结合高帽变换和曲线演化模型的两阶段图像分割算法。该算法有效的扩展了曲线演化模型的适用范围,即使对于背景不均匀的自然图像和医学图像也能有效的加以分割。实验结果表明:该算法既保留了原有曲线演化模型的优势,又能高效稳健、快速地完成曲线演化过程。

A Two-stage Image Segmentation Algorithm Combined Top-hat Transformation with Level Set Method

Traditional level set method is a very popular image segmentation algorithm for its highly flexibility.However,it is un able to handle images with uneven background which is a very common phenomenon existed among the natural optical and medical images.Meanwhile the morphological top-hat transform is very effective for estimating the uneven background.Thus we propose a two-stage algo rithm for image segmentation by combining top-hat transformation and level set method in this paper to remedy the drawback for the level set method to the uneven background.And the experimental results show that our proposed algorithm works perfectly to segment the images with uneven background.