基于PCNN图像分割新算法

New Algorithm of Image Segmentation Based on PCNN

在原有PCNN的基础上提出一种新的参数自适应PCNN模型,该模型中的各参数均由像素点本身决定,而不受人为的外界干涉,因此,该算法在图像处理领域有较好的运用价值。此外提出另一类分割判定准则——最大相关系数判定准则。最后通过对3种不同的图像分割实验结果分析后得出。该算法和最大相关系数判定准则在图像分割领域中具有较强的适应性和较好的辅助意义。

Based on the original PCNN model, this paper proposes a novel parameters adaptive PCNN model,in which all parameters are decided by pixel itself other than jamming in this model. The algorithm of the model has a great value in practical applications. Another segmentation criterion - Maximum Correlation Coefficient Criterion is also proposed in this paper. Based on the three different image segmentation results, this paper finally shows that the proposed algorithm and maximum correlation coefficient have significant application and assistance in image segmentation processing.