优化的PCNN自适应三维图像分割算法*

脉冲耦合神经网络(pulse coupled neural network,PCNN)对图像分割具有天然的优势,但是传统的PCNN模型参数难以确定,且算法耗时多。对多种PCNN模型进行研究改进,并利用统计学知识提出了一种精简高效的自适应三维分割算法。将其用于脑部磁共振成像(magnetic resonance imaging,MRI)图像的分割,把脑组织分成白质、灰质和脑脊液。与标准PCNN、传统的Otsu阈值方法、SPM8工具箱及专家手动分割结果的对比实验表明,该自适应算法具有精确性、高效性。

Adaptive 3D image segmentation based on optimized PCNN

The segmentation algorithm with PCNN model has many natural advantages. But the typical PCNN model has too many parameters difficult to determinate and consumes too much time. This paper proposed an effective three dimension image segmentation model that integrated multiple PCNN models and the statistical model. It was used to segment brain MRI image into gray matter(GM), white matter(WM) and cerebrospinal fluid(CSF). And the segmentation results were compared with those of standard PCNN, traditional Otsu threshold, SPM8 toolbox and expert segmentation. It demonstrates this adaptive method is fairly accurate and effective.