结合PCNN和自适应DGA算法的视网膜血管分割

为辅助诊断眼底疾病，提出一种眼底图像血管自动分割方法。首先利用对比度受限制的自适应直方图均衡化（CLAHE）技术与二维高斯匹配滤波器增强血管与背景对比度；然后利用自适应分布式遗传算法（ADGA）对PCNN参数设置自动寻优，将寻优得到的参数用于PCNN血管分割；最后采用面积滤波和区域连通性方法对分割结果进行后处理，得到优化后的血管检测结果。通过在国际上公认的彩色眼底图像库STARE中的实验结果表明，相比于利用传统的DGA算法对PCNN参数寻优，所提方法将分割的平均准确度从0.929?3提高到0.945?4，具有更高的鲁棒性、有效性和可靠性。

Retinal image segmentation based on PCNN and adaptive distributed genetic algorithm

A new blood vessels automatic detection method is proposed to diagnose fundus disease. In preprocessing, Contrast Limited Adaptive Histogram Equalization（CLAHE） and two-dimensional Gaussian matched filtering are adopted to improve the contrast between blood vessels and background. Then Adaptive Distributed Genetic Algorithm（ADGA） is used to search the best parameters and the blood vessel network is segmented by Pulse Coupled Neural Network（PCNN）. Finally, the final blood vessels detection result is obtained via post processing including area filtering and section connectivity filtering. The experiments implemented on the public STARE database indicate that the method has relatively higher robustness, effectiveness and reliability comparing with the traditional distributed genetic algorithm because of the average accuracy which is increased from 0. 929?3 to 0. 945?4.