基于遗传算法和简化PCNN神经网络的裂缝检测方法*

为能够在复杂背景下检测裂缝、分析裂缝图像特征,由脉冲耦合神经网络(pulse coupled neural networks, PCNN)的运行特征和神经元的状态变化分析简化PCNN模型,将简化PCNN模型用于裂缝图像的目标检测。针对PCNN无法确定裂缝图像的最优检测以及脉冲门限具有非线性因子的问题,提出了一种基于遗传算法(genetic algorithm, GA)和简化PCNN的裂缝图像检测方法。该方法采用最小误差准则作为遗传算法的适应度函数,并且根据遗传算法具有全局最优解的特点确定简化PCNN中各因子的值,实现了简化PCNN的裂缝图像自动分割。将本文方法与不同的分割方法对实际裂缝图像的处理结果进行比较,通过区域对比度、准确率和召回率等客观指标进行定量分析,表明了该方法对裂缝图像检测的有效性与通用性。

The Method of Crack Detection Based On Genetic Algorithm and simplified Pulse Coupled Neural Network

In order to detect cracks and analyze the characters of crack images in complicated background,PCNN model is simplified through analyzing of its running characters and the state changes of nerve cells,the simplified PCNN model is used in target detection of crack image.For PCNN model is not sure of the optimal detection of the crack images and pulse threshold with nonlinear factor,a method of crack detection of crack images is proposed based on genetic algorithm and optimized PCNN model.The method is on the minimum error principle as the fitness function of genetic algorithm, and according to the characteristics of the genetic algorithm has the global optimal solution to determine value of each factor in simplified PCNN model to realize automatic segmentation of simplified PCNN crack images.Comparing the processed results with other methods of segmentation come from the real crack images, quantitative analysis is conducted for the image after segmentation using region contrast,precision and recall objective indexes,experiment results show that the proposed crack detection method is effective and universal.