一种改进的细胞神经网络图像边缘提取方法  

An improved image edge extraction method for cellular neural networks

作　　者：张宪红[1] 张春蕊[2] 

机构地区：[1]东北林业大学机电工程学院,黑龙江哈尔滨150040 [2]东北林业大学理学院数学系,黑龙江哈尔滨150040

出　　处：《暨南大学学报（自然科学与医学版）》2017年第2期178-184,共7页Journal of Jinan University(Natural Science & Medicine Edition)

基　　金：国家自然科学基金项目(11601070);黑龙江省自然科学基金项目(2015A016)

摘　　要：目的:为了进一步提高细胞神经网络的处理速度,减少计算的迭代次数.方法:首先对细胞神经网络模型的输出函数进行改进,并对模板的取值范围进一步确定.其次,用反应扩散方程来改进阈值,并融入了图像梯度信息和构造的调整函数,使其可以自动根据图像的梯度生成一个扩散矩阵,自适应地为每一个像素点选取不同的阈值.结果:图像处理的运行时间和迭代次数都减少了一半,图像边缘提取的结果更加精确.结论:用此方法选取2种不同类型的图像进行MATLAB仿真,与几种常用的边缘提取算法处理结果相比,在视觉效果上取得了很大提升,并提高了边缘提取的效率和准确性.Aim： In order to improve the processing speed and reduce the number of iterations.Methods： In this paper,the output function of cellular neural network model is improved,and the range of the template is further determined. Secondly,this paper uses reaction diffusion equation to improve the threshold,and the gradient information of image reconstruction and adjustment function,which can automatically generate a diffusion matrix according to the gradient information of the image,adaptively selecting a different threshold for each pixel. Results： The running time and the number of iterations are reduced by half,and the result of image edge detection is more accurate. Conclusion： This method uses MATLAB to simulate 2 different types of images. Compared with several commonly used edge extraction algorithms,the visual effect is greatly improved,and the efficiency and accuracy of edge detection are improved.

关 键 词：边缘提取 细胞神经网络 反应扩散方程 边缘停止函数 梯度 

分 类 号：TP391.4[自动化与计算机技术—计算机应用技术]