基于脉冲耦合神经网络的图像分割

doi:10.11896/j.issn.1002-137X.2017.04.064

摘要/Abstract

摘要： 针对传统脉冲耦合神经网络(PCNN)模型在图像分割时需要设置较多参数和不能准确分割低对比度图像的问题,提出一种简化的PCNN模型和改进算法。在简化模型中减少了在传统PCNN模型中需要设置的参数的数量；在改进算法中根据图像像素空间和灰度特征自适应设置模型参数,并根据图像灰度直方图求出灰度期望均值作为图像分割阈值,因此该算法无需选择循环迭代次数,只需一次点火过程就能实现图像的有效分割。实验结果表明,该方法能准确分割图像,纹理细节清晰,分割结果优于人工调整参数的PCNN方法和Otsu方法。

关键词: 脉冲耦合神经网络,图像分割,参数设置,灰度期望均值

Abstract: Traditional pulse coupled neural network (PCNN) model needs to set a lot of parameters in processing image segmentation and can not segment the images with low contrast precisely.In order to solve the problems,an improved image segmentation algorithm was proposed based on a simplified PCNN model.In the simplified mode,the number of parameters required in the traditional PCNN model was reduced.In the improved algorithm,the model parameters were set adaptively according to the image pixel space and gray features,and the image grayscale expected mean was obtained as the image segmentation threshold according to the image gray histogram.Therefore,the improved algorithm has no iteration stop condition need to choose,just once the ignition process the method can complete the image segmentation effectively.The experimental results show that this method is accurate in image segmentation,especially in image texture details,and the final result is better than some methods,such as manual adjustment method of PCNN parameters and Otsu method.

Key words: Pulse coupled neural network,Image segmentation,Parameter setting,Grayscale expected mean

王爱文,宋玉阶. 基于脉冲耦合神经网络的图像分割[J]. 计算机科学, 2017, 44(4): 317-322. https://doi.org/10.11896/j.issn.1002-137X.2017.04.064

WANG Ai-wen and SONG Yu-jie. Image Segmentation Based on Pulse Coupled Neural Network[J]. Computer Science, 2017, 44(4): 317-322. https://doi.org/10.11896/j.issn.1002-137X.2017.04.064

参考文献

[1] WANG K J,DUAN S L,FENG W X.A survey of face recogni-tion using single training sample[J].PR &AI,2008,21(5):635-642.(in Chinese) 王科俊,段胜利,冯伟兴.单训练样本人脸识别技术综述[J].模式识别与人工智能,2008,1(5):635-642.
[2] MA Y D,DAI R L,LI L.Automated image segmentation using pulse coupled neural networks and images entropy[J].Journal of China Institute of Communications,2002,3(1):46-51.(in Chinese) 马义德,戴若兰,李廉.一种基于脉冲耦合神经网络和图像熵的自动图像分割方法[J].通信学报,2002,3(1):46-51.
[3] LIU Q,MA Y D,QIAN Z B.Automated image segmentationusing improved PCNN model based on cross-entropy[J].Journal of Image and Graphics,2005,10(5):579-584.(in Chinese) 刘勍,马义德,钱志柏.一种基于交叉熵的改进型PCNN图像自动分割新方法[J].中国图像图形学报,2005,0(5):579-584.
[4] LIU Q,XU L P,MA Y D,et al.Automated image segmentation using the ULPCNN model with ultra-fuzzy entropy [J].Journal of Xidian University,2010,7(5):817-823.(in Chinese) 刘勍,许录平,马义德,等.一种超模糊熵ULPCNN图像自动分割新方法[J].西安电子科技大学学报(自然科学版),2010,7(5):817-823.
[5] ZHANG Y D,WU L N.Image segmentation based on 2D Tsallis entropy with improved pulse coupled neural networks [J].Journal of Southeast University(Natural Science Edition),2008,8(4):579-583.(in Chinese) 张煜东,吴乐南.基于二维Tsallis熵的改进PCNN图像分割[J].东南大学学报(自然科学版),2008,8(4):579-583.
[6] QI Y F,HUO Y L,ZHANG J S.A automatic image segmen-tation method based on simplified PCNN and minimum scatter within clusters [J].Journal of Optoelectronics Laser,2008,9(9):1258-1264.(in Chinese) 齐永锋,火元莲,张家树.基于简化的PCNN与类内最小离散度的图像自动分割方法[J].光电子激光,2008,9(9):1258-1264.
[7] WANG H M,ZHANG K,LI Y J.Image segmentation method based on PCNN [J].Opto-Electronic Engineering,2005,2(5):93-96.(in Chinese) 王红梅,张科,李言俊.一种基于PCNN的图像分割方法[J].光电工程,2005,2(5):93-96.
[8] XIN G J,ZOU B J,LI J F,et al.Image segmentation with PCNN model and maximum of variance ratio [J].Journal of Image and Graphics,2011,16(7):1310-1316.(in Chinese) 辛国江,邹北骥,李建锋,等.结合最大方差比准则和PCNN模型的图像分割[J].中国图像图形学报,2011,6(7):1310-1316.
[9] HUO Y L.Automated image segmentation method based onPCNN [J].Automation & Instrumentation,2008,28(6):97-99.(in Chinese) 火元莲.一种基于PCNN的图像自动分割方法[J].自动化与仪器仪表,2008,28(6):97-99.
[10] MA Y D,QI C L.Study of automated PCNN system based on genetic algorithm [J].Journal of system simulation,2006,18(3):722-725.(in Chinese) 马义德,齐春亮.基于遗传算法的脉冲耦合神经网络自动系统的研究[J].系统仿真学报,2006,8(3):722-725.
[11] LU G F,WANG Y,DOU Y W.Automated PCNN image segmentation method with optimal parameters [J].Computer Engineering and Applications,2010,46(13):145-157.(in Chinese) 卢桂馥,王勇,窦易文.一种参数自动寻优的PCNN图像分割算法[J].计算机工程与应用,2010,6(13):145-157.
[12] SHEN W,ZHAO Z J,LI X G.Research on automated PCNN system based on cultural algorithm[J].Applied Science and Technology,2008,35(1):32-35.(in Chinese) 沈蔚,赵峙江,李晓刚.基于文化算法的PCNN自动系统的研究[J].应用科技,2008,5(1):32-35.
[13] WEN C J,WANG S S,YU H L,et al.Image segmentationmethod for maize diseases based on pulse coupled neural network with modified artificial bee algorithm [J].Transactions of the Chinese Society of Agricultural Engineering,2013,9(13):142-149.(in Chinese) 温长吉,王生生,于合龙,等.基于改进蜂群算法优化神经网络的玉米病害图像分割[J].农业工程学报,2013,9(13):142-149.
[14] OTSU N.A threshold selection method from gray level histograms [J].IEEE Transactions on Systems,Man and Cyberne-tics,1979,9(1):62-66.
[15] ECKHORN R,REITBOECK H J,A RNDT M,et al.Feature linking via synchronization among distributed assemblies:simulation of results form cat visual cortex [J].Neural Computation,1990,2(3):293-307.
[16] JOHNSON J L,PADGETT M L.PCNN models and applications[J].IEEE Transactions on Neural Netmorks,1999,10(3):480-498.
[17] SHEN Y,ZHANG X M,HAN K G,et al.Research of image segmentation technology based on PCNN [J].Modern Electro-nics Technique,2014,7(2):38-41.(in Chinese) 沈艳,张晓明,韩凯歌,等.PCNN图像分割技术研究[J].现代电子技术,2014,7(2):38-41.
[18] AN Q,LI M,HE Y J,et al.Novel PCNN Model and its Application on Image Segmentation[J].Computer Science,2014,41(6A):215-217.(in Chinese) 安琦,李敏,何玉杰,等.一种优化脉冲耦合神经网络模型及在图像分割中的应用[J].计算机科学,2014,41(6A):215-217.
[19] JIN W B,SHEN J J,ZHANG Z F,et al.Approach to image segmentation with spatial moments based on PCNN [J].Application Research of Computers,2009,6(12):4800-4802.(in Chinese) 金文标,沈晶晶,张智丰.一种基于空间矩的PCNN图像分割方法[J].计算机应用研究,2009,26(12):4800-4802.

Metrics

Viewed

Full text

Abstract

Cited

Shared

Discussed