Modified-PCNN based detection of gyroscope pivot surface defects
-
摘要: 脉冲耦合神经网络(PCNN,Pulse Coupled Neural Network)与传统神经网络不同,不经过训练即可用于图像处理.针对PCNN模型中结构参数较多,且需要人工反复试验进行设置的困难,改进模型结构,简化了馈送输入和连接输入,减少了待定参数;根据邻域灰度动态地计算内部连接系数,由邻域的欧氏距离计算权值矩阵,再由图像的灰度特征计算动态阈值.将改进的PCNN用于陀螺轴尖表面缺陷图像的分割,用基于完整性与正确性指标的缓冲区匹配方法评价所提方法、最大熵法及Canny方法.针对不同缺陷图像的实验表明:所提算法的完整性与正确性都高于0.9,证明所提方法更有效.Abstract: Pulse coupled neural networks (PCNN) differs from traditional neural networks. PCNN can be applied to image processing without training. There are many structure parameters in PCNN model, and it is difficult to determine these parameters by manually trying. The model structure was improved by simplifying feedback input and connection input, and thus the number of the parameters was reduced. The inside connection coefficient was calculated dynamically based on neighborhood grayscale. The weight matrix was obtained by utilizing neighborhood Euclidean distance. The dynamic threshold was calculated from image grayscale character. The modified PCNN was used to segment several gyroscope pivot surface defects images. Based on the buffer region matching method, the completeness and correctness measures were used to compare the presented method, maximum entropy and Canny segmentation, and the results showed the two measures were not less than 0.9, which means that the proposed method is more effective.
-
Key words:
- pulse coupled neural network(PCNN) /
- gyroscope pivot /
- defect detection
-
[1] Eckhorn R,ReitBoeck H J,Arndt 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 [2] Ji Luping,Zhang Yi.A mixed noise image filtering method using [JP2]weighted-linking PCNNs[J].Neurocomputing,2008(71):2986-3000 [3] 张煜东,吴乐南.基于SPCNN和Nagao滤波的图像去噪[J].中国科学F辑:信息科学,2009,39(6):598-607
Zhang Yudong,Wu Lenan.Image denoising based on SPCNN and Nagao fileter[J]. Science in China,Series F:Information Science,2009,39(6):598-607(in Chinese)[4] Ji Luping,Zhang Yi,Shang Lifeng.An improved pulse coupled neural network for image processing[J].Neural Comput & Applic,2008(17):255-263 [5] Mure 瘙 塂 an R C.Pattern recognition using pulse coupledneural networks and discrete Fourier transforms [J].Neurocomputing,2003(51):487-493 [6] Gu Xiaodong.Feature extraction using unitlinking pulse coupled neural network and its applications[J].Neural Process Lett,2008(27):25-41 [7] 赵峙江,张田文,张志宏.一种新的基于PCNN的图像自动分割算法研究[J].电子学报,2005,33(7):1342-1344
Zhao Shijiang,Zhang Tianwen,Zhang Zhihong.A study of a new image segmentation algorithm based on PCNN[J].Acta Electronica Sinica,2005,33(7):1342-1344(in Chinese)[8] Berg H,Olsson R,Lindblad T,et al.Automatic design of pulse coupled neurons for image segmentation [J].Neurocomputing,2008(71):1980-1993 [9] Johnson J L,Padgett M L.PCNN models and applications[J].IEEE Transactions On Neural Networks,1999,10(3):480-498 [10] 彭真明,蒋彪,肖峻.基于并行点火PCNN模型的图像分割新方法[J].自动化学报,2008,34(9):1169-1173
Peng Zhenming,Jiang Biao,Xiao Jun.A novel method of image segmentation based on parallelized firing PCNN[J].Acta Automatica Sinica,2008,34(9):1169-1173(in Chinese)[11] Stewart R D,Fermin I,Opper M.Region growing with pulse-coupled neural networks:an alternative to seeded region growing[J].IEEE Transactions on Neural Networks,2002,13(6):1557-1662 -
点击查看大图
计量
- 文章访问数: 2274
- HTML全文浏览量: 134
- PDF下载量: 581
- 被引次数: 0
下载:
