首页 | 本学科首页   官方微博 | 高级检索  
相似文献
 共查询到20条相似文献,搜索用时 135 毫秒
1.
基于 NSCT 域特征和 PCNN 的SAR 图像目标分割   总被引:1,自引:0,他引:1  
针对 SAR 图像的目标自动分割问题,在分析非下采样轮廓波变换和脉冲耦合神经网络的基础上,提出了一种基于非下采样轮廓波域特征图和 PCNN 的 SAR 图像目标分割算法.对 SAR 图像经过 NSCT 分解后的高、低频图像分别运用不同方式进行处理.对低频图用 PCNN 进行分割以获取目标所在的区域,对高频子带构造了特征图,对特征图利用 PCNN 进行分割以获取目标的精细结构.利用 MSTAR 数据进行了仿真实验,并与基于模糊 C 均值的分割算法、基于马尔可夫随机场的分割算法进行了对比.实验结果表明,所提出算法对 SAR 图像目标的分割结果更为准确,同时较其它算法具有更强的抗噪性能,是一种有效可行的 SAR 目标分割算法.  相似文献   

2.
马志荣  吴坚  孙富康 《硅谷》2010,(8):68-69
脉冲耦合神经网络(PCNN)理论是基于哺乳动物视觉皮层视觉神经细胞活动而产生的一种神经网络理论,在图像处理、语音识别领域有着广泛的应用。将描述在Xilinx公司的FPGA开发平台上应用Handel-C语言实现脉冲耦合神经网络(PCNN)模型的设计方案。同时,还将介绍应用该方法来实现图像预处理的过程。  相似文献   

3.
针对滚动轴承故障信号具有明显的非线性和非平稳特征,提出一种基于最优广义S变换和脉冲耦合神经网络(PCNN)的故障特征提取方法。首先采用基于时频聚集性最优化的广义S变换获取轴承故障信号的时频表示,然后利用脉冲耦合神经网络对最优广义S变换时频图进行二值分解,提取二值图像的捕获比序列用于表达故障信号的故障特征。对滚动轴承4种状态信号进行分析,验证方法的有效性。结果表明该方法能够提取出更加有效的轴承故障特征参数,有利于提高轴承故障诊断的精度。  相似文献   

4.
丁惠忠 《硅谷》2010,(9):113-114
在脉冲耦合神经网络(PCNN,Pluse Coupled Neural Network)模型基础上,提出简化PCNN模型用于对灰度图像进行伪彩色处理,以达到图像增强的目的。该方法通过快速并行计算,最大化实现RGB三色的相关性,合理地实现灰度级与彩色颜色的对应,与传统方法相比体现较好的优势。  相似文献   

5.
基于单阈值PCNN的边缘-参数运动模糊辨识算法   总被引:1,自引:1,他引:0  
针对存在着无旋转角度的线性位移模糊的图像,本文提出了一种新的基于脉冲耦合神经网络(PCNN)的边缘-参数曲线模糊分析法来辨识位移参数.该方法利用归一化的局部熵变换和改进的单阈值PCNN模型构造一个新的边缘因子,提取出模糊图像的边缘和纹理信息;通过边缘因子与对应的线性匀速运动参数构成的边缘-参数曲线,可以准确地辨识该模糊图像的位移量.实验结果表明,该算法产生的辨识曲线性能稳定,辨识方便准确,可辨识的有效线性移动模糊参数范围可到4~30个像素.  相似文献   

6.
基于NSCT和PCNN的红外与可见光图像融合方法   总被引:8,自引:2,他引:6  
提出了一种基于非采样Contourlet变换(NSCT)和脉冲耦合神经网络(PCNN)的红外与可见光图像融合方法.首先用NSCT对已配准的源图像进行分解,得到低频子带系数和各带通子带系数;其次对低频子带系数采取一种基于边缘的方法以得到融合图像的低频子带系数;对各带通子带系数提出了一种改进的基于PCNN的图像融合方法来确定融合图像的各带通子带系数;最后经过NSCT逆变换得到融合图像.实验结果表明,本文方法优于Laplaeian方法、小波方法和传统的NSCT方法.  相似文献   

7.
基于脉冲耦合神经网络和遗传算法的图像增强   总被引:1,自引:0,他引:1  
提出了基于脉冲耦合神经网络(PCNN-Pulse Coupled Neural Network)与遗传算法的图像增强算法.对PCNN模型进行了改进使之更适用于图像处理,并利用遗传算法的全局寻优能力自动搜索图像的最佳灰度阈值,再对图像进行处理.该算法既可有效地去除噪声,又能根据图像灰度性质自动选取最佳阈值,并对自适应分割后图像进行不同的灰度变换,从而实现了图像的自适应增强.试验结果表明,该算法显著提高了图像增强效果.  相似文献   

8.
《中国测试》2016,(8):108-112
为消除图像降噪过程中传统降噪方法对图像边缘和细节的影响,提出一种基于改进脉冲耦合神经网络(pulse coupled neural network,PCNN)赋时矩阵的有效滤除高斯噪声算法。该算法将PCNN模型的突触联结强度改进为随神经元与其周围神经元相似程度不同而变化的可变值,并将PCNN神经元的点火时间记录在赋时矩阵中,根据点火时刻判断噪声点,选择滤波方式。实验结果表明:该算法能够有效去除高斯噪声,具有较强的降噪性能及很好的边缘与细节保护能力。  相似文献   

9.
在对掌纹原始图像进行去噪、分割等预处理之后,利用平移不变的Zernike矩特征矢量(TIZMs)作为掌纹特征建立特征库,根据已知分类信息建立样本集。并将问题分解为多个小规模的两类问题,然后采用模块化神经网络(MNN)作为分类器进行掌纹识别。对香港理工大学的Polyu PalmprintDB数据库中的3200个掌纹进行实验,在响应时间和识别精度等方面获得了很好的结果。  相似文献   

10.
传统的脉冲耦合神经网络(PCNN)在图像降噪时不能准确地定位噪声数据并去除图像噪声.提出一种基于改进的PCNN有效去噪方法.该方法在PCNN模型上采用自适应的突触连接系数,使之随不同神经元与其周围神经元相似程度的不同而自适应变化,提高噪声数据的辨识度;同时将PCNN神经元的点火频次记录在点火时间序列中,根据神经元点火次数判断并滤出噪声点,实现更好地降噪效果.实验测试结果表明,该方法不仅可以准确地辨识噪声数据,而且能够有效地去除图像的噪声点,具有较强的适应性和较好的边缘与细节保护能力.  相似文献   

11.
A GPU-based discrete element method (DEM) with bonded particles is investigated to simulate the mechanical properties of sea ice in uniaxial compressive and three-point bending tests. Both the uniaxial compressive strength and flexural strength of sea ice are related to the microparameters in DEM simulation including particle size, sample size, bonding strength, and interparticle friction coefficient. These parameters are analyzed to build the relationship between the material macrostrengths of sea ice and the microparameters of the numerical model in DEM simulations. Based on this relationship, the reasonable microparameters can be calculated by given macrostrengths in the applications of simulating the failure processes of sea ice. In this simulation, both uniaxial compressive strength and flexural strength of ice increase with the increasing ratio of sample size and particle size. The interparticle friction coefficient is directly related to the compressive strength but has little effect on the flexural strength. In addition, numerical simulations are compared with experimental data to show the performance of the proposed model, and a satisfactory agreement is achieved. Therefore, this microparameter validation approach based on macrostrengths can be applied to simulate the complicated failure process of sea ice interacting with offshore platform structures.  相似文献   

12.
This paper addresses the fracture toughness and strength characteristics of ice taken from Notoro Lagoon in the Okhotsk Sea close to Hokkaido. Experimental values of tensile and bending strengths, and fracture toughness of sea ice conformed to Weibull statistical distribution. The proposed model predicts variation in fracture toughness as a function of the statistical distribution of ice grain sizes, effective surface energy, and elastic constants of ice. A very good agreement between experimental cumulative probability of fracture toughness and predicted distribution of fracture toughness of sea ice has been found. Computing the Weibull stress of sea ice, the dependence of fracture probability on stress intensity factor has been established. This result is in very good agreement with the presented method for the prediction of fracture toughness of sea ice.  相似文献   

13.
粘弹-塑性海冰动力学本构模型中的Drucker-Prager屈服准则   总被引:2,自引:1,他引:2  
王刚  季顺迎  吕和祥  岳前进 《工程力学》2006,23(6):154-161,153
针对中小尺度下海冰动力作用过程中的漂移和堆积特征,在粘弹-塑性海冰动力学本构模型中引入了Drucker-Prager(D-P)屈服准则。该模型在海冰屈服前采用Kelvin-Vogit粘弹模型,屈服后采用相关联的正则流动法则。采用该模型对规则海区内的海冰堆积过程进行了数值试验,计算的海冰堆积高度与其解析解一致。另外,在对渤海海冰动力过程中的海冰厚度、密集度、速度以及冰内应力进行的48小时数值模拟中,计算的冰厚分布与卫星遥感资料相符合。基于D-P准则的计算结果与Mohr-Coulomb(M-C)准则的结果相一致,但D-P准则克服了M-C准则计算塑性应力时的奇异现象,进而简化了计算过程。在以上数值模拟中,均采用了光滑质点动力学计算方法。以上数值计算均验证了基于D-P屈服准则的粘弹-塑性本构模型在海冰动力学中的可靠性。  相似文献   

14.
Diabetic retinopathy (DR) diagnosis through digital fundus images requires clinical experts to recognize the presence and importance of many intricate features. This task is very difficult for ophthalmologists and time-consuming. Therefore, many computer-aided diagnosis (CAD) systems were developed to automate this screening process of DR. In this paper, a CAD-DR system is proposed based on preprocessing and a pre-train transfer learning-based convolutional neural network (PCNN) to recognize the five stages of DR through retinal fundus images. To develop this CAD-DR system, a preprocessing step is performed in a perceptual-oriented color space to enhance the DR-related lesions and then a standard pre-train PCNN model is improved to get high classification results. The architecture of the PCNN model is based on three main phases. Firstly, the training process of the proposed PCNN is accomplished by using the expected gradient length (EGL) to decrease the image labeling efforts during the training of the CNN model. Secondly, the most informative patches and images were automatically selected using a few pieces of training labeled samples. Thirdly, the PCNN method generated useful masks for prognostication and identified regions of interest. Fourthly, the DR-related lesions involved in the classification task such as micro-aneurysms, hemorrhages, and exudates were detected and then used for recognition of DR. The PCNN model is pre-trained using a high-end graphical processor unit (GPU) on the publicly available Kaggle benchmark. The obtained results demonstrate that the CAD-DR system outperforms compared to other state-of-the-art in terms of sensitivity (SE), specificity (SP), and accuracy (ACC). On the test set of 30,000 images, the CAD-DR system achieved an average SE of 93.20%, SP of 96.10%, and ACC of 98%. This result indicates that the proposed CAD-DR system is appropriate for the screening of the severity-level of DR.  相似文献   

15.
基于PCNN的图像融合新方法   总被引:5,自引:0,他引:5  
余瑞星  朱冰  张科 《光电工程》2008,35(1):126-130
本文提出了一种基于PCNN的新型图像融合算法.首先,对待融合的两幅图像进行平稳小波分解得到两组多尺度图像;接着,取其中任意一组作为主PCNN的输入、另一组相应的图像作为从PCNN的输入,在每次迭代时,经并行PCNN点火后,得到一系列多尺度融合图像;然后,对它们进行平稳小波反变换得到每次迭代的融合结果;最后,计算每次迭代结果的信息熵,取信息熵值最大的融合图像作为最终结果.大量的实验以及与其它融合算法的比较分析,表明了本文算法的有效性和优越性.  相似文献   

16.
17.
The effect of long ice-coupled waves impinging on a tabular iceberg, an ice island or a thick sea ice floe trapped within a thin veneer of shore fast sea ice of substantial extent is considered. The waves most likely originate as ocean waves in the open sea beyond the fast ice boundary, from where they propagate into the sea ice. There their character is altered because of the flexural properties of the ice. The geophysical/engineering problem posed is solved by a Green's function method that redevelops, for a different surface boundary condition, an earlier study concerned with a freely floating ice floe. Reflection and transmission coefficients for the berg are found to depend strongly on its thickness and length. Amongst other things, the work relates to the operational safety of natural and artificially thickened Arctic ice platforms located in a contiguous ice sheet.  相似文献   

18.
蒋沛  火晓越  刘宝林  刘威  梁玮 《制冷学报》2020,41(2):159-166
细胞低温保存过程中,胞内冰损伤是其主要损伤因素之一,这可以通过控制冰晶成核来降低。本文总结了物理、化学及生物控制成核方法在低温保存领域的应用,归纳了各类控制成核方法的机理及对细胞产生的影响,对比得出生物及化学试剂控制成核方法较为优异。并提出最佳成核温度的概念,在该温度下,胞外冰产生且引导细胞适当脱水,减少胞内自由水,降低胞内冰的产生几率,为控制成核在细胞低温保存领域的应用指明了方向。  相似文献   

19.
冯常慧  刘强  张杰 《声学技术》2017,36(6):509-515
冰介质中的声速特性受温度、盐度、密度等因素影响,是一个随时间和空间变化的不稳定值。研究了冰介质中温度、盐度、密度三个物理量在垂直方向上的变化特性,引入孔隙率作为冰中声速特性的表征因子,利用孔隙率与温度、盐度、密度已有的函数关系,通过大量天然冰的实验数据,得到淡水冰中声速关于温度、盐度和密度的经验公式。进行了河冰声速测量的外场试验,验证了淡水冰中声速的经验公式的有效性,并利用该声速经验公式研究了淡水冰垂直方向上的声速剖面。文中所做的相关研究可为以后研究利用冰介质中的声速特性提供借鉴,为以后利用声学方法,在极地或我国北方高纬度地区进行冰层厚度监测,了解寒地冰层的更替、演化以及气候变迁等工作奠定研究基础。  相似文献   

20.
Single-beam upward-looking sonars mounted on British submarines have provided a vast amount of Arctic sea ice thickness data since the early 1970s. In this paper we develop a systematic procedure to estimate the bias introduced in the measurement of the ice draft by the standard procedure of taking only the first arrival of the returned echo. Because the magnitude of this error varies significantly with the depth of the submarine, the beamwidth of the sonar and the topography of the underside of the ice, it has to be calculated for each individual transect. Based on data collected during a submarine cruise in the Arctic Ocean in the winter of 2004, we estimate that for conic beams of semi-angles 3° and 6° the observed drafts are typically 7-20% and 15-35% higher than the real drafts. In view of the size of these errors, much higher than previously reported, we argue that beamwidth corrections must always be taken into account when measuring sea ice draft from below.  相似文献   

设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号