基于脉冲耦合神经网络和遗传算法的图像增强

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．     

一种基于脉冲耦合神经网络的图像降噪方法

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.     

一种新的红外图像增强算法研究

为了解决传统红外图像增强方法存在的问题,该文提出了一种新的红外图像增强方法。首先,使用韦伯定理并结合人类视觉特性对红外图形进行预处理,预处理包括亮度增强和图像分割。其次,针对分割后的不同区域的特点使用不同的方法进行增强处理,使用多尺度Retinex对包括细节信息较多的区域2进行处理,使用改进的自适应直方图均衡算对区域一和区域三进行处理。最后,对处理过的红外图像进行重构。测试结果表明,该文提出的算法整体增强效果较好,更适合人类视觉观察,同时还可以缩短运行时间并且可以消除“光晕”现象,具有广阔的应用空间。     

基于神经网络的红外图像实时去斑方法

在分析红外图像斑纹产生机理基础上,提出利用人眼视网膜自适应调节原理的去斑方法。去斑时,根据距离不同作用不同原则,对中心及其周围像素分别予以不同权重;将斑纹信息加入求解过程,能有效快速去斑;去斑后,采用基于阈值的直方图规定化方法,以弥补对比度可能降低情况。研制出的DSP去斑系统实现了实时去斑功能。     

基于最优广义S变换和脉冲耦合神经网络的轴承故障诊断

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。     

一种结合IGM和改进PCNN的图像增强方法

针对部分对比度低、噪声大的图像,提出一种基于大脑内部生成机制(IGM)和改进脉冲耦合神经网络(PCNN)的图像增强方法。首先,根据IGM有关理论将原始图像分解为细节子图与粗糙子图;然后,采用改进的PCNN增强方法对粗糙子图进行处理,以提高整体对比度,采用PCNN与模糊集理论结合的增强方法对细节子图进行处理,增强边缘等细节信息并去除部分噪声;最后,将处理后的细节子图与粗糙子图重构,得到最终的增强图像。实验结果表明,该方法能够有效增强图像的对比度和纹理细节,减少部分噪声,较好地保留原图细节信息。     

红外图像实时增强的新算法

针对红外图像的特点,提出了一种红外图像实时增强的新算法。该算法通过分析图像的直方图,得到图像中目标像素数峰值的估计值,并作为平台直方图均衡化的阈值。用该阈值对直方图进行修正,然后通过修正后的直方图进行直方图均衡化。在FPGA内通过采用并行处理结构及流水线技术实现了该算法,并且每秒可处理25帧128×128×8bits的红外图像。理论分析和实验结果均表明,本算法克服了采用一般直方图均衡化增强红外图像的缺点?对背景和噪声增强过度,抑制了目标的增强。该算法对红外图像增强后,图像对比度是直方图均衡化增强后图像对比度的1.8倍。     

Contourlet变换和遗传算法相结合的沥青红外图像增强方法

沥青施工过程中,采集的红外图像容易受到周围环境噪声的影响,使图像变得模糊、信噪比低,从而导致后续图像处理分析的准确度降低。针对该噪声特性,提出了一种Contourlet变换和遗传算法相结合的红外图像增强方法。首先对原始红外图像进行Contourlet变换,得到带有多尺度、多方向信息的带通子带,然后对其进行模糊增强,并通过自适应遗传算法优化模糊增强参数,最后对增强后的带通子带进行Contourlet逆变换,得到效果增强的红外图像。实验结果表明,与其它几种常用的红外图像增强方法相比,此方法能更有效地抑制噪声,提高清晰度,取得了较好的增强效果。     

基于NSCT和PCNN的红外与可见光图像融合方法

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.     

基于小波Contourlet系数相关性的红外图像增强算法

针对现有红外图像存在噪声干扰与边缘模糊等问题,提出一种基于方向高频子带各层间系数相关性的红外图像增强算法.利用各层间小波Contourlet系数相关性的不同,合理对图像进行去噪和边缘优化处理.提出一种适用于文章算法的阈值选择方法,从而得到更为精确的去噪和边缘优化效果.仿真结果表明,实验结果与预期目标相符,能获得相比传统增强算法更高的PSNR值与PAI值.     

一种基于PCNN的图像分割方法

针对脉冲耦合神经网络无法确定最优分割的问题,提出了一种将脉冲耦合神经网络和类间方差准则相结合的图像分割方法。在每次迭代时将脉冲耦合神经网络点火的神经元对应的像素作为目标,未点火的神经元对应的像素作为背景,计算目标和背景之间的类间方差,取类间方差值最大的分割图像作为最终结果。实验结果表明该方法能获得视觉效果较好的分割结果并具有较强的普适性,对一幅大小为256×256的图像进行分割所需要的时间是0.8秒左右。     

利用波束形成和神经网络进行语音增强

语音增强在语音信号处理的前端非常重要,直接影响后端语音识别等效果。目前用神经网络进行单通道语音分离对于解决鸡尾酒会问题取得了很大的进步,但是用于复杂混合语音时分离效果仍不令人满意。针对单通道情形下的不足,使用多通道结构形成4个方向的超指向波束,结合神经网络算法实现对于指定方向的目标语音增强。仿真和实验结果表明,该算法相较于超指向波束形成算法和谱减法在多种评价指标上均有了明显的提升。     

Performance enhanced PSO-based modified Kohonen neural network for retinal image classification

Image classification is one of the significant applications in the field of ophthalmology for abnormality detection in retinal images. Image classification is a pattern recognition technique in which abnormal retinal images are categorized into different groups based on similarity measures. Accuracy and convergence rate are the important parameters of this automated diagnostic system. Artificial neural networks (ANNs) are widely used for automated image analysis systems. Kohonen neural networks (KNNs) are one of the prime unsupervised ANNs suitable for image processing applications. Besides the numerous advantages, KNNs suffer from two drawbacks: (a) lack of standard convergence conditions and (b) less accurate results. In this study, a novel approach is adopted to eliminate these disadvantages by performing suitable modifications in the conventional KNN. Initially, the fuzzy approach is an integrated one within KNN in the training algorithm to overcome the convergence difficulties. Second, a particle swarm optimization algorithm is used in feature selection for better accuracy. This proposed approach is tested on four different abnormal retinal image categories. The system is analyzed using several performance measures and the experimental results suggest promising results for the proposed system. Comparative analyses with other systems are also presented to show the superior nature of the proposed system.     

基于神经网络和方差的快速分形图像编码算法

为克服快速分形图像编码带来的解码图像质量下降问题,提出了一种神经网络与方差混合编码的快速分形图像编码算法.该算法结合图像子块复杂度与方差值的对应关系,根据每个区块的方差值大小选择适当的映射编码方法,即对于方差值相对小的区块采用方差编码以提高编码速度,对于方差值相对大的区块采用神经网络编码以提高编码质量.该算法可以较好地修正传统分形编码中由于自仿射映射结构限制所带来的解码质量偏低的问题,在大幅提高编码速度的同时,很好地保持了图像的编码质量.实验结果表明,该算法对比基本分形编码算法可以加速24倍,解码图像的质量对比方差快速分形编码算法有1.1dB的提高.同时,该算法的硬件实现比较容易,非常贴近实用化.     

红外序列图像的支持向量机分割方法

红外序列图像的准确分割是自动目标识别的关键,而当图像背景复杂时,传统的图像分割技术往往难以满足要求,为此,提出了基于支持向量机的红外序列图像分割方法。序列图像中的部分帧被作为训练样本,通过选择适合的模型参数,运用支持向量机方法建立学习机器,将后续图像帧中的目标从复杂的背景中识别出来,从而实现红外图像分割。实际红外序列图像分割表明,基于支持向量机的图像分割方法不需要复杂的预处理和后处理工作,分割效果理想,对于小目标的图像,识别正确率可达 99%。     

Effect of image resolution on intensity based scene illumination classification using neural network

In this paper, a framework for testing scene illumination classification with different image resolutions is proposed. The testing aims to provide the researchers with valuable information about the effect of image resolution on scene illumination classification using a neural network. The experiment is done by extracting three types of features from the images. These three types consist of statistical features, physic based features and histogram based features. It has been demonstrated that scene illumination classification can be affected by changing the image resolution. Despite the popular belief that high resolution images lead to better results, scene illumination classification by the proposed method performed best using low resolution images. At the second part of discussion, the reason behind this phenomenon is mathematically analysed and explained.     

Efficient image compression techniques for compressing multimodal medical images using neural network radial basis function approach

Image compression technique is used to reduce the number of bits required in representing image, which helps to reduce the storage space and transmission cost. Image compression techniques are widely used in many applications especially, medical field. Large amount of medical image sequences are available in various hospitals and medical organizations. Large images can be compressed into smaller size images, so that the memory occupation of the image is considerably reduced. Image compression techniques are used to reduce the number of pixels in the input image, which is also used to reduce the broadcast and transmission cost in efficient form. This is capable by compressing different types of medical images giving better compression ratio (CR), low mean square error (MSE), bits per pixel (BPP), high peak signal to noise ratio (PSNR), input image memory size and size of the compressed image, minimum memory requirement and computational time. The pixels and the other contents of the images are less variant during the compression process. This work outlines the different compression methods such as Huffman, fractal, neural network back propagation (NNBP) and neural network radial basis function (NNRBF) applied to medical images such as MR and CT images. Experimental results show that the NNRBF technique achieves a higher CR, BPP and PSNR, with less MSE on CT and MR images when compared with Huffman, fractal and NNBP techniques.     

An improved adaptive detail enhancement algorithm for infrared images based on guided image filter

Detail enhancement algorithms are important for raw infrared images to improve their overall contrast and highlight important information in them. To solve the problems that current algorithms like GF&DDE have, an improved adaptive detail enhancement algorithm for infrared images based on a guided image filter is proposed in this paper. It chooses the threshold for the base layer image adaptively according to the histogram statistical information and adjusts the mapping range of the histograms according to the dynamic range of the image. Besides, the detail layer is handled by a simpler adaptive gain control method to achieve the good detail enhancement effect. Finally, the base layer and the detail are merged according to the approximate proportion of the background and the details. Experimental results show that the proposed algorithm can adaptively and efficiently enhance different dynamic range images in different scenarios. Moreover, this algorithm has high real-time performance.     

顺序形态变换的图像增强算法

根据顺序形态变换的相关概念和性质,提出了一种新的图像增强算法。该算法通过对图像做局部加权均值滤波,得到图像增强的基值分量;采用多方位结构元素与图像边缘匹配,计算图像关于各个方位结构元素的加权均值并选取其中的最大值来确定边缘;将此最大值与基值分量之差作为增强分量来扩大图像灰度梯度的动态范围;针对图像中的高灰度区和灰度剧变区,应用图像局部均值和方差自适应调节增强系数。因此,算法在抑制图像中的高频噪声的同时,能有效提升图像中的边缘和目标。实验结果表明,增强前后图像标准差由41.1515,36.9133提高到62.0535,52.8331;图像熵由15.8463,16.8998减少到15.8156,16.8324。