小波变换在图像处理中的应用探析

简要介绍小波变换理论的发展背景,根据傅立叶变换(FFT)方法和短时傅立叶变换(SFFT)方法基本原理和处理信号的特点,结合小波变换原理的时频分析特性和对时变信号的处理能力,文章将小波变换方法引入到了图像信息处理过程中。由于此方法能从图像中获取更多的特征信息,且具有较高的分析精度,并具有较好的重构效果,所以使用该方法对图像进行分析和处理时,效果显著。     

基于Daubechies小波的图像边缘检测技术

利用Daubechies正交小波变换的性质,通过Mallat多尺度分析方法对图像进行小波变换,把图像分解成低频轮廓,水平高频、垂直高频和斜线高频四个部分。针对图像边缘主要集中在高频部分,该文先保持小波变换后的高频小波系数,同时对低频小波系数进行再次小波变换,提取出次高频信号的边缘信息。最后对保留下来的高频小波系数和次高频小波系数进行逆变换获取最大边缘信息。     

一种基于小波变换的目标融合跟踪方法

本文在分析小波变换的基础上，将小波分析应用到目标图像的融合跟踪技术上，利用小波的多尺度和多分辨特性，不仅能够获得不同分辨力下的图像序列，进行目标图像融合；还能有效地从信号中提取突变信号。对函数或信号进行多尺度的细化分析。图像边缘用小波变换进行处理和提取并对图像形心进行计算。能够得到较好的轮廓提取效果和形心定位精度，进而说明了小波变换可能成为目标跟踪中较好的数学方法。     

基于小波变换模极大值的自适应阈值图像去噪算法

依据白噪声小波变换性态与信号奇异性相比具有显著不同的特点，在大尺度下设置阈值，去掉噪声信号而保留图像细节信号引起的模极大值点。在阈值设置问题上，采用自适应阈值的方法，克服单一阈值不能在每级尺度上将信号与噪声作最大分离的缺点。实验表明，与单一阈值去噪方法相比，该方法不仅可以保留图像边缘信息，而且能提高去噪后图像的峰值信噪比2～5dB。     

正交小波包分析在遥感图像融合中的应用

小波变换具有优良的时频局部特性，但由于其尺度是按二进制变化的，存在“高频低分辨”这一缺陷．正交小波包分析能够将信号(图像)频带进行多层次划分，对多分辨分析没有细分的高频部分进一步分解，从而提高了频率分辨率，能有效地提取特定的频率成分．推导了小波包分析的基本原理，给出了基于正交小波包分析的遥感图像融合算法．最后，通过实例说明正交小波包分析的有效性和优越性．     

图像边缘检测二维小波算法研究与实现

边沿作为图像视觉的最主要特征，成为图像信息获取的重要内容。小波变换具有检测局域突变的能力，而且可以结合多尺度信息进行检测，因此成为图像信息边缘检测的优良工具。根据二维小波变换的特点,分析了利用二维小波进行图像边缘检测的基本原理，并设计了利用二维小波变换进行多尺度边缘匹配的检测算法。基于研究结果，编写了计算机应用程序，进行实例分析。     

小波变换遥感图像的数据融合

利用小波变换方法进行多卫星遥感图像数据融合，分析不同长度的小波基对融合图像的影响，从信息的保持性、视觉效果及运用灵活性等方面与IHS、PCA融合算法进行了比较，从而探讨这一新算法在遥感图像分析应用中的可行性。     

图像边缘检测的二维连续小波变换法

提出一种新的基于二维连续小渡变换的图像边缘检测方法。算法充分利用连续小渡变换探测信号奇异性的能力和小波的多分辨率特性。对图像进行二维连续小波变换后,直接对各个分解层的正的小波系数或负的小波系数利用闽值法初步确定图像的边缘点,然后利用“投票选举法”综合各个尺度的边缘信息。这种方法不需要进行边缘点连接,可以直接得到光滑连续的边缘。实验结果表明,新方法能够准确探测图像中目标的整体轮廓和细节信息,在性能上优于传统的两种边缘检测算子。     

小波变换在信号处理中的应用

介绍了常用小波基函数和连续小波变换，重点将小波变换应用到信号的去噪、分解和重构信号分析中。通过仿真实验可见小波变换已成为信号处理中行之有效的方法，必将在信号处理中起着重要作用。     

基于连续小波变换的旋转机械振动信号灰度矩研究

通过对机械振动信号的连续小波变换，利用小波滤波器良好的时频特性，研究了振动信号连续小波变换灰度图的统计特征，提出了“一阶灰度矩向量”指标。对8种典型故障信号的计算表明，一阶灰度矩向量可以较好地展现小波灰度图的局部信息，能够有效地提取信号的特征，区分振动故障。     

高速旋转物体的频闪结构光三维面形测量系统

提出了一个用于测量高速旋转物体三维面形的频闪结构光照明系统。系统自动跟踪检测旋转物体的转动信息,再用该信息同步控制光源发光和摄像系统工作,记录下旋转物体在闪光时刻的瞬间“静止”图像。使用该系统产生持续时间为424s的同步频闪结构光,对转速为每分钟1080转的家用电风扇旋转叶片三维面形进行测量,证明该系统能够准确地获得旋转叶片的瞬间静止图像,便于重建旋转叶片每个瞬间时刻的三维面形。同时,该系统还具备人工设定光源发光频率的功能,可以拓宽应用到无重复特征信号的高速运动物体三维面形测量。     

Depth Transfer by an Imaging System

In many applications such as three-dimensional (3-D) data acquisition, the scanning of 3-D objects or 3-D display, it is necessary to understand how an imaging system can be used to obtain information on the structure of an object in the direction perpendicular to the image plane, i.e. depth information. In certain cases the formation of a 3-D image can be described by a theory based on optical transfer functions (OTF): the image intensity distribution is given by the 3-D convolution of the object and a 3-D point spread function (PSF); equivalently, in 3-D Fourier space the image spectrum is the product of the object spectrum and a 3-D OTF. This paper investigates the 3-D PSFs and OTFs that are associated with different pupil functions of the imaging system.     

Two-Dimensional Image Transmission Based on the Ultrafast Optical Data Format Conversion between a Temporal Signal and a Two-Dimensional Spatial Signal

We have experimentally demonstrated a two-dimensional (2-D) image transmission based on the ultrafast optical data format conversion between a temporal signal and a spatial signal with an ultrashort optical pulse. In the proposed system we adopt a spectral holography technique to transmit a one-dimensional (1-D) spatial signal and use a spatial-domain time-frequency transform to realize a transform between 1-D and 2-D spatial signals. By use of these techniques, a low-optical-loss transmission system can be constructed. To demonstrate a 2-D image transmission with this technique, we achieved experimentally transmission of the alphabet letter T as a 3 x 3 pixel 2-D spatial image.     

Three-dimensional matching by use of phase-only holographic information and the Wigner distribution

Matching of three-dimensional (3-D) objects is achieved by Wigner analysis of the correlation pattern between the phase-only holographic information of a reference object and that of a target object. First, holographic information on the reference object and on the target object is extracted by use of optical scanning holography as a form of electrical signal. This electrical information is then stored in a computer for digital processing. In the digital computer, the correlation between the phase-only information of the hologram of the reference object and that of the target object is calculated and analyzed by use of a Wigner distribution. The Wigner distribution yields a space-frequency map of the correlation pattern that indicates whether the 3-D image of the target object matches that of the reference object. When the 3-D image of the target object matches that of the reference object, the Wigner distribution gives a well-defined line that directly indicates the 3-D location of the matched target object. Optical experiments with digital processing are described to demonstrate the proposed matching technique.     

快速傅里叶变换FFT及其应用

利用快速傅里叶变换FFT将图像信号从空间域转换到频域进行分析,使快速卷积、目标识别等许多算法易于实现;然后根据图像信号的灰度结构特征和频谱分布,用Butterworth带通滤波器和二维维纳滤波器进行滤波处理,去除图像信号中的低频干扰和噪声信号;再利用傅里叶反变换将信号还原。结果显示,处理后的模拟远程高空卫星照片轮廓清晰可见。     

Three-dimensional image compression with integer wavelet transforms

A three-dimensional (3-D) image-compression algorithm based on integer wavelet transforms and zerotree coding is presented. The embedded coding of zerotrees of wavelet coefficients (EZW) algorithm is extended to three dimensions, and context-based adaptive arithmetic coding is used to improve its performance. The resultant algorithm, 3-D CB-EZW, efficiently encodes 3-D image data by the exploitation of the dependencies in all dimensions, while enabling lossy and lossless decompression from the same bit stream. Compared with the best available two-dimensional lossless compression techniques, the 3-D CB-EZW algorithm produced averages of 22%, 25%, and 20% decreases in compressed file sizes for computed tomography, magnetic resonance, and Airborne Visible Infrared Imaging Spectrometer images, respectively. The progressive performance of the algorithm is also compared with other lossy progressive-coding algorithms.     

Three-dimensional blind deconvolution of ultrasound images

Three-dimensional ultrasound images are blurred by the ultrasound pulse through the convolution between the 3-D tissue signal and the 3-D pulse. The blurring reduces the spatial resolution of the 3-D ultrasound images and, consequently, their diagnostic value. This paper presents a method for 3-D blind homomorphic deconvolution of medical 3-D ultrasound images to improve their spatial resolution. The blind estimate of the 3-D pulse is necessary because the pulse changes in spatial extent and frequency composition as it passes through the tissues and because the pulse is not separable in its spatial dimensions. The method was tested on a 3-D image of a phantom with anechoic spheres of known size in a uniform diffuse scattering matrix. The spheres were clearly better defined and had volumes much closer to the true volume in the deconvolved image than in the original image     

基于哈夫曼编码的关联成像算法的图像传输机理研究

目的在图像传输过程中存在信道噪声干扰、频谱资源紧张和误码率较高等问题,限制了图像传输质量和效率,降低了信道的可靠性,研究能够在降低图像传输信息量的同时提高传输图像的分辨率,进而提高信道的可靠性和安全性。方法通过在最有效地利用哈夫曼编码信道传输能力的前提下,尽可能多地降低信息传输中的误码率,综合信道编码的冗余信息和关联成像传输信道信息。结果提高了传输信号的可靠性,保证了数据的传输质量,实现了稳定快速通信。结论该研究基于信道编码关联成像的图像传输机理,充分利用了哈夫曼编码的无损耗数据压缩特性,在降低图像传输信息量的同时提高了传输图像的分辨率,进而提高了信道的可靠性和安全性。     

Real-time 3-D ultrasound imaging using sparse synthetic aperture beamforming

A method for real-time three-dimensional (3-D) ultrasound imaging using a mechanically scanned linear phased array is proposed. The high frame rate necessary for real-time volumetric imaging is achieved using a sparse synthetic aperture beamforming technique utilizing only a few transmit pulses for each image. Grating lobes in the two-way radiation pattern are avoided by adjusting the transmit element spacing and the receive aperture functions to account for the missing transmit elements. The signal loss associated with fewer transmit pulses is minimized by increasing the power delivered to each transmit element and by using multiple transmit elements for each transmit pulse. By mechanically rocking the array, in a way similar to what is done with an annular array, a 3-D set of images can be collected in the time normally required for a single image.     

Visual communication with retinex coding

Visual communication with retinex coding seeks to suppress the spatial variation of the irradiance (e.g., shadows) across natural scenes and preserve only the spatial detail and the reflectance (or the lightness) of the surface itself. The separation of reflectance from irradiance begins with nonlinear retinex coding that sharply and clearly enhances edges and preserves their contrast, and it ends with a Wiener filter that restores images from this edge and contrast information. An approximate small-signal model of image gathering with retinex coding is found to consist of the familiar difference-of-Gaussian bandpass filter and a locally adaptive automatic-gain control. A linear representation of this model is used to develop expressions within the small-signal constraint for the information rate and the theoretical minimum data rate of the retinex-coded signal and for the maximum-realizable fidelity of the images restored from this signal. Extensive computations and simulations demonstrate that predictions based on these figures of merit correlate closely with perceptual and measured performance. Hence these predictions can serve as a general guide for the design of visual communication channels that produce images with a visual quality that consistently approaches the best possible sharpness, clarity, and reflectance constancy, even for nonuniform irradiances. The suppression of shadows in the restored image is found to be constrained inherently more by the sharpness of their penumbra than by their depth.