Description: 此设计要求能够实现将医学图像进行识别的过程，包括了图像预处理、图像特征提取及分类判决三大模块。在预处理这一步中主要实现的是将彩色图像转换为灰度图像，灰度图像的二值化，直方图修正，去除干扰、噪声以及差异，边缘增强等；第二模块是图像的特征提取。由于对象的物理与几何特性差异，在影像中表现为局部区域的灰度产生明显变化，形成影像特征，而图像特征提取就是对其进行加工、整理、分析、归纳以便提取构成目标影像的特征，得到能反映图像内容区别于其他事物的本质特征；分类判决作为第三模块，则是要在第二步的基础上采用某种分类判别函数与判别规则，通过对目标特征的分析和匹配来识别目标。-this design requirements to achieve medical image identification process, including the image preprocessing, Image feature extraction and classification ruling three modules. This pretreatment step in the main achievement of the color image is converted to grayscale images, the two gray-scale image value, histogram amendment to remove interference and noise variance Edge Enhancement; the second module is the image feature extraction. Because of the physical objects with geometric characteristics difference in the images showed local area Gray significant changes, the video features, feature extraction and image is its processing, compilation, analysis, summarized in order to extract constitute the target image features can be reflected image as distinct from the other characteristics of th Platform: |
Size: 6630 |
Author:uhih |
Hits:
Description: 此设计要求能够实现将医学图像进行识别的过程，包括了图像预处理、图像特征提取及分类判决三大模块。在预处理这一步中主要实现的是将彩色图像转换为灰度图像，灰度图像的二值化，直方图修正，去除干扰、噪声以及差异，边缘增强等；第二模块是图像的特征提取。由于对象的物理与几何特性差异，在影像中表现为局部区域的灰度产生明显变化，形成影像特征，而图像特征提取就是对其进行加工、整理、分析、归纳以便提取构成目标影像的特征，得到能反映图像内容区别于其他事物的本质特征；分类判决作为第三模块，则是要在第二步的基础上采用某种分类判别函数与判别规则，通过对目标特征的分析和匹配来识别目标。-this design requirements to achieve medical image identification process, including the image preprocessing, Image feature extraction and classification ruling three modules. This pretreatment step in the main achievement of the color image is converted to grayscale images, the two gray-scale image value, histogram amendment to remove interference and noise variance Edge Enhancement; the second module is the image feature extraction. Because of the physical objects with geometric characteristics difference in the images showed local area Gray significant changes, the video features, feature extraction and image is its processing, compilation, analysis, summarized in order to extract constitute the target image features can be reflected image as distinct from the other characteristics of th Platform: |
Size: 6144 |
Author:uhih |
Hits:
Description: ：为实现番茄收获机械自动化，提出了利用计算机视觉代替人眼对成熟番茄进行自动识别的方法。首先对番茄图像进
行了各种去除噪声方法的比较，提出用HSI 颜色系统的H 色调分量作为识别的颜色特征参数。利用最大方差自动取阈法及
数学形态运算等构成的识别算法对采集的番茄图像进行自动识别.-: In order to realize automation of tomato harvesting, the use of computer vision to replace the human eye of a mature tomato automatic identification method. First, a variety of tomato image noise removal methods, the use of HSI color hue component H system as a means of identification of the color feature parameter. Automatic check using the maximum variance threshold method and mathematical form of computing, such as recognition algorithm consisting of tomato collected images of automatic identification. Platform: |
Size: 94208 |
Author:daiyu |
Hits:
Description: 对图像进行斑点噪声的添加，用方程f=f+n*f将乘性噪音添加到图像f上，其中n是均值为零，方差为var的均匀分布的随机噪声。-Image speckle noise addition, the equation f = f+ N* f will be added to the multiplicative noise on the image f, in which n is zero mean, variance var of the uniform distribution for the random noise. Platform: |
Size: 1024 |
Author:刘昊天 |
Hits:
Description: 先由原始图像（任选）产生待恢复的图像；(产生方法如下：冲激
函数为 ，将原始图像与冲激函
数卷积产生模糊，然后再迭加均值为0，方差为8，16，32的高斯
随机噪声而得到一组待恢复的图像 ；
-By the original images (optional) have to be the restoration of the image (the method is as follows: impulse function, the original image with the impulse function arising convolution fuzzy, and then stacking the mean is 0, the variance for the 8,16 , 32 the Gaussian random noise has been a group of images to be restored Platform: |
Size: 459776 |
Author:李 |
Hits:
Description: 分块有损压缩图像忽略了块间相关性，重构时会产生块效应，该文提出一种空域自适应去块效应算法。对块边缘采用方向自适应
有理滤波，以弱化块效应。根据块的内部活动性将图像块分成平坦块和纹理块2 类，利用基于方差的空域检测方法检测出平坦块，并对平
坦块进行邻块边缘自适应平滑。实验结果表明，该算法有效去除了块效应，一定程度上提高了信噪比，算法简单且鲁棒性较好。-Block lossy compression image ignores the inter-block correlation, Reconstruction would have a blocking effect, the paper presents a airspace Adaptive Deblocking Algorithm. To block the edge of rational use of the direction of adaptive filtering, in order to weaken the block effect. According to block the internal activity of the image block is divided into a flat block and texture block 2 categories, the use of airspace based on the variance detection methods to detect the flat block, and a flat block adjacent block edge adaptive smoothing. Experimental results show that the algorithm effectively blocking go except, to some extent, improve the signal to noise ratio, the algorithm is simple and better robustness. Platform: |
Size: 147456 |
Author:子登 |
Hits:
Description: 计算SENSE重建图像中的g-factor,这是并行磁共振成像SENSE算法的关键一步-G-factor is the metric to quantify the amplificaiton of noise power in reconstructing SENSE accelerated image. The detail was presented in Pruessmann s 1999 Magn. Reson. Med. paper. In theory, g-factor is the pixel-by-pixel ratio of the image variance between the SENSE reconstructed image and un-accelerated image. And g-factor depends on: (1) the k-space data trajectory (2) coil sensitivit profiles (3) noise covariance matrix of the array coil Platform: |
Size: 5120 |
Author:李荣智 |
Hits:
Description: 最小邻域均值投影函数及其在眼睛定位算法.提出一种投影函数：最小邻域均值投影函数．该函数通过计算每条投影线上各像素点邻域均值的最小值
来跟踪图像中的低灰度特征．与传统的积分投影函数和方差投影函数相比，它以求最小值的局部选择性代替传统投
影函数的全局累加性，因此具有对片状噪声不敏感的特点、此外，在计算过程中，它还能记录最小值点的二维位置信
息，是一个二维的搜索算子、最小邻域均值投影函数的这些特点使其非常适合于眼睛定位．它对眼睛，特别是瞳孔，总
能够产生精确、鲁棒的响应通过在CAS—PEAL数据库和BioID数据库上的实验表明，其定位正确率与精确度均高
于传统的投影函数．-A projection function called minimal neighborhood mean projection function(MNMPF)is proposed．
The projection function calculates and stores the minimal neighborhood mea1]of each pixel on each projection line，
SO that it is able to trace the low grayscale features in image．Compared with traditional projection functions，i．e．
integral projection function(IPF)and variance projection function(VPF)，MNMPF is insensitive to sheet noise，due
to the local selectivity of its mimmum operation．During the computation of MNMPF,the image locations of
minima are recorded at the same time．This makes MNMPF a 2D operator．All these properties of MNM PF are very
suitable for eye location．It can bring precise and robust response to eyes，especially pupils．Experiments on
CAS—PEAL and BioID databases show its excellent correct rate and precision over traditional projection functions． Platform: |
Size: 436224 |
Author:郭事业 |
Hits:
Description: Abstract—The fingerprint segmentation algorithm based on gray
variance can t segment those fingerprint images with high noise.
After analyzing limitation of the fingerprint image segmentation
algorithm based on gray variance, the paper proposes the
improved algorithm to acquire the gray average and gray
variance combining to the basic gray distributing character in the
valid fingerprint image region. Experimental results indicate that
the gray average and gray variance based on the improved
algorithm are more representative of the original information of
the fingerprint image region than the gray variance based on the
classical algorithm. The segmented results of the Platform: |
Size: 736256 |
Author:oopsingh |
Hits:
Description: 本实验开发一个通用程序， 用在后续的几个实验中。(a) 编写一个给图像中添加高斯噪声的程序，程序的输入参数为噪声的均值与方差。
(b) 编写一个给图像中添加椒盐噪声的程序，程序的输入参数为两个噪声分量的概率值。
-In this study, the development of a common procedure, used in the follow-up of several experiments. (A) preparing a program to add Gaussian noise to the image, the input parameters of the program is the noise mean and variance. (B) prepare a program to add salt and pepper noise in the image, the input parameters of the program is the probability that the value of the two noise components. Platform: |
Size: 123904 |
Author:张亚丽 |
Hits:
Description: 此文档中的代码可以计算5种图像的质量参数，分别是：图像灰度均值、图像方差、图像平均梯度、熵和峰值信噪比及均方误差。-The code in this document can be calculated five kinds of image quality parameters, namely: the image gray value, image variance, average gradient image, entropy and peak signal to noise ratio and mean square error. Platform: |
Size: 7168 |
Author:玄林江 |
Hits:
Description: 通过观察图像lena_noise.bmp可以看到，该图像主要包含高斯噪声、椒盐噪声等。
通过查阅资料可知，维纳滤波可以很好的滤除高斯噪声，而中值滤波对于椒盐噪声的滤出效果优异，因此本次采用维纳+中值的滤波方式进行图像的去噪处理。
图像常用评价方法有：信噪比、相关系数、清晰度。信噪比较为常用；相关系数 由于滤波会使原图像失去某些细节，因此这种评价方式为参考性指标而非绝对评判标准；清晰度 为图像子块的方差
综合考虑采用图像的峰值信噪比(PSNR)作为评价去噪效果指标(自编PSNR.m函数).-By observing the image lena_noise.bmp can see that the image consists mainly of Gaussian noise, salt and pepper noise. Through access to information shows that the Wiener filter can be well filtered Gaussian noise, and median filter for impulse noise filter out excellent results, so this value using Wiener filtering method+ in image denoising. Image evaluation methods commonly used are: noise ratio, correlation coefficient, clarity. SNR is more commonly used correlation coefficient as filtering the original image will lose some of the details, so this evaluation method as the reference index rather than absolute criteria clarity of image sub-block variance considering the peak signal using image noise ratio (PSNR) as an evaluation index denoising (self PSNR.m function). Platform: |
Size: 64512 |
Author:献诗 |
Hits:
Description:
Wavelet denoising
For using this code need to use signal toolbox and general toolbox in your matlab
In the first part of this assignment, we asked to obtain a (black-and-white) digital image of size
512 by 512 and then generate noisy image by adding a Gaussian noise but under the condition of
having SNR=20dB by select the suitable value of variance for Gaussian noise formula.
Second step is performing wavelet denoising using the hard thresholding (Use the db 6 for four
levels) in the condition of finding the optimal thresholding value of T in terms of the SNR obtained. It
means that, we should find the highest SNR value by finding the suitable value for threshold. Then we
asked to do the same process but this time using soft thresholding.
Finally for the last part of question one, we should compare the results of the obtained SNR
with the recommendations of 3*sigma for the hard thresholding and 3/2*sigma for the soft
thresholding.-
Wavelet denoising
For using this code need to use signal toolbox and general toolbox in your matlab
In the first part of this assignment, we asked to obtain a (black-and-white) digital image of size
512 by 512 and then generate noisy image by adding a Gaussian noise but under the condition of
having SNR=20dB by select the suitable value of variance for Gaussian noise formula.
Second step is performing wavelet denoising using the hard thresholding (Use the db 6 for four
levels) in the condition of finding the optimal thresholding value of T in terms of the SNR obtained. It
means that, we should find the highest SNR value by finding the suitable value for threshold. Then we
asked to do the same process but this time using soft thresholding.
Finally for the last part of question one, we should compare the results of the obtained SNR
with the recommendations of 3*sigma for the hard thresholding and 3/2*sigma for the soft
thresholding. Platform: |
Size: 91136 |
Author:jams1166 |
Hits:
Description: we present an improved fuzzy
C-means (FCM) algorithm for image segmentation by introducing
a tradeoff weighted fuzzy factor and a kernel metric. The
tradeoff weighted fuzzy factor depends on the space distance of
all neighboring pixels and their gray-level difference simultaneously.
By using this factor, the new algorithm can accurately
estimate the damping extent of neighboring pixels. In order to
further enhance its robustness to noise and outliers, we introduce
a kernel distance measure to its objective function. The new
algorithm adaptively determines the kernel parameter by using
a fast bandwidth selection rule based on the distance variance
of all data points in the collection. Furthermore, the tradeoff
weighted fuzzy factor and the kernel distance measure are both
parameter free. Experimental results on synthetic and real images
show that the new algorithm is effective and efficient, and is
relatively independent of this type of noise. Platform: |
Size: 1024 |
Author:李蕾 |
Hits:
Description: 在本文中，我们专注于为添加剂和多折扇状的模型提出了一种简单而新颖的方法为此自动噪声参数估计问题。我们表明，如果图像的工作有一个足够大的量的变异率低的地区（这是一个典型的在大多数图像的特征），噪声的方差（如果添加剂）可作为估计的分布模式在图像局部方差的分布与变化噪声系数（如果乘法）可以估计的变异系数局部估计的分布模式。此外，模型的样本方差分布的图像加噪声的建议和研究。实验表明，所提出的方法的优点，特别是在递归或迭代滤波方法。-In this paper, we focus on the problem of automatic noise parameter estimation for additive and multi-
plicative models and propose a simple and novel method to this end. Specifically we show that if the
image to work with has a sufficiently great amount of low-variability areas (which turns out to be a typ-
ical feature in most images), the variance of noise (if additive) can be estimated as the mode of the dis-
tribution of local variances in the image and the coefficient of variation of noise (if multiplicative) can be
estimated as the mode of the distribution of local estimates of the coefficient of variation. Additionally, a
model for the sample variance distribution for an image plus noise is proposed and studied. Experiments
show the goodness of the proposed method, specially in recursive or iterative filtering methods. Platform: |
Size: 761856 |
Author:陈怀兵 |
Hits:
Description: In the existing bilateral filtering algorithm, the domain parameters and range parameters need to
be predefined. Parameters of a bilateral filter are fixed and cannot guarantee to be optimal. A new adaptive
bilateral filtering (ABF) is proposed in this paper. The ABF obtains the domain parameters by estimating
the local object scale to minimize edge blurring. The range parameters are set adaptively according to noise
variance estimated in smooth areas of a sub image. The method can improve the filtering performance. To
filter out strong noise, the value of domain parameters is increased. ABF avoids setting parameters solely
by experience, and the domain parameters are set adaptively according to the local image features. ABF
can improve the noise filtering ability and reserves edges. Experiments show that the adaptive bilateral filter
is superior to traditional bilateral filters, anisotropic diffusion filters, and modified bilateral filters in both
subjective and objective uations.-In the existing bilateral filtering algorithm, the domain parameters and range parameters need to
be predefined. Parameters of a bilateral filter are fixed and cannot guarantee to be optimal. A new adaptive
bilateral filtering (ABF) is proposed in this paper. The ABF obtains the domain parameters by estimating
the local object scale to minimize edge blurring. The range parameters are set adaptively according to noise
variance estimated in smooth areas of a sub image. The method can improve the filtering performance. To
filter out strong noise, the value of domain parameters is increased. ABF avoids setting parameters solely
by experience, and the domain parameters are set adaptively according to the local image features. ABF
can improve the noise filtering ability and reserves edges. Experiments show that the adaptive bilateral filter
is superior to traditional bilateral filters, anisotropic diffusion filters, and modified bilateral filters in both
subjective and objective ua Platform: |
Size: 4019200 |
Author:杨松 |
Hits:
Description: 正在成像的物体的光谱反射率的重建对于在各种观察光源下再现颜色是重要的。在这项工作中，导出了一个简单的公式来评估一组旨在重建光谱反射率的彩色图像传感器的质量，并将其应用于多光谱图像采集系统。由于质量不仅取决于光谱灵敏度，还取决于系统中存在的噪声，所以不可能在没有其中存在噪声的情况下对一组传感器进行评估。因此，多光谱相机的噪声方差由新方案估算，并首次应用于评估。结果表明，实验结果与评估模型的预测吻合良好，估计噪声方差估计方法对评估是有用的。(The reconstruction of spectral reflectances of objects being imaged is important in reproducing a color under a variety of viewing illuminants. In this work, a simple formula is derived to evaluate the quality of a set of color image sensors aimed at reconstruction of spectral reflectances, and it is applied to multispectral image acquisition systems. Since the quality depends not only on the spectral sensitivities but also on the noise present in the systems, it is impossible to evaluate a set of sensors without prior knowledge of the noise present in it. Therefore, the noise variance of the multispectral cameras is estimated by a new proposal, and it is applied to the evaluation for the first time. It is shown that the experimental results agree well with the predictions by the evaluation model, and that the method to estimate the noise variance is useful for the evaluation.) Platform: |
Size: 324608 |
Author:feiku
|
Hits:
Description: J = imnoise(I,'localvar',IMAGE_INTENSITY,VAR) adds zero-mean, Gaussian
noise to an image, I, where the local variance of the noise is a
function of the image intensity values in I. IMAGE_INTENSITY and VAR
are vectors of the same size, and PLOT(IMAGE_INTENSITY,VAR) plots the
functional relationship between noise variance and image intensity.
IMAGE_INTENSITY must contain normalized intensity values ranging from 0
to 1. Platform: |
Size: 2048 |
Author:Hoang Cuong
|
Hits:
Description: adds zero-mean, Gaussian noise to an image, I, where the local variance of the noise is a
function of the image intensity values in I. IMAGE_INTENSITY and VAR
are vectors of the same size, and PLOT(IMAGE_INTENSITY,VAR) plots the
functional relationship between noise variance and image intensity.
IMAGE_INTENSITY must contain normalized intensity values ranging from 0
to 1. Platform: |
Size: 1024 |
Author:Hoang Cuong
|
Hits:
Description: 主要包括DIP 3/e—Student Projects的说明文档以及实验05-04的详细代码，实验的要求为：
Instruction manual:
(a) Implement a blurring filter as in Eq. (5.6-11).
(b) Blur image 5.26(a) in the +45o direction using T = 1, as in Fig. 5.26(b).
(c) Add Gaussian noise of 0 mean and variance of 10 pixels to the blurred image.
(d) Restore the image using the parametric Wiener filter given in Eq. (5.8-3).(A file of the "DIP 3/e" and code of the experiment 05-04.) Platform: |
Size: 89088 |
Author:爱传奇
|
Hits:
Search - variance of noise in image