# CS代写|图像处理作业代写Image Processing代考|EEE6512

## CS代写|图像处理作业代写Image Processing代考|Blur Kernel Estimation

According to the image degradation model, to restore the blurred image (de-blurring), it is necessary to determine the blur function, that is, estimate the blur kernel. In practice, the blur function is often difficult to be completely determined from the image, but it can be estimated with the help of some prior knowledge. In the case that the blur function cannot be obtained directly, performing image restoration to eliminate blur is also called blind de-convolution.
The estimation methods for blur functions can be divided into three categories:

1. Estimation with the help of image observation.
2. Estimation by using point source image experiment.
3. Estimation with the help of modeling degradation.
3.1.3.1 Using the Image Observation to Estimate the Blur Function
Consider the case where the image is affected by linear space invariant degradation. If only a degraded image $g(x, y)$ is given without any knowledge about the image degradation function, only the information contained in this image can be used to estimate the degradation function.

When the degradation is caused by the blur process, a (sub) region with a typical structure in the image can be selected. To reduce the influence of noise, the region should preferably contain obvious edges or high-contrast borders between the object and the background. If the grayscale contrast between the object and the background is $C_{o b}$, and the mean square error of the noise is $\sigma$, then the signal-to-noise ratio can be defined as (Kitchen and Rosenfeld 1981).
$$\mathrm{SNR}=\left(\frac{C_{o b}}{\sigma}\right)^2$$
Here, it is required to choose a region with a large signal-to-noise ratio.
Suppose the region in the selected blurred image is $g_s(x, y), g_s(x, y)$ needs to be processed to obtain $f_s(x, y)$, where $f_s(x, y)$ is the estimation of the original image $f(x, y)$ at the position corresponding to $g_s(x, y)$.

## CS代写|图像处理作业代写Image Processing代考|IMAGE RESTORATION AND DE-BLURRING

Research on de-blurring has a history of many years, and people have proposed a variety of classic methods, which have been widely used in practice. The following paragraph briefly introduces the methods based on image restoration technology to eliminate blur.

Image restoration is a large category of technology in image processing. Image restoration is closely related to image enhancement. The similarity between image restoration and image enhancement is that they can both improve the visual quality of the input image. The difference between them is that image enhancement technology generally only uses the characteristics of the human visual system to obtain good-looking visual results, while image restoration considers that the image (quality) is degraded or deteriorated under certain circumstances/conditions (i.e., the image quality has been reduced and distorted), and now it is necessary to reconstruct or restore the original image based on the corresponding degradation model and knowledge. In other words, the image restoration technology is to model the image degradation process and restore it according to the determined image degradation model to obtain the original desired effect.

Under the conditions of a given model, image restoration techniques can be divided into two categories: unconstrained and constrained. The method of unconstrained restoration only regards the image as a digital matrix, without considering the physical constraints that the image should be subjected to after restoration, and mainly deals with it from a mathematical point of view. The constrained restoration method also considers that the restored image should be subject to certain physical constraints, such as being relatively smooth in space and the image gray value is always positive.

By the way, although noise is random, it often has certain statistical laws. If a certain model of noise can be established, or the process of image degradation affected by noise can be modeled, then image restoration technology can also be used to denoise the image based on the noise degradation model.

Based on the basic image degradation model established in the previous section, the following summarizes three typical methods, namely, inverse filter restoration, Wiener filter restoration, and constrained least squares restoration (Zhang 2017). In addition, it also introduces, with examples, how to use human-computer interaction methods to improve the flexibility and efficiency of image restoration.

# 图像处理代考

## CS代写|图像处理作业代写Image Processing代考|Blur Kernel Estimation

1. 借助图像观察进行估计。
2. 利用点源图像实验进行估计。
3. 在建模退化的帮助下进行估计。
3.1.3.1 使用图像观察估计模糊函数
考虑图像受到线性空间不变退化影响的情况。如果只有退化的图像G(X,是)在没有任何关于图像退化函数的知识的情况下给出，只有该图像中包含的信息可以用来估计退化函数。

## CS代写|图像处理作业代写Image Processing代考|IMAGE RESTORATION AND DE-BLURRING

myassignments-help数学代考价格说明

1、客户需提供物理代考的网址，相关账户，以及课程名称，Textbook等相关资料~客服会根据作业数量和持续时间给您定价~使收费透明，让您清楚的知道您的钱花在什么地方。

2、数学代写一般每篇报价约为600—1000rmb，费用根据持续时间、周作业量、成绩要求有所浮动(持续时间越长约便宜、周作业量越多约贵、成绩要求越高越贵)，报价后价格觉得合适，可以先付一周的款，我们帮你试做，满意后再继续，遇到Fail全额退款。

3、myassignments-help公司所有MATH作业代写服务支持付半款，全款，周付款，周付款一方面方便大家查阅自己的分数，一方面也方便大家资金周转，注意:每周固定周一时先预付下周的定金，不付定金不予继续做。物理代写一次性付清打9.5折。

Math作业代写、数学代写常见问题

myassignments-help擅长领域包含但不是全部: