kingtrans1.txt

纯粹是外文,图象去噪处理的外文....可以用来做外文翻译的很不错的素材!

不一致是他们在发展方面负担的柔性简单的。

 Yet powerful 
 仍然有力 

enhancement technique based on statistical measure that has a close 
基於有一个结束的统计尺寸的提高技术 

predictable corresponding with image appearance. 
可预期的用  图像外表符合。 

3.
3.

 Arithmetic/logic operations 
 算术/逻辑操作 

Arithmetic/logic operations involving images are preformed on a 
包括图像的算术/ 逻辑操作被预先形成在一之上 

pixel-by-pixel basis between two or more image (this exclude the logic 
在二或较多的图像之间的图素-被- 图素基础 (这排除逻辑 

operation NOT, which is preformed on a single image), logic operations 
操作不, 在一个单一图像上被预先形成),逻辑操作 

similarly operate on a pixel-by-pixel basis.
同样地在一种图素-被- 图素基础之上操作。

 We need only be concerned 
 我们只需要是关心的 

with the ability to implement the AND, OR and NOT logic operations. 
藉由能力实现及, 或和不逻辑操作。 

3.1
3.1

 Image subtraction 
 图像减少 

The difference between two images f(x, y) and h(x, y) expressed as: 
在二图像 f(x,y) 和 h(x,y) 之间被表示成的不同: 

( , ) ( , ) g x y .
(,)(,) g x y 。

 f x y .h(x, y) [12] 
 f x y.h(x,y)[12] 

is obtained by computing the difference between all pairs of 
被藉由计算在所有的双之间的不同获得 

corresponding pixels from f and h.
来自 f 和 h 的对应图素。

 The key usefulness of subtraction is the 
 减少的主要有用是那 

enhancement of difference between images. 
在图像之间的不同提高。 

3.2
3.2

 Image averaging 
 图像平均 

Consider a noisy image g(x, y) formed by addition of noise .
考虑被噪音的附加形成的吵杂图像 g(x,y) 。

 (x,y) to 
 (x,y)到 

an original image f(x, y); that is: 
最初的图像 f(x,y); 是: 

g(x, y).
g.(x,y)

 f (x, y)..
 f(x,y)..

 (x, y) 
 (x,y) 

[13] 
[13] 

where the assumption is that at every pair of coordinates (x, y) the noise is 
假定在哪里在坐标 (x,y) 的每双噪音是 

uncorrelated and has zero average value.
uncorrelated 而且零的平均有评价吗。

 The objective of the following 
 下列各项的目的 

procedure is to reduce the noise content by adding a set of noising 
程序要藉由增加一组谣传减少噪音内容 

image{ ( , )} i g x y .
图像{(,)} i g x y 。

 If the noise satisfies the constraints just stated, it can 
 如果噪音仅仅使限制满意陈述, 它能 

be shown that if the image ( g x,y) is formed by averaging K different 
如果图像 (g x,y) 被藉由平均 K 不同的形成，被显示那 

noisy images 
吵杂的图像 

1 
1 

1 
1 

( , ) ( , ) 
(,)(,) 

k 
k 

i 
i 

i 
i 

g x y g x y 
g x y g x y 

K . 
K 。 

.
。

 .
 。

 [14] 
 [14] 

Then it following that 
然后它在那之后 

E{g(x, y)}.
E{g(x,y)}.

 f (x, y) [15] 
 f(x,y)[15] 

And 
而且 

2 2 
2 2 

( , ) ( , ) 
(,)(,) 

1 
1 

g x y x y K .
g x y x y K 。

 .
 。

 .
 。

 .
 。

 [16] 
 [16] 

Where E{g(x, y)} is the expected value of g , and 
哪里 E{g(x,y)} 是 g 的预期价值 ,和 

2 
2 

g (x,y) .
g.(x,y)

 and 
 而且 

2 
2 

.
。

 (x,y) .
 (x,y).

 are the variance of g and.
 g 的不一致是吗和。

 .
 。

 All at coordinates (x, y) the standard 
 所有的在坐标 (x,y) 标准 

derivation at any point in the average image as 
引出在平均的图像中的任何点当做 

( , ) ( , ) 
(,)(,) 

1 
1 

g x y x y K 
g x y x y K 

.
。

 .
 。

 .
 。

 .
 。

 [17] 
 [17] 

As K increase equation [16] and [17] indicate that the variability of the 
如 K 增加相等 [16] 和 [17] 指示那易变那 

pixel values at each location (x, y) decrease. 
在每个位置 (x,y) 减少的图素价值。 

Because E{g(x, y)}.
因为 E{g(x,y)}.

 f (x, y) , the mean thatg(x,y) approach f (x, y) 
 f(x,y), 低劣的 thatg(x,y) 接近 f(x,y) 

as the number of noisy images used in the averaging process increases.
如吵杂的图像数字用在那平均处理增加。

 In 
 在 

practice, the image ( , ) i g x y must be registered (aligned) in order to 
练习, 图像 (,) i g x y 一定是注册的 (排列) 在次序中到 

avoid the introduction of blurring and other artifacts in the output image. 
避免使的介绍模糊和在输出图像中其他的人工品。 

4.
4.

 Spatial filtering 
 空间的过滤 

We use the term spatial filtering to differentiate this type of process 
我们使用术语空间的过滤区别程序的这个类型 

from the more traditional frequency domain filtering.
从比较传统的频率领域过滤。

 The processing 
 那处理 

consists simply of moving the filter mask from point to point in an image. 
只是移动来自点的过滤器假面具组成在一个图像中指出。 

At each point (x, y) the response of the filter at that point is calculated 
在点的在每点 (x,y) 过滤器的回应是有计画的 

using a predefined relationship for linear spatial filtering.
使用被预先定义的关系作为线的空间过滤。

 The response is 
 回应是 

given by a sum of products of the filter coefficients and responding image 
被过滤器系数的产品总数给而且回应图像 

pixels in the area spanned by the filter mask.
被过滤器假面具跨越的区域图素。

 In general, linear filtering of 
 大体上, 线的过滤 

an image f of size M N .
大小 M N 的图像 f 。

 with a filter mask of size m n .
 藉由一个过滤器，大小 m n 的假面具。

 is given by the 
 被给被那 

expression: 
表达: 

( , ) ( , ) ( , ) 
(,)(,)(,) 

a b 
a b 

s a t b 
s a t b 

g x y w s t f x s y t 
g x y w s t f x s y t 

..
..

 .. 
 .. 

.
。

 .
 。

 .
 。

 .
 。

 .
 。

 [18] 
 [18] 

Where a= (m-1)/2 and b= (n-1)/2.when interest lies on the response R, of 
哪里一=(m-1)/2 和 b=(n-1)/2. 当兴趣在回应 R 之上躺卧的时候, 

a m n .
a m n 。

 mask at any point (x, y), and not on the mechanics of 
 在任何的点 (x,y) 戴面具, 而且不在技巧之上 

implementing mask convolution, it is common practice to simplify the 
实现假面具回旋, 单一化是常见的做法那 

notation by using the following expression: 
记号法藉由使用下列各项表达: 

1 1 2 2 
1 1 2 2 

1 
1 

... 
... 

mn 
mn 

mn mn i i 
mn mn i i 

i 
i 

R w z w z w z wz 
R w z w z w z wz 

. 
。 

.
。

 .
 。

 .
 。

 .
 。

 .
 。

 .
 。

 [19] 
 [19] 

Where the w is mask coefficients the z’s are the values of an image gray 
w 是假面具系数 z's 哪里是图像灰色的价值 

levels corresponding to those coefficients and mnis the total number of 
水平符合到那些系数和 mnis 总数 

coefficient in the mask. 
在假面具中的系数。 

Nonlinear spatial filter also operate on neighborhoods, and the 
非线性空间的过滤器也在附近之上操作, 和那 

mechanics of sliding a mask past an image are the same as previous 
滑假面具过去一个图像的技巧是相同於早先的 

mentioned.
提到。

 In general, the filtering operation is based conditionally on the 
 大体上，过滤操作被有条件地建立在那之上 

values of the pixels in the neighborhood under consideration.
在考虑之下的在附近里面的图素价值。

 For example, 
 举例来说, 

noise reduction can be achieved effectively with a nonlinear filter whose 
谣传减少能与～一起有效地达成一个非线性过滤器谁的 

basic function is to complete the median gray-level value in the 
基本功能要完成中央的灰色- 水平价值在那 

neighborhood in which the filter is located. 
过滤器位於的附近。 

4.1
4.1

 Smoothing spatial filter 
 抹光空间的过滤器 

Smoothing filters are used for blurring and for noise reduction blurring 
抹光过滤器被用为使和和模糊作为噪音减少模糊 

is used in preprocessing steps.
在 preprocessing 步骤中被用。

 Such as removal of small details from an 
 例如  小的细节移动从一 

image prior to object extraction and bridging of small gaps in lines or 
图像在～之前在线中反对小的缝隙抽出和剪刀撑或 

curves.
曲线。

 Noise reduction can be accomplished by blurring with a linear filter 
 谣传减少藉由用  一个线的过滤器模糊可能是完成的 

and also by nonlinear filtering. 
以及藉着非线性过滤。 

4.1.1
4.1.1

 Smoothing linear filter 
 抹光线的过滤器 

The output of a smoothing linear spatial filter is simply the average of 
抹光线的空间过滤器的输出只是平均 

the pixels contained in neighborhood of the filter mask.
包含在过滤器假面具的附近图素。

 These filter 
 这些过滤 

sometimes are called averaging filter.
有时被叫做平均过滤器。

 They also are referred to a lowpass 
 他们也被提到 lowpass 

filter. 
过滤器。 

The idea behind smoothing filter is straightforward by replacing the 
在抹光过滤器後面的主意藉由更换是笔直的那 

value of every pixel in the neighborhood by the averaging of the gray 
在灰色的平均附近里面的每个图素的价值 

levels n the neighborhood defined by the filter mask.
消除 n 被过滤器假面具定义的附近。

 This process results 
 这个程序产生 

in an image with reduced “sharp” transitions in gray levels.
在一个图像中用  减少的 "高调" 转变在灰色的水平中。

 Because noise 
 因为噪音 

random noise typically consists of sharp transitions in gray levels, the most 
在灰色的水平中随意噪音典型地由～所组成锐利的转变,大部分 

obvious application of smoothing is noise reduction.
抹光的明显申请是噪音减少。

 An important 
 一重要的 

application of spatial filtering is to blur an image for the purpose getting a 
空间的过滤申请要为目的使一个图像模糊得到一 

gross representation of objects of interest.
兴趣的物体总共的表现。

 Such that the intensity of 
 以致于强烈 

smaller objects blend with the background and larger objects become 
比较小的物体由于背景混合，而且比较大的物体变成 

“blodlike” and easy to detect.
"blodlike" 和容易的发现。

 The size of the mask establishes the relative 
 假面具的大小建立亲戚 

size of the objects that will be blended with background. 
将会与～一起混合背景的物体大小。 

4.1.2
4.1.2

 Order-statistics filter 
 次序- 统计学过滤器 

Order-statistics filter are nonlinear spatial filters whose response is 
次序- 统计学过滤器是非线性空间的过滤器谁的回应是 

based on ordering(ranking) the pixels contained in the image area 
基於命令 (排名) 被包含在图像区域的图素 

encompass by the filter, and then replacing the value of the center pixel 
藉着过滤器包含, 然後更换中央的图素价值 

with the value determined by the ranking result.
藉由被上级的结果决定的价值。

 The best-known example 
 流传久远的例子 

in this category is the median filter, which, as its name implies, replacing 
在这个种类中是中央的过滤器,,如它的名字暗示,更换 

the value of a pixel by the median of the gray levels in the neighborhood 
在附近里面的灰色水平的中动脉的图素价值 

of that pixel.
那个图素。

 Median filter are quite popular because, for certain types of 
 中动脉过滤相当流行因为, 为某类型的 

random noise.
随意噪音。

 They provide excellent noise-reduction capabilities with 
 他们用～提供优良的噪音- 减少能力 

considerably less blurring than linear smoothing filter of similar size. 
非常地比较少量模糊比较相似的大小线抹光过滤器。 

Median filter are particularly effective in the presence of impulse noise, 
中动脉过滤是特别地有效的在～之前冲动噪音, 

also called salt-and-pepper noise because of its appearance as white and 
也呼叫了盐-和- 胡椒粉因为它的外表如白色的谣传和 

black dots superimposed on an image.
黑色的点在一个图像上重叠。

 In order to perform median 
 为了要运行中动脉 

filtering at a point in an image, we first sort the values of the pixel in 
过滤在一个图像中的点，我们首先分类图素的价值在 

question and its neighbors determine their median and assign this value to 
问题和它的邻居决定他们的中央而且分配这价值到 

that pixel.
那个图素。

 For example in a 3x3 neighborhood the median is the 5th largest 
 举例来说在 3x3 neighborhood 中中动脉是第 5 个最大的 

values. 
价值。 

Although the median filter is by far the most useful order-statistics 
虽然中央的过滤器显然是最有用的次序-统计学 

filter in image processing.
图像处理的过滤器。

 It is by no means the only one.
 它绝不是一个唯一的。

 The median 
 中动脉 

represents the 50th percentile of a ranked set of numbers. 
表现数字的被排列的组合第 50 percentile 。 

4.2
4.2

 Sharpening spatial filter 
 使空间的过滤器尖锐 

The principal objective of sharpening is to highlight fine detail in an 
使的主要目的尖锐是对加亮区罚款细节在一 

image or to enhancement detail that has been blurred either in error or as 
图像或对有被模糊的提高细节或在错误中或当做 

a natural effect of a particular method acquisition.
特别的方法获得的天然效果。

 In the last section, we 
 在最後一个区段中,我们 

saw that image blurring could be accomplished in the spatial domain by 
看见那个图像模糊在空间的领域中可能是完成的被 

pixel averaging in a neighborhood.
图素在附近中平均。

 Since averaging is analogous to 
 自平均以后是类似的到 

integration, it is logic to conclude that sharpening could be accomplished 
整合, 它是逻辑得出结论使可能是完成的尖锐 

by spatial differentiation.
藉着空间的区别。

 Image differentiation enhances edges and other 
 图像区别提高边缘和其他人 

discontinuities (such as noise) and deemphasizes areas with sharpening 
断绝 (例如  噪音) 和 deemphasizes 由于尖锐 

filters that are based on first-
以～为基础第一的过滤器-

 and second-order derivatives. 
 而且秒- 次序引出之物。