Difference between revisions of "General Image Processing"
(Created page with "'''Genaral Image Processing'''では,一般的なImage Processing方法の基礎について,Eosを使ってExample示しながら解説します. == basis of image proces...") |
(→Image Averaging) |
||
| (17 intermediate revisions by the same user not shown) | |||
| Line 1: | Line 1: | ||
| − | '''Genaral Image Processing''' | + | On '''Genaral Image Processing''', describe about general foundation of image processing with some examples by using Eos. |
| − | == [[ | + | == [[Foundation of image processing]]と[[Simple Image Processing]] == |
| − | + | Tutorial on simple image processing using the Eos has been published. | |
| − | == [[Input of image]] | + | == [[Input of image]] and [[Lens]] == |
| − | + | Before performing image processing using a computer, you should keep in mind about the input device that digitize images.Then, it is important to understand optical element called lens which creates image by condensing for light occurred at object. | |
| − | == [[CTF]],[[PSF]] | + | == [[CTF]],[[PSF]] and [[MTF]]== |
| − | + | True image: f(x,y) is degraded due to the method of [[Input of image]] and performance of the [[Lens]].こThen, It might occur uniform blur(degradation) at whole field. Here, the function about degradation of original point image is called [[PSF]](Point Spread Function). If this Point Spread Function PSF(x, y) is determined, it can be regarded that an observed image g(x, y) is convoluted PSF(x, y) to a true image f(x, y). | |
== [[Sampling]] == | == [[Sampling]] == | ||
| − | + | To convert analog images to digital images, it is important that discretize the space by decomposing. This step is called Sampling. This failure might make incorrect interpretation like false image. Therefore, you must process carefully. | |
== [[Quantization]] == | == [[Quantization]] == | ||
| − | + | First, in digital image process the operation that expresses density values within certain number of bits (Optical density) is required. This operation is called quantization(AD convert). Information lost by this process can't be regained. | |
== [[Noise reduction]] == | == [[Noise reduction]] == | ||
| − | + | In order to deal the image including a lot of noise, knowing well the nature of the noise is important. | |
| − | + | Show as following some factors about noise related to electron microscope. | |
| − | # | + | # Quantum noise by lack of electron dose: White Noise(Whole field) |
| − | # | + | # Noise which exists many in the low-resolution side by energy lack of electron or chromatic aberration: Colored Noise(Whole field) |
| − | # | + | # Noise by radiation or dust on CCD, fluorescent screen, or film of camera: Local Noise |
| − | + | and so on. | |
| − | + | Quantum Noise is one kind of White noise, and it is known that this is often noise distribution depending on Poisson Process. | |
| − | + | In the low-resolution side There is noise by Inelastic scattered electrons or chromatic aberration, with blur. Therefor, this noise is Colored. | |
| − | + | Noise by radiation or dust on CCD, fluorescent screen, or film of camera is one example of Local Noise. Factor by cosmic rays or radiation by collapse of the fluorescent screen gives pixels which have considerable high contrast. Because of [[MTF]](e.g. by CCD), it is not one point, but occurs blur. | |
=== [[Smoothing]] === | === [[Smoothing]] === | ||
| − | + | This is an image processing method with the primary purpose of eliminating the noise in the image. By understanding well the nature of the noise, the noise can be reduced properly. | |
== [[Edge extraction]] == | == [[Edge extraction]] == | ||
| − | + | This is an important step in order to understand the shape of the object, but it is also a very difficult step. | |
== [[Binarization]] == | == [[Binarization]] == | ||
| − | + | This is the process of isolating the background and the signal, and extracting the representative point and the skeletal. And it is the start process of the analysis and image processing. | |
| − | == [[Fourier Space]] | + | == Image processing using the [[Fourier Space]] == |
| − | + | It is an useful method for an image which has repetition period. | |
| − | == | + | == Image processing using the [[Kernel]] of the Real space == |
| − | == [[Mathematical morphology]] | + | == Image processing using the [[Mathematical morphology]] == |
== [[Image Averaging]] == | == [[Image Averaging]] == | ||
| − | + | If many images about the same photo field or the same particle exist, image averaging can raise the quality of the images. | |
Latest revision as of 02:03, 2 October 2014
On Genaral Image Processing, describe about general foundation of image processing with some examples by using Eos.
Contents
- 1 Foundation of image processingとSimple Image Processing
- 2 Input of image and Lens
- 3 CTF,PSF and MTF
- 4 Sampling
- 5 Quantization
- 6 Noise reduction
- 7 Edge extraction
- 8 Binarization
- 9 Image processing using the Fourier Space
- 10 Image processing using the Kernel of the Real space
- 11 Image processing using the Mathematical morphology
- 12 Image Averaging
Foundation of image processingとSimple Image Processing
Tutorial on simple image processing using the Eos has been published.
Input of image and Lens
Before performing image processing using a computer, you should keep in mind about the input device that digitize images.Then, it is important to understand optical element called lens which creates image by condensing for light occurred at object.
CTF,PSF and MTF
True image: f(x,y) is degraded due to the method of Input of image and performance of the Lens.こThen, It might occur uniform blur(degradation) at whole field. Here, the function about degradation of original point image is called PSF(Point Spread Function). If this Point Spread Function PSF(x, y) is determined, it can be regarded that an observed image g(x, y) is convoluted PSF(x, y) to a true image f(x, y).
Sampling
To convert analog images to digital images, it is important that discretize the space by decomposing. This step is called Sampling. This failure might make incorrect interpretation like false image. Therefore, you must process carefully.
Quantization
First, in digital image process the operation that expresses density values within certain number of bits (Optical density) is required. This operation is called quantization(AD convert). Information lost by this process can't be regained.
Noise reduction
In order to deal the image including a lot of noise, knowing well the nature of the noise is important. Show as following some factors about noise related to electron microscope.
- Quantum noise by lack of electron dose: White Noise(Whole field)
- Noise which exists many in the low-resolution side by energy lack of electron or chromatic aberration: Colored Noise(Whole field)
- Noise by radiation or dust on CCD, fluorescent screen, or film of camera: Local Noise
and so on.
Quantum Noise is one kind of White noise, and it is known that this is often noise distribution depending on Poisson Process.
In the low-resolution side There is noise by Inelastic scattered electrons or chromatic aberration, with blur. Therefor, this noise is Colored.
Noise by radiation or dust on CCD, fluorescent screen, or film of camera is one example of Local Noise. Factor by cosmic rays or radiation by collapse of the fluorescent screen gives pixels which have considerable high contrast. Because of MTF(e.g. by CCD), it is not one point, but occurs blur.
Smoothing
This is an image processing method with the primary purpose of eliminating the noise in the image. By understanding well the nature of the noise, the noise can be reduced properly.
Edge extraction
This is an important step in order to understand the shape of the object, but it is also a very difficult step.
Binarization
This is the process of isolating the background and the signal, and extracting the representative point and the skeletal. And it is the start process of the analysis and image processing.
Image processing using the Fourier Space
It is an useful method for an image which has repetition period.
Image processing using the Kernel of the Real space
Image processing using the Mathematical morphology
Image Averaging
If many images about the same photo field or the same particle exist, image averaging can raise the quality of the images.
