Differences

This shows you the differences between two versions of the page.

Link to this comparison view

Next revision
Previous revision
eyes:logics:pcatransform [2017/06/12 16:01]
flambre created
eyes:logics:pcatransform [2017/06/12 19:11] (current)
jschlie1
Line 1: Line 1:
-====== ​Template Logic Description ​====== +====== ​PCATransform ​====== 
-THIS position should be used for a brief introduction (max 2 sentences!) ​of the logic, ​stating WHAT and WHY it is doing something. +This logic uses the result ​of the PCA logic (the eigenvectors) to transform images into a new space spanned by these eigenvectors. I.e. Images can be represented as factors of eigenvectorsa so called linear combination
 ===== Usage ===== ===== Usage =====
-Here, a general/​generic description of HOW the logic is USED should be given. Try to be as general ​as possible, but also mention prerequisites,​ restrictions,​ advantages, requirements which are specific ​of this logic. Basically everything ​the user needs to know to successfully use this logic.+Firstly PCA-transform can be used with any set of eigenvectors (as long as the x/y/z- dimensions ​of the PCA-input and the images used in the PCA transform are the same) to yield a set of linear factors. This is called a forward transform. Secondly, these linear factors can be used to generate images by multiplying eigenvectors and linear factors (back-transform).
  
-===== Example ==== 
-Here, a very specific example should be given/​described. In the future, this can be supported by screenshots etc.. For the moment, give an example easy enough for the user to understand, but specific enough to elaborate why a given parameter is a good set for this very situation. 
  
 ===== Modes/​Processes ===== ===== Modes/​Processes =====
-==== thisIsTheNameOfMode1 ​==== +==== Pca forward transform ​==== 
-Here, a short introduction for the given mode should be placedAgain, state WHAT and WHY this mode us useful in not more than 2 sentences. +Here, linear factors are calculated for given set of INPUT images using the provided eigenvectors.
 |< 100% 30% >| |< 100% 30% >|
 ^ Parameters ​                ^ Description ​    ^ ^ Parameters ​                ^ Description ​    ^
-Some changeable parameter ​ ​| ​Description ​of this parameter | +Desired Output ​ ​| ​Lets you choose the kind of output. Linear factor vectors are just 1D plots of the linear factors per eigenvector as used in the back-transformImages with linear factors as headerkeys just append the linear factors to the headerThis is used as input for example the energy landscape calculation |
--> and its sub-parameter ​  | more description | +
-| Next main parameter ​       | and more more more | +
-| -> and its sub-parameter ​  ​| ​... descriptions ​|+
  
-|< 100% 30% >| 
-^ Input   ^ Description ^ 
-| FirstInput ​ | Input Description 1 | 
-| SecondInput | Input Description 2 | 
-| //​ThirdInput// ​ | Input Description 3: Optional Input in Italic | 
  
-|< 100% 30% >| +==== Pca backward transform ==== 
-^ Output ​  ^ Description ^ +Here the linear factor vectors are used as INPUT. The eigenvectors times the linear factors gives you the original data (minus the amount of information you discarded by choosing the number of eigenvectors).  ​
-| FirstOutput | Output Description |+
  
-|< 100% 30% >| 
-^ New/Changed Header Values ^ Description ^ 
-| headerValue1 | what does it say? how is it changed? | 
-| headerValue2 | what does it say? how is it changed? | 
-| headerValue3 | what does it say? how is it changed? | 
-| headerValue4 | what does it say? how is it changed? | 
- 
-==== thisIsTheNameOfMode2 ==== 
-Here, a short introduction for the given mode should be placed. Again, state WHAT and WHY this mode us useful in not more than 2 sentences. 
- 
-|< 100% 30% >| 
-^ Parameters ​                ^ Description ​    ^ 
-| Some changeable parameter ​ | Description of this parameter | 
-| -> and its sub-parameter ​  | more description | 
-| Next main parameter ​       | and more more more | 
-| -> and its sub-parameter ​  | ... descriptions | 
- 
-|< 100% 30% >| 
-^ Input   ^ Description ^ 
-| FirstInput ​ | Input Description 1 | 
-| SecondImput | Input Description 2 | 
-| ThridImput ​ | Input Description 1 | 
- 
-|< 100% 30% >| 
-^ Output ​  ^ Description ^ 
-| FirstOutput | Output Description | 
- 
-|< 100% 30% >| 
-^ New/Changed Header Values ^ Description ^ 
-| headerValue1 | what does it say? how is it changed? | 
-| headerValue2 | what does it say? how is it changed? | 
-| headerValue3 | what does it say? how is it changed? | 
-| headerValue4 | what does it say? how is it changed? | 
  
 ===== Concept ===== ===== Concept =====
-In this paragraph, the "HOW a logic works under the hood" and WHY someone should use it can be elaborated with higher detail. Describes a scenario in an image processing workflow where this logic can be used to solve the resulting problemAlsowikipages, publications or anything else describing the theory behind an algorithm should ​be linked here, if applicable.+Based on the eigenvectors calculated by PCA, this logic calculates linear factors or vice versa calculates ​"raw" images or volumes by using linear factors ​and the eigenvectorsHoweverby simply choosing low numbers of eigenvectors filtering through [[https://​en.wikipedia.org/​wiki/​Dimensionality_reduction|Dimensionality reduction]] can be done