Differences

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

--- eyes:logics:angularreconstitution [2017/06/08 15:32]
vzinch [Template Logic Description]
+++ eyes:logics:angularreconstitution [2017/06/12 15:29]
vzinch [Concept]
@@ Line 1: / Line 1: @@
 ====== AngularReconstitution ======
-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 determines Euler angles for a given set of input images based on the common line theorem. This is needed in order to correctly place images (2D projections) in 3D space prior to reconstruction.
 ===== 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.
+The input of this logic - class averages - should be of high quality with enough visible structural features. Choosing more class averages for this logic is not always a good idea, since similar views will result in worse angular assignment. So ideally class averages chosen should represent all possible distinct views, but without overlapping. If symmetry of the particle is known in advance, it is recommended to try running reconstitution both with and without applying symmetry restraints.
-===== Example ====
+===== Modes =====
-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.
+==== Euler ====
+This mode is used for de-novo assignment of angles only from the input projections themselves
-===== Modes/Processes =====
-==== thisIsTheNameOfMode1 ====
-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 |
+| Sinograms  | Gives options to read already generated sinograms or to write down sinograms for the images being processed |
-| -> and its sub-parameter   | more description |
+| Symmetry  | Symmetry of the particle (if known) |
-| Next main parameter        | and more more more |
-| -> and its sub-parameter   | ... descriptions |
 |< 100% 30% >|
 ^ Input   ^ Description ^
-| FirstInput  | Input Description 1 |
+| In  | Image or stack of images (a set of class averages)|
-| SecondInput | Input Description 2 |
-| //ThirdInput//  | Input Description 3: Optional Input in Italic |
 |< 100% 30% >|
 ^ Output   ^ Description ^
-| FirstOutput | Output Description |
+| Out | The input projections with assigned Euler angles |
+==== Anchor set ====
+This logic determines the angles of stack of projections if the set with known angular relations is available (e.g. 3D model from already performed angular reconstitution)
 |< 100% 30% >|
-^ New/Changed Header Values ^ Description ^
+^ Parameters                 ^ Description     ^
-| headerValue1 | what does it say? how is it changed? |
+| Symmetry  | Symmetry of the particle (if known) |
-| headerValue2 | what does it say? how is it changed? |
-| headerValue3 | what does it say? how is it changed? |
+|< 100% 30% >|
-| headerValue4 | what does it say? how is it changed? |
+^ Input   ^ Description ^
+| In  | Image or stack of images (a set of class averages)|
+| Anchor set | Projection set with known angular relations (mostly projected 3D-model) |
+|< 100% 30% >|
+^ Output   ^ Description ^
+| Out | The input projections with assigned Euler angles (no anchor dataset in the output) |
-==== thisIsTheNameOfMode2 ====
+==== Add projection ====
-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.
+This mode is used to adds projections to a set with known/determined Euler angles
 |< 100% 30% >|
 ^ Parameters                 ^ Description     ^
-| Some changeable parameter  | Description of this parameter |
+| Symmetry  | Symmetry of the particle (if known) |
-| -> 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 |
+| In  | Image or stack of images (a set of class averages)|
-| SecondImput | Input Description 2 |
+| Dataset | Stack of images (projections) with assigned Euler angles |
-| ThridImput  | Input Description 1 |
 |< 100% 30% >|
 ^ Output   ^ Description ^
-| FirstOutput | Output Description |
+| Out | The input dataset with added input projections with assigned Euler angles |
+==== Sinograms ====
+This mode is used to create sinograms from the images. These sinograms can be later used in the same logic to reconstitute the angles.
 |< 100% 30% >|
-^ New/Changed Header Values ^ Description ^
+^ Parameters                 ^ Description     ^
-| headerValue1 | what does it say? how is it changed? |
+| Symmetry  | Symmetry of the particle (if known) |
-| 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? |
+|< 100% 30% >|
+^ Input   ^ Description ^
+| In  | Image or stack of images |
+|< 100% 30% >|
+^ Output   ^ Description ^
+| Out | Sinogram |
 ===== 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 problem. Also, wikipages, publications or anything else describing the theory behind an algorithm should be linked here, if applicable.
+This logic is needed as an intermediate step to get from 2D images to a 3D volume. In order to reconstruct the full volume we need to know the relationship between the images (projections) we have - how they are located in 3D relative to each other.
+In case any reference model for the particle is available, it can be used as 'anchor set' that would increase the precision of the output.
+However, when no prior model is available for the particles of interest, determination of [[https://en.wikipedia.org/wiki/Euler_angles|Euler angles]] should be done de novo. This logic does it based on [[http://www.sciencedirect.com/science/article/pii/0304399187900787|Common line theorem]].