Divisão de Observação da Terra e Geoinformática

Essa página está em modo somente de leitura. Você pode visualizar a fonte, mas não alterá-la. Informe-se com o administrador do Wiki, caso você ache que isso está incorreto.
[[interimage:|InterIMAGE wiki]] 

====== BREC Operators Documentation ======
Operators reference version: 1.0.0

InterIMAGE is a multi-platform framework, written in C++, currently with implementations for LINUX and Windows operational systems. Moreover, InterIMAGE provides support for the integration of image processing operators in the interpretation process and, as such operators are treated as external programs by its control mechanism, they can be coded in any computer language, and even in proprietary programs. The InterIMAGE framework offers, nonetheless, a repository of operators, called **TerraAIDA** ([[http://www.dpi.inpe.br/terraaida]]), assembled with software classes and functions supplied by the [[http://www.terralib.org/|TerraLib]] library and another repository, called **BREC**, with routines supplied by the [[http://tolomeofp7.unipv.it/SoftwareTools/BREC/|BREC]] library.









===== BREC Urban Focus =====

**Authors**

Gianni Lisini, [[gianni.lisini@unipv.it]]\\
Gianni Cristian Iannelli, [[giannicristian.iannelli@unipv.it]]\\
Rodrigo da Silva Ferreira, [[rsilva@ele.puc-rio.br]]

**Description**  

This operator extracts urban area polygons based on a built-up presence index and on the given threshold.

**Type**

Top-Down operator.

**Applies to**

Any node.

**Multi-Class support**

No.

**Processing sequence**

  -Read input Image
  -Compute the built-up presence index
  -Apply thresholding
  -Generate the output hypothesis

**Input parameters**
 
^ Parameter Name  ^ Type            ^ Description               ^   Valid Values      ^   Note   ^   
| Input Image       | TIFF | Input image.   | Valid image file name.  | The image type must be supported by TerraLib. |
| Low-Pass Filter Kernel   | Real Number   | Filter Kernel (in meters). | Positive real numbers greater than three times the resolution. | Average moving kernel. Must be an exact multiple of the resolution. If not, it will be automatically rounded. |
| Threshold     | Real Number | Threshold applied to the Build-Up presence Index. | Positive real numbers. |  |
| Label Image Resolution     | Real Number | The output label image resolution. | Positive real numbers. |  |
| Reliability     | Real Number | The reliability (higher priority will be given to nodes with higher weights in cases where there are geographic overlays).  | [0.0, 1.0] |  |









===== BREC Building Regularizer =====

**Authors** 

Gianni Lisini, [[gianni.lisini@unipv.it]]\\
Rodrigo da Silva Ferreira, [[rsilva@ele.puc-rio.br]]

**Description**  

Building Regularizer is a BREC routine that regularizes the shapes of the buildings after classification. Using 
morphological operations the operator fills some holes and smoothes out some branches. 

**Type**

Bottom-Up operator.

**Applies to**

Any node with children nodes.

**Multi-Class support**

No.

**Processing sequence**

  -Read input Image
  -Polygon raster-based shape regularization
  -Generate the output hypothesis

**Input parameters**
 
^ Parameter Name  ^ Type            ^ Description               ^   Valid Values      ^   Note   ^   
| Detail Threshold       | Real Number  | A threshold given in squared meters. | Positive real numbers.  | Must be an exact multiple of the squared resolution. If not, it will be automatically rounded. |
| Building Classes   | String | Comma-separated list of the building class names. | The classes must be children nodes of the node the BU operator will be run on.  |  |
| Reliability     | Real Number | The reliability (higher priority will be given to nodes with higher weights in cases where there are geographic overlays).  | [0.0, 1.0] |  |



===== BREC Shape Regularizer =====

**Authors** 

Gianni Lisini, [[gianni.lisini@unipv.it]]\\
Rodrigo da Silva Ferreira, [[rsilva@ele.puc-rio.br]]

**Description**  

This operator is in many ways similar to the previous one. The main difference is that instead of working with 
morphological operations to “regularize” the shape of the segments, it simplifies the corresponding polygon. So, in this 
sense, the previous one works in the raster domain and this one in the vector domain.  

**Type**

Bottom-Up operator.

**Applies to**

Any node with children nodes.

**Multi-Class support**

No.

**Processing sequence**

  -Read input image
  -Polygon vector-based shape regularization
  -Generate the output hypothesis

**Input parameters**
 
^ Parameter Name  ^ Type            ^ Description               ^   Valid Values      ^   Note   ^   
| Distance       | Real Number | Distance in meters. | Positive real numbers. | Must be an exact multiple of the resolution. If not, it will be automatically rounded. |
| Tolerance       | Integer |  |  |  |
| Building Classes   | String | Comma-separated list of the building class names. | The classes must be children nodes of the node the BU operator will be run on.  |  |
| Reliability     | Real Number | The reliability (higher priority will be given to nodes with higher weights in cases where there are geographic overlays).  | [0.0, 1.0] |  |















===== BREC Building Shadow Raiser =====

**Authors** 

Gianni Lisini, [[gianni.lisini@unipv.it]]\\
Rodrigo da Silva Ferreira, [[rsilva@ele.puc-rio.br]]

**Description**  

Based on date and time of image acquisition, and given the classification of shadow and building classes, this operator calculates the mean height for each polygon.

**Type**

Bottom-Up operator.

**Applies to**

Any node.

**Multi-Class support**

No.

**Processing sequence**

  -Read input Image
  -Height image computation
  -Mean height computation for each polygon
  -Generate the output hypothesis

**Input parameters**
 
^ Parameter Name  ^ Type            ^ Description               ^   Valid Values      ^   Note   ^   
| Latitude       | Real Number |  |  |  |
| Longitude       | Real Number  |  |  |  |
| Day       | Integer  | Day of image acquisition.   | [1,31] |  |
| Month       | Integer  | Month of image acquisition.   | [1,12] |  |
| Year       | Integer  | Year of image acquisition.  | 4-digit years e.g. 2009. |  |
| Hour       | Integer  | Hour of image acquisition.   | [0,23] |  |
| Label Image Resolution | Real Number | The output label image resolution.  | Positive real numbers or zero for automatic.  | Choose a value good enough matching the resolution of the best used image. Leave blank and the system will automatically choose the resolution. The automatic method can be used only if the operator is not in the semantic network node right below the “scene” node.  |
| Max. Building-Shadow Gap | Real Number | Maximum distance between building and shadow (in meters).   | Positive real numbers. | Must be an exact multiple of the resolution. If not, it will be automatically rounded.  |
| Max. Shadow Length       | Real Number | Maximum shadow length (in meters).   | Positive real numbers. | Must be an exact multiple of the resolution. If not, it will be automatically rounded.  |
| Min. Percentage Shadow Sample      | Integer  | Minimum building edge (percentage) generating shadow.   | Positive real numbers. |  |
| Max. Percentage Deviation Sample      | Integer  | Mean square error for the building heights.   | [10,30]  |  |
| Shadow Class       | String  | The shadow class name.   | The class must be a child node of the node the BU operator will be run on.  |  |
| Building Classes   | String | Comma-separated list of the building class names. | The classes must be children nodes of the node the BU operator will be run on.  |  |
| Attribute Name  | String | The name of the attribute that will hold the mean height value of each polygon an will be available in the corresponding BU decision rule. | The name must not match the default attribute names.  |  |
| Reliability     | Real Number | The reliability (higher priority will be given to nodes with higher weights in cases where there are geographic overlays).  | [0.0, 1.0] |  |









===== BREC Road Extraction =====

**Authors** 

Gianni Lisini, [[gianni.lisini@unipv.it]]\\
Rodolfo Georjute Lotte, [[rodolfo.lotte@inpe.br]]\\
Rodrigo da Silva Ferreira, [[rsilva@ele.puc-rio.br]]

**Description**  

Enables extraction of road network in urban and non-urban areas on high-resolution synthetic aperture radar (SAR) images.

**Type**

Top-Down operator.

**Applies to**

Any node.

**Multi-Class support**

No.

**Processing sequence**

  -Read input Image
  -Apply the Multi-scale feature fusion detector, composed by: (a) Multiple feature extraction, (b) Feature binarization, (c) Multi-scale fusion and (d) Candidate area selection.
  -Apply the Segment extraction, where the results obtained in the previous stage are submitted to the (a) Shape regularization and then (b) Best fitting segment extraction.
  -Generate the output hypothesis

**Input parameters**
 
^ Parameter Name  ^ Type            ^ Description               ^   Valid Values      ^   Note   ^   
| Input Image      | TIFF | Input image.   | Valid image file name.  | The image type must be supported by TerraLib. |
| Minimun Radius   | Real Number | Minimun size (in meters) of the radius of the circle around to scanning the radiance. | [0, 50*res] | Must be an exact multiple of the resolution. If not, it will be automatically rounded.  |
| Maximum Radius   | Real Number | Maximum size (in meters) of the radius of the circle around to scanning the radiance. | [5*res, x] | Must be an exact multiple of the resolution. If not, it will be automatically rounded.  |
| Step Radius   | Real Number | Step increment (in meters) of the radius of the circle around to scanning the radiance. | [1*res, x]  | Must be an exact multiple of the resolution. If not, it will be automatically rounded.  |
| Reliability     | Real Number | The reliability (higher priority will be given to nodes with higher weights in cases where there are geographic overlays).  | [0.0, 1.0] |  |










===== BREC Cross Detection =====

**Authors** 

Gianni Lisini, [[gianni.lisini@unipv.it]]\\
Rodolfo Georjute Lotte, [[rodolfo.lotte@inpe.br]]\\
Rodrigo da Silva Ferreira, [[rsilva@ele.puc-rio.br]]

**Description**  

Allows extracting road junctions based on their type.

**Type**

Top-Down operator.

**Applies to**

Any node.

**Multi-Class support**

No.

**Processing sequence**

  -Read input Image
  -Apply the Multi-scale feature fusion detector, composed by: (a) Multiple feature extraction, (b) Feature binarization, (c) Multi-scale fusion and (d) Candidate area selection.
  -Apply the Segment extraction, where the results obtained in the previous stage are submitted to the (a) Shape regularization and then (b) Best fitting segment extraction.
  -Apply the crossing detection algorithm.
  -Generate the output hypothesis

**Input parameters**
 
^ Parameter Name  ^ Type            ^ Description               ^   Valid Values      ^   Note   ^   
| Input Image      | TIFF | Input image.   | Valid image file name.  | The image type must be supported by TerraLib. |
| Minimun Radius   | Real Number | Minimun size (in meters) of the radius of the circle around to scanning the radiance. | [0, 50*res] | Must be an exact multiple of the resolution. If not, it will be automatically rounded.  |
| Maximum Radius   | Real Number | Maximum size (in meters) of the radius of the circle around to scanning the radiance. | [5*res, x] | Must be an exact multiple of the resolution. If not, it will be automatically rounded.  |
| Step Radius   | Real Number | Step increment (in meters) of the radius of the circle around to scanning the radiance. | [1*res, x]  | Must be an exact multiple of the resolution. If not, it will be automatically rounded.  |
| Junction Tipology   | Enumeration | The junction type. | [L, T, X, Y, M, all]  | M stands for multiple-line crossings. |
| Angle Threshold   | Integer | Sets a minimum angle between two streets for them to be differentiated and not to be considered as a single street bending in the hypothesised junction point. | Positive number.  |  |
| Max. Stem   | Real Number | The maximum distance between the intersection point and the extreme of one street crossing the junction. | Positive real number.  | Must be an exact multiple of the resolution. If not, it will be automatically rounded. |
| Max. Gap   | Real Number | It is the maximum distance between the extreme of one road (external, non-crossing) and the junction. | Positive real number.  | Must be an exact multiple of the resolution. If not, it will be automatically rounded. |
| Min. Junction Distance   | Real Number | Minimum distance between two junctions for them to be differentiated and not considered as a single junction. | Positive real number.  | Must be an exact multiple of the resolution. If not, it will be automatically rounded. |
| Angle Tolerance   | Integer | Minimum angle between non-coincident streets.  | Positive number.  |  |
| Max. Number of roads   | Integer | Maximum number of roads springing from a junction.  | Positive number.  |  |
| Road Network shapefile   | *.shp | Shapefile with a road network.  |  | If not given, the operator will extract the roads automatically before computing the crossings. |
| Reliability     | Real Number | The reliability (higher priority will be given to nodes with higher weights in cases where there are geographic overlays).  | [0.0, 1.0] |  |