MapAlgebra
Properties
| Property | Characteristic | Dimension | Description |
|---|---|---|---|
| PixelType | compulsory, user-defined, default: float |
0 | The data type of the input and output image pixels. |
| NumDimensions | compulsory, user-defined, default: 2 |
0 | The number of dimensions of the input and output images. |
| InputTables | optional, user-defined |
2 | absolute path to the image/RAT to be read; uses forward slash / as path separator |
| MapExpressions | compulsory, user-defined |
1 | List of map algebra expressions per iteration. Multiple expressions per iteration are separated by a comma (,). Each expression creates a separate output image. |
| OutputNames | optional, user-defined |
2 | List of short names for individual output images per ieration. The specified names must not contain any colons (:)! The order of the names corresponds with the order of the comma separated MapExpressions. These short names can be used instead of an index number as part of an input image specification for another process component. |
| UseTableColumnCache | optional, user-defined |
0 | Checkbox to indicate wether to cache table columns prior to processing. Note, this option is only effective when SQLite-based attribute tables are processed. |
Supported Image & Pipeline Features
| Feature | Details |
|---|---|
| Image dimensions | 1D, 2D, 3D |
| Multi-band images | no |
| Pipeline role | process |
| Sequential processing | yes |
| Parallel processing | yes |
| Number of inputs | 1..multiple |
| Number of outputs | 1..multiple |
Overview
The MapAlgebra process component performs pixel-wise mathematical and logical operations (s. MapExpressions) on multi-dimensional single-band (scalar) input images. Operations may be performed on the actual image pixel values or on raster attribute table values associated with a given pixel value. The component accepts multiple input images of the same PixelType. MapExpressions refer to different inputs by the input components’ UserID.
Important: Irrespective of the specified PixelType, the MapAlgebra component casts all input pixel values into double before applying the specified algebra expression. The result is then cast into the specified PixelType and written into the output image. Therefore, information loss may occur if the precision and/or scale of PixelType is larger than the precision and/or scale of double.
MapExpressions
LUMASS uses the muparser library for parsing and evaluating mathematical and logical expressions. It supports a large set of mathematical functions and operators (cf. Math Expressions). Image pixel values or associated values in a raster attribute table (RAT) are referred to by the input image’s UserID (e.g. img1) and, if applicable, the RAT’s column name (e.g. biomass) appended to the UserID using a double underscore __, i.e. img1__biomass. Please refer to the Examples section for more details.
Important: The look-up of RAM-based RAT values is based on the assumption that image pixel values represent 0-based row numbers of the RAT. If this assumption is not valid, wrong table values may be returned as a result.
Examples
Input images’ UserIDs |
MapExpression(s) | Explanation |
|---|---|---|
dem |
dem - 0.5 |
subtraction of 0.5 from dem’s pixel values |
rainfall |
rainfall^2 |
raising rainfall’s pixel values to the power of 2 |
dz_dx, dz_dy |
atan(sqrt(dz_dx^2 + dz_dy^2)) * 180 / pi |
calculation of slope angle in degree from slope in x and y direction |
land_use |
land_use__ProductionHa * land_use__AreaHa |
calculation of total production per land use (category) |
dem |
dem < 0 ? 0 : dem |
reclassification: assign all values of dem that are smaller than 0 value 0 and assign all other values their original (i.e. input) value |
slope_deg |
slope_deg >= 0 && slope_deg < 5 ? 1 : slope_deg >= 5 && slope_deg < 15 ? 2 : slope_deg >= 15 ? 3 : 0 |
reclassify the slope (in degree) image into 3 classes 1: 0-5, 2: 5-15, 3: >= 15, all other pixel are assigned the value 0 |
dem |
dem < 0 ? dem : 0, dem >= 0 ? dem : 0 |
splitting the value range of dem into two different output images; output image #1 represents all areas in dem that have below zero elevation and output image #2 represents all areas in dem that are at or above sea level |
Note #1: LUMASS does not support the assignment operator. The result of the ith MapExpression is assigned to the current pixel of the ith ouput image depending on the values of the input image pixels.
Note #2: To save both output images of the last example in the table above, you need to connect an ImageWriter component (double link!) and specify two output names for the ImageWriter component. Please refer to the section ‘Multiple Inputs and Outputs’.
UseTableColumnCache
This option is only effective when SQLite-based attribute tables are processed. SQLite tables are either provided by ImageReader components whose RATType property is set to ATTABLE_TYPE_SQLITE, or by TableReader components who always output SQLite tables. The option is ignored when a RAM-based table is provided, for example by an ImageReader whose RATType is set to ATTABLE_TYPE_RAM.