Kernel Map

The Kernel Map is a simple operation to analyze the behavior of point pattern and it is an easy to use and understand indicator. Essentially the Kernel map provides the punctual density of the process in the study region through interpolation. As a result, a general view of the process intensity in the study region is obtained.

About the Algorithm

In the algorithm fields, the user selects the function to be used in the calculation, what will be calculated, and the radius to be used. Typical options in Kernel maps, and TerraAmazon defaults, are Quartic Function and Compute Density.

The three Compute options are:

Density: calculates density of crimes per area unit.

Spatial Moving Average: calculates the expected number of crimes per area unit.

Probability: calculates the probability of an already executed crime to be located in a given area.

About Radius

Radius is an important parameter to create a Kernel map and requires the user to understand the study region.

TerraAmazon facilitates the definition task by allowing the creation of Adaptive Kernel map. Adaptive option calculates a radius automatically based on the number of events and the area. Results of maps created with adaptive radius are reasonably good.

However, adaptive option can be unselected and the user can set the radius based on the width of the study area. This value is set using percentage compared to the total width.



It is accessible through:

PROCESSING → SPATIAL ANALYSIS → KERNEL MAP


Input Information:

Layer Name: Select the input layer.

Kernel Parameters:

Attr Name: Attribute name that identifies the desired information.

Function: Quartic, Normal, Triangular, Uniform and Negative Exponential.

Estimation: Density, Spatial Moving Average and Probability.

Use adaptative radius: Uses adaptative radius or defines the percentage value of total width.

Output Information:

Grid: The output is a grid (image) where the kernel map information is a pixel value.

Data Set: The output is the input data plus an attribute with kernel map information.

Repository: Select one type of repository by clicking on the button to save the output layer as a file.

Layer Name: Defines the name to create the output layer.

Click on the OK button and then the kernel map will be calculated.