Hexagon Geospatial
MENU

Spatial Modeler Tutorials

Learn more about our products, find answers, get the latest updates, and connect with other Hexagon Geospatial product users, or get support from our professional service team.
Showing results for 
Search instead for 
Do you mean 

Urban Density Enhancement

by donn.rodekohr on ‎12-04-2015 01:08 PM - edited on ‎04-14-2016 05:13 AM by Community Manager (1,022 Views)

Download model

RoadDensity_model.png

 

Description:

Data Requirements and Assumptions

  1. Road data in line or polyline vector format

  2. Model developed using ERDAS IMAGINE 2015 v15.1

Model Functions

This model was written to reflect the research by Yang, et.al. (see attachment) to detect heavily urbanized areas based upon the density of roads in an area. It is reasonable to use road network information to improve the urban classification accuracy of land use/land cover categories, especially for Urban or Developed classes. Road density is assumed to be a surrogate for actual urban density since we are not examining population or building density. The output is not intended to be used as a land cover classification in its own right. Instead it is intended as input to further analysis to help recode urban classes based on the probable density of human development indicated by the road density. This technique provides a practical solution to the problem of generating accurate and timely land use / land cover inventories.

 
The model requires the operator to provide two file names:

  1. The name of an existing vector data layer containing a polyline file designating roads. This file must be georeferenced.
  2. An output file name of a raster image that will contain a rating for density of roads detected within an area.

Operations

The input vector file is converted to a raster image with a spatial resolution of 15m. Subsequently, the raster file is analyzed for the presence of road values in a 23x23 circular matrix by computing the sum of the road values within the matrix. The frequency of road values are then categorized to a 0 to 9 scale using the following CONDITIONAL statement:

 

Conditional {
($RDS EQ 0) 0,
($RDS LE 10) 1,
($RDS LE 25) 2,
($RDS LE 45) 3,
($RDS LE 60) 4,
($RDS EQ 61) 5,
($RDS LE 85) 6,
($RDS LE 110) 7,
($RDS LE 140) 8,
($RDS GT 140) 9}

where the $RDS value is the number of occurrences of roads in the 23x23 circular matrix window.


Class names are assigned as follows:

Value Class Name  

0

0% Urban  
1 <10% Urban  
2 <60% Urban  
3 <75% Urban  
4 <80% Urban  
5 80% Urban  
6 >90% Urban  
7 >95% Urban  
8 >98% Urban  
9 100% Urban  

 

Class names, colors, and opacity are attached to the raster output.

 

Example Analysis

Input road vector file over a developed suburban area:
RoadDensity_intput.png
Output product of RoadDensity model showing how differing road densities indicate differing urban densities:
RoadDensity_product.png

  

Input parameters:

Roads Vector File: Georeferenced road vector file.

Raster Road Density: User Defined

 

Courses