GIS/GPS Chapter 10

Raster distance measurement operations

distance of buffering rasters is based on Pythagorean formula, based off on centroid of each pixel using the measured x and y distance

Buffering using raster data

create 30 m polygons within a raster, very pixelated due to being raster data, raster cannot tell if in the middle/side of a pixel, buffering point features watch out for the overlap

Local raster operations

each pixel (cell) at location i, j in a raster dataset may be considered a local object, value of said pixels can be operated upon independent of neighboring pixels (cells), force every pixel to become an integer

Reclassification/recoding using a single raster dataset

sometimes it’s necessary to reclassify/recode the individual pixel values in a single raster dataset

Raster reduction and magnification

analysts routinely maps/images that have been reduced in size/magnified during the map/image interpretation process, reduction techniques allow analysts to zoom out and obtain a regional perspective of the map/remotely sense data, raster map/image magnification allows the analyst to zoom in and view very sit-specific pixel characteristics

Local operations applied to multiple registered raster datasets

raster map overlay occurs when local operations are applied to multiple raster layers, map algebra

Map algebra

when multiple raster layers are processed using arithmetic and algebraic operations

Example of map algebra

min/max, range, sum, mean, median, mode, standard deviation

Sea level rise prediction

freshwater entering a saltwater system (estuary), taken by satellite, zoomed in to far and fix by clipping and legend will update, helpful for coastal changes within communities

Neighborhood raster oeprations

neighborhood raster operations modify the value of each focal pixel in the context of the values of the pixels surrounding it

Qualitative raster neighborhood modeling

majority, minority, diversity, minimum, maximum, mean, standard deviation, clean up rasters of elevation/water data and elevation used heavily in social sciences

Spatial frequency

chracteristic of maps and remotely sensed images is a parameter, defined as the number of changes in digital number value/unit distance for any particular part of a map/image

Spatial frequency in raster thematic maps/remotely sense imagery

spatial convolution filtering, fourier analysis

Spatial convolution

uses a convolution mask to enhance low and high frequency detail, as well as edges

Fourier analysis

mathematically separates a raster map/image into its spatial frequency components

Low-frequency filtering in the spatial domain

coefficients, Ci, in the mask template are multiplied by the following individual values, Vi, in the input digital image/raster map

Example of low-frequency in the spatial domain

usually set equal to 1 in the filter, median filter makes it more blurry and zone color are smoother

Linear edge enhancement

performed by convolving the original data with a weighted mask/kernel

Parts of linear edge enhancement

embossing with a shaded-relief format, direction of the embossing controlled by changing the value of the coefficients around the periphery of the mask, compass gradient masks may be used to perform 2D, discrete differentiation directional edge enhancement, compass mames suggest the slope direction of maximum response, laplacian filters

Laplacian filters

2nd derivative (as opposed to gradient which is a 1st derivative) and is invariant to rotation (insensitive to the direction in which the discontinuities run)

Nonlinear enhancement

sobel edge detector uses a 3×3 window numbering scheme and is computed according to the relationship

Example of nonlinear edge enhancement

Roberts edge detector based on use of 4 elements of a 3×3 mask, new pixel value at location computed

Zonal operations

zonal statistics

Embossing

shaded-relief format