Applications in GIS Module 1

Grid Overlay

Kernel Density:

Local Moran's I:

This weeks module focused on crime analysis using the grid overlay, kernel density and local Moran's methods. The grid overlay method (image one) involved layering a grid of half mile cells over the city of Chicago, isolating the cells in the top 20% of total homicides in 2017 and using that to compare it to locations of all the homicides in 2018 to determine where homicides where recurring. The kernel density method (image two) creates a density map by searching for clusters of points in an area and assigning points closer to the center of the search area higher weight values. All area total homicides numbering more than three times the average were isolated. Local Moran's I (picture three) determines whether points inside a cell are either clustered or dispersed and then determines whether or not they are statistically significant by comparing each cells value to the cells around it. All cells with high clustering and high statistical significance were isolated for analysis.

Comments

Module 5 - Unsupervised and Supervised Image Classification

This weeks module focused on classifying images using multispectral signatures. Above you can see the completed classified land cover of Germantown, Maryland. To create this image above signatures were collected that correlated to each required feature. Then bands were chosen (R:4 G:6 B:5) that contained the largest separation amongst features to minimize spectral confusion. In the above image roads and urban areas were often confused leading to a much larger area of roads than actually exist. The inset map contains a classification distance map which displays the distance each cell is (spectrally) from the sample points with brighter pixels being further than darker pixels. This indicates that brighter areas have a higher chance of error.

Lab 5: M 2.2 Interpolation

This weeks module focused on identifying the best interpolation method for modeling the air quality over Tampa Bay. Four methods were tested using the same set of sample points Thiessen, Inverse Weighted Distance (IDW), Tensioned Spline (seen above), Regularized Spline. Thiessen Interpolation assigns all cells in the raster with the value of the nearest sample point. IDW calculates the value of all cells by considered multiple sample points nearby and giving closer points a higher weight than further points. Both Spline methods create a smooth surface over the sample points but the regularized version creates a smooth curvature regardless of the range of values in the sample meaning cell values can end up both above and below the minimum and maximum values found in the sample. The tension model attempts to fix this by constricted the curvature of values to the ranges found in the sample points.

GIS Portfolio - Michael Malone

Search This Blog