Posts

Posted on

Presentations This Week

Hi Class,

Next week are the presentations.  The schedule is:

Tuesday: Javan, Jason, Patrick, Tom

Wednesday: Luisa, Ammara, Michael, Patrick, Jason, Kaitlin

Here are a few resources for you to use to improve your powerpoint presentation:

Jack

Posted on

The NEON Program and Regression in ArcGIS

Hi class,

This week, in addition to talking about your projects, we will look at the NEON program at NSF.  NEON stands for the National Ecological Observatory Network.  As part of the network, they are building planes with an imaging spectrophotometer, a LiDAR and a multi-spectral imager on board (one has been delivered, two more planes on the way).  A page describing these planes (including a video) may be hound here: http://www.neoninc.org/science/aop

In addition, some of you are planning on doing regression analysis with some of your map data.  There are some things you need to worry about when doing spatial regression.  A good introduction is presented in the free training course on ESRI.com called ‘Regression Analysis Basics in ArcGIS 9.3’ (http://training.esri.com/gateway/index.cfm?fa=catalog.webCourseDetail&courseid=1640).  This is a video lecture and demonstration that has good advice for any regression, especially spatial.  The tools haven’t changed very much in 10.0, but there are a few new ones in 10.1, so it is worth looking at the talk if you are going to do regression.  To view it, you will have to register, but it is free.

In class today (Thursday), I’ll go through a regression tutorial that you can download from the ArcGIS.com gallery (or you can download my copy for ArcGIS 10.0:  https://udrive.oit.umass.edu/xythoswfs/webui/_xy-12027721_1-t_eO3ed61U).

The first tool to try is called ‘Ordinary Least Squares Regression’.  This tool is primitive compared to R or SAS.  For example, it will not automatically use categorical data!  To make it do categories, you have to construct dummy (0/1) variables (just like R does) by yourself.  I’ll show you how this works in class today.  To make it work you need a set of points or polygons with attributes for each one.  One of the attributes will be the response variable you are trying to predict, and one or more will be predictor variables.  You must know the response for some of your points, so that you can construct your statistical models.  You then specify the variables to use in the OLS tool and you will get out a regression equation and a bunch of statistics.  You can use the equation to predict the response of other places on the map that you didn’t measure!

You need to look at whether their is spatial autocorrelation in the data that isn’t accounted for yet.  Calculating Moran’s I on the residuals will tell you that.  If there is autocorrelation, then you should use the Geographically Weighted Regression (GWR) tool, which takes into account spatial autocorrelation and computes a set of coefficients for each point in the dataset (and nearby points as well).  This can be very time consuming, so be careful.  If you need to do this, I’ll show you how.

Jack

Posted on

Selecting Random Points for Class Evaluation

Hi class,

Today, we’ll look at how to randomly sample from an image.  The idea is that you want to assess how well you did in classifying an image.  One way  is to select a bunch of locations and see if you got those locations correct or not.  The problem is that you need random samples, and those are hard to do by hand.

So we’ll look at several ways to randomly select sample points.  NOAA has an Add-in that is in the ARCGis Gallery called ‘The Sampling Design Tool’.  It is useful if you have a polygon to select from (like town boundaries, or a polygon class map like those distributed by MassGIS.  Installing the Add-in is a little tricky, but I’ll walk through it today.

For rasters, you do it by creating a uniform random grid, and use the setnull statement to select the number of random points you want.  You may then want to convert that raster of selected cells to a point file that is easier to work with.  I’ll go over this in class too.  You might need to do this for each class; this is a statified random sample.

Once you have a random point file, you can use the points to store the class that you map says that point is, and the class that you (or you ground truth) say it is.  You then have a way to calculate the %error in your map byt class.

Jack

Posted on

LIDAR Lecture & Lab

Hi class,

This week we will look at another Remote Sensing technology that I have mentioned before.  We will look at it in more detail, and look at its potential for the future.

LIDAR.zip : https://udrive.oit.umass.edu/xythoswfs/webui/_xy-11945030_1-t_wMyzspiy

Also, we’ll talk some more about the projects.  From now on, please bring up questions/problems you are having with your projects.  In particular, I want to make sure that everyone has the data they need.  We have less than a month before the presentations, so let me know if you are still looking for data!

I we get done with LIDAR this week, I’ll talk about Hyperspectral data.

Jack

Posted on

Some tools and techniques to get you through your project

Hi class,

This week I’m going to show you a few tools that you many not have seen before.  First, go to the Change Matters page at ESRI.  This site let’s you pick anywhere in the world and look at various images (color infrared, enhanced vegetation, etc.) from 1975, 1990, 2000 and 2010.  It also displays a change image for the chosen years.  For those of you that are studying change, this is a good way to see what you might be looking for, and better understand some of the methods for detecting change.  It is also very fast, compared to downloading data for yourself.

One reason this works is the data set being used: the Landsat Global Land Survey (GLS). These are orthorectified, and terrain corrected images, so they overlay each other quite well, and they are worldwide for the four years mentioned above.  If you have downloaded an image that is not lining up correctly with your ground truth, then you might think about using a GLS image to help you shift or georectify your image.

Lab this week will cover several Spatial Analyst techniques for working with images, including masking, using the con() and setnull() statements to eliminate unsightly excess zeros, and split images apart or paste them back together.  The lab outline is here:

https://udrive.oit.umass.edu/xythoswfs/webui/_xy-11907615_1-t_n5PyBTwQ

Jack

Posted on

Radar Remote Sensing Lecture

Hi Class,

Thursday’s lecture this week will be on Radar Remote Sensing.  I am also posting a radar image centered on the Quabbin Reservoir.  It is not georeferenced (as you can see, but I will show you in class what it looks like when it is georeferenced.

Jack

Posted on

Thermal Remote Sensing (Lecture and Lab)

Hi class,

This week we will be looking at doing thermal remote sensing.  In lab we will convert the relative digital numbers we get on Landsat images to physical values such as power or temperature.  This is sometimes a useful step in preparing all imagery for classification or for ratioing (e.g., NDVI).

Jack