GEOM30009: Imaging the Environment

4 assignments worth 10% each, 3 hour exam worth 60%

This has the makings of a really interesting and valuable subject for geomatics, engineering and breadth students interested in the huge field of remote sensing. However, this semester it was poorly organised and executed by the staff.

As mentioned, it focuses on remote sensing, and you will broadly look at topics including image interpretation, planetary remote sensing, geological and environmental imaging, image display and rectification, and techniques of satellite and aerial imaging. None of this is particularly complicated as this subject does not even approach the math that underlies most of these concepts, just introduces the concepts itself. Not a lot of new material is really introduced if you have already done GEOM subjects in first and second year, so its a very, very incremental subject and doesn't require too much effort to understand the content.

The lectures themselves are a bit dull. As mentioned, a lot of the content rehashes or slightly expands on what other GEOM subjects have already covered, so its really about consolidation rather than any real challenge. Joe speaks extremely slowly, and I found you can basically speed a lecture recording up by about 1.7x without missing a beat. The slides themselves aren't particularly well constructed; they are basically just a series of scanned images with very little text. In some cases, the lecture material is quite dated and in need of a complete revamp. Joe also never seems to respond to emails, and it wasn't until week 7 (and presumably 10s or 100s of emails) that the link to the current year's lecture recordings was provided.

But that said, the strength of this subject is the hand-on practicals and assignments. Unlike many other geomatics subjects, Imaging doesn't have fieldwork pracs, rather it has computer lab pracs. All of the assignments are heavily completed within the pracs, and Joe, the lecturer, turns up to introduce all the assignments.

The assignments involve read-world applications of the things you learn in the lectures. The first is probably the easiest assignment you will ever do at Uni, as it basically involves describing what you see in a set of images. That's it. The difficulty picks up a bit with the second assignment, where you have to do a crater count using satellite imagery of Mars to discern its age. Then you move onto image processing, as with assignment 3 you use band selection and image stretching to display the Port Phillip bay area in a range of spectral configurations. You then delve into the imagery and discern how features change under different configurations. Assignment 4 involves doing more complicated band ratios and algorithms for essentially the same task (i.e. identifying land use patterns). Below is an example of the sort of results that this yields.

Yes, with screen capture

Joe Leach

Yes, 3 under the format

2.75 out of 5

Nein

1x2 hour hour lecture (actually went for about 1:15 on avg.) and 1x2 hour practical computer lab class

Sem 1, 2014

H1