Spatio Temporal Aquarium

Monday, August 01, 2005

datacursormode problem solved. Finally.

How stupid of me!!!

I have been trying to figure out why I could use datacursormode with everything except my data in my GUI (my data outside of my GUI worked fine). It turns out that somewhere in my screeds of nicely commented code I had helpfully set my data in my GUI to "Unclickable." Apparently I did this to speed things up (thank god I actually commented the code in the first place). This is why nothing happens when I try to click it with the data cursor! I have been struggling with this for weeks! Aaarghhhh!!!!!!

At least I know now...

Useful piece of information: datacursormode does not work on things that have "HitTest" set to "off".

Now, maybe I can make some progress...

¶ Monday, August 01, 2005 3 comments

Tuesday, July 19, 2005

Data Cursor Again

Aaargh!!! How frustrating.

The data cursor does not seem to work on my grid data. It only works on my vector data.
I can select a vertex on a line, or part of a patch/surface, but I am a little stuck on how to select a voxel. This could be a major stumbling block.

Major hacking ahead.

¶ Tuesday, July 19, 2005 8 comments

Tuesday, July 05, 2005

Meeting Minutes

Since the last meeting I have done the following:

Converted STVTool to new version of matlab
Started implementing frizzle (am stuck on getting 3d point information into program)
Asked TMCM for data
Started this blog to record progress
I have had extra work committments, so thesis work time was limited to one weekend.

Over the next couple of weeks I have a relatively light workload at work (ie standard 40 hour week) so I plan to:

Continue Frizzle implementation
Obtain TMCM's data before he flits off overseas
Obtain Halifax data from Pip
Tidy up data input/selection
Write something
Investigate TerraSeer
Investigate NZ ESRI User conference

Next meeting: 8am 20th July

¶ Tuesday, July 05, 2005 1 comments

To Do

To Do Over next two weeks:
1. Get Data sets from Otto (he is making me a dvd of a voxel data set)
2. Implement that thing that uses datacursor mode.
3. Tidy up Data set input/scaling.
4. Write something.

¶ Tuesday, July 05, 2005 0 comments

Wednesday, June 08, 2005

Matlab Problem: datacursormode

I need to customize the datacursor, but have no idea how to go about it. I have looked at the datacursormode function, but there isn't a lot of detail in the documentation.

Basically I want to run another function when the user clicks on a data point with the data cursor.

I'm sure it is possible, I'm just a bit annoyed with myself that how to do this doesn't seem obvious.

Technorati Tags: Matlab, cursor, datacursor, customize

¶ Wednesday, June 08, 2005 0 comments

Monday, June 06, 2005

GIS Tip: Joining Large Shapefiles

If you are joining a very large shapefile on an attribute field, it is much faster to join the dbfs and calculate new fields in Microsoft Access.

Then just overwrite the dbf part of the shapefile with the new improved dbf.

Technorati Tags: GIS, shapefile, join, tips

¶ Monday, June 06, 2005 0 comments

Sunday, June 05, 2005

It works!

...although "works" is probably an exaggeration. It is more like: I have gotten rid of the obvious error messages and the prototype seems to be doing what it did before.

I think that is enough brain strain for today.

¶ Sunday, June 05, 2005 0 comments

Matlab Solution: setappdata and structure arrays

Setting up the structure of the array first IS the way to go about it.

Does Not Work:

setappdata(STVFig,'dataset.isosurf.M',[]);
setappdata(STVFig,'dataset.isosurf.O',[]);
setappdata(STVFig,'dataset.isopatch.M',[]);
setappdata(STVFig,'dataset.isopatch.O',[]);
setappdata(STVFig,'dataset.isopatch.XY',[]);
setappdata(STVFig,'dataset.isopatch.XZ',[]);

Works:

dataset.isosurf.M =[];
dataset.isosurf.O = [];
dataset.isopatch.M=[];
dataset.isopatch.O=[];
dataset.isopatch.XY=[];
dataset.isopatch.XZ = [];
setappdata(STVFig,'dataset',dataset);
clear dataset;

Technorati Tags: matlab, gui, setappdata, appdata, structure, field

¶ Sunday, June 05, 2005 0 comments

Matlab Problem: setappdata and structure arrays

It turns out my old code is not as messy as I thought. There are lots of comments, and it is relatively logically structured. I have decided to just modify the old code.

So, I try and run it. And it crashes on the Initialise routine.

For some reason, the code that used to work, no longer works. That's OK. It is what I expected.

However, I have spent the last hour trying to figure out why it doesn't work, and I am stuck. I think it should work. I think the documentation says it should work, but the documentation is not very explicit.

The problem is on the setappdata function:

setappdata(STVFig,'dataset.isosurf.M',[]);

It doesn't like the fields (ie "isosurf" and "M"). It is quite happy if I just do this:

setappdata(STVFig, 'dataset',[]);

But I don't want to do just that. I want dataset to be a structure array like this:

dataset
- isosurf
- M
- O
- isopatch
- M
- O

I can't figure out how to define the dataset structure in the setappdata function. Usually, you can just add a field like this:

dataset.isosurf.M = [];

I have tried setting up the fields, and then using that in the setappdata function, eg:

dataset.isosurf.M =1;
dataset.isosurf.O = 2;
setappdata(STVFig,'dataset',dataset);

but that doesn't seem to be working either.

I think I need more coffee. I really want chocolate, but I know if I go home, I won't come back.

Technorati Tags: matlab, gui, setappdata, appdata, structure, field

¶ Sunday, June 05, 2005 0 comments

I Am Starting Again

Well, not again exactly.

I used to have a tool that allowed interaction visualization of spatio-temporal data sets. It was written in Matlab. I need to update it and add functionality.

I have the option of modifying my old code or recoding. Given that the old code was written for an older version of Matlab, and it was rather messy, I think it will be easier to start again.

Here are screenshots from the old tool:
isosurf2d

isosurf3d

Today, I am going to figure out the new version of Matlab, and start setting up the User Interface.

¶ Sunday, June 05, 2005 0 comments

Purpose of this Blog

Blog Purpose: To track my thesis progress.

Thesis Title: Visualizing Spatio-Temporal Data using a Time Geographic Approach.

Keywords: GIS, Visualization, spatio-temporal, time, Time Geography, aquarium, Matlab

¶ Sunday, June 05, 2005 1 comments