[ECE 453] Re: Meeting and Important Information
- From: "Logen Watkins" <slapkey@xxxxxxxxx>
- To: 453_all@xxxxxxxxxxxxx
- Date: Tue, 2 Dec 2008 21:42:39 -0500
Hi everyone,
I don't know what time everyone has available tomorrow, but unfortunately I
have to be at work at 6 -- that is, assuming I feel better than I do now. I
have a test at 230, so I would like to have everyone meet at 3:30 in Riggs.
*You need to be there.* I realize that some of you may be unable to make it
because of class, and that is fine. I imagine as tomorrow is Wednesday and
we are doing a run through Thursday, that there will be plenty enough to do
when you can get there. I would like to have some rough drafts of a
presentation by tomorrow night, and we'll need to meet again Thursday to
make sure we're all on the same page, and for individual groups to run
through their presentations.
Let me know if you have any questions.
-logen
On Tue, Dec 2, 2008 at 9:32 PM, Lee Boaman <lboaman@xxxxxxxxxxx> wrote:
> I imagine we are going to have a meeting tomorrow to discuss progress
> and/or
> presentation. My fearless leaders, any idea what time?
>
> Thanks,
> Lee
>
> -----Original Message-----
> From: 453_all-bounce@xxxxxxxxxxxxx [mailto:453_all-bounce@xxxxxxxxxxxxx]
> On
> Behalf Of Matthew Warren
> Sent: Tuesday, December 02, 2008 7:18 PM
> To: 453_all@xxxxxxxxxxxxx
> Subject: [ECE 453] Re: Meeting and Important Information
>
> I have no class at all tomorrow, so if it would be helpful for me to
> come to the 449 lab to help somebody work on anything, I'm free - just
> let me know.
>
> -Matthew
>
> On Tue, Dec 2, 2008 at 6:20 PM, Logen Watkins <slapkey@xxxxxxxxx> wrote:
> > Hey everyone,
> >
> > I'm finally out of my comatose state -- which was better than the state
> > before it, but we won't go there; Not feeling great but I'm going to get
> > some serious work in tonight because I know we're on a strict deadline
> here.
> >
> > The basic idea behind my code is the producer/consumer model you may
> > remember from ECE 329 (I think?). I have four threads, a pair for images
> > and a pair for data. The "Receiver" of each pair is listening for new
> > data/images, and when it finds something adds all of that information too
> a
> > queue to be processed. That enables the Receiver to quickly look again
> for
> > data, instead of going through the writing process (in the images
> example)
> > and THEN looking for new data. In my tests I think it worked better than
> > doing it all in one class. So the "Writer" of each pair is scanning the
> > data and image queues for stuff to be processed. For data, this involves
> > gleaning from the data transmission all of the vehicle information
> > structures that DBusch and I set up, and subsequently inserting those
> > records into the database. For images, a filename/location is made and
> the
> > image is written to the harddisk, followed by an entry into the database
> of
> > this frame number (what links the images to the data, as chris said) and
> > file location.
> >
> > i've been in touch with those at the TMC so we're trying to get this
> thing
> > up and going. I will say that the code on the TMC currently is very
> > outdated, and I had stripped parts of it to do the testing chris
> described,
> > so don't try and use it.
> >
> > -logen
> >
> >
> > On Tue, Dec 2, 2008 at 6:08 PM, Chris Rodgers <Rodger4@xxxxxxxxxxx>
> wrote:
> >>
> >> Hey Kenny,
> >>
> >> I'm not really sure if I can talk intelligently about the code that
> Logen
> >> was writing. However, I believe his app was going to create threads for
> >> each of the four fixed cameras. Each thread was going to use the
> network
> >> API to ask and receive data. I believe these threads were going to
> check
> >> for an image/data pair before inserting the image path and vehicle
> >> information pertaining to that image into the database. However, I'm
> not
> >> sure how much of that is done. I don't know if dummy threads were
> created
> >> or if you can just plug the API function calls in there and be
> good-to-go.
> >>
> >> A separate thread in the program is also going to periodically delete
> all
> >> of the images older than a day (we believe that's our worst-case storage
> >> limitation). I know Dan was working with Logen to do that part of it,
> but I
> >> think you'd have to ask him to see if that got done.
> >>
> >> Some of the code that Logen wrote is actually on the TMC. You just need
> >> to open visual studio express 2008 on the server and go to the
> LogenCCode
> >> tab. I have only looked at this code as an interface for testing other
> >> components on the webpage, so I don't know to what extent it has been
> >> altered to act as a test rather than a functional app.
> >>
> >> The website is ready to go for when the API is up, so nothing needs to
> be
> >> done there. At least, we hope... I know that probably sounds
> terrible...
> we
> >> did extensive testing without the API up to see to it that everything
> will
> >> work. However, without an API going on the server, we don't know for
> sure.
> >>
> >> Sorry I can't be of more help.
> >> -Chris
> >>
> >> On Tue, 02 Dec 2008 17:32:20 -0500, Ken Knox <kknox@xxxxxxxxxxx> wrote:
> >>
> >>> Guys,
> >>>
> >>> Guess somebody canceled the meeting? About four of us showed up tonight
> >>> and Lee set up the routing for the last IP camera. We are getting great
> >>> feeds from the fixed cameras and the IP cameras.
> >>>
> >>> We need to do video tomorrow since the weather will be sunny with a
> high
> >>> of 56.
> >>>
> >>> Since Logen is sick we need to band together to try and get the control
> >>> center up and running by sometime early tomorrow. We at least need to
> >>> figure out a way to get the fixed camera feed to the control center
> >>> website. I'm willing to volunteer my time to get the system functional
> by
> >>> tomorrow.
> >>>
> >>> Worst-comes-worst we can run the algorithm at each location and take
> >>> video
> >>> of each. I hate to come this close and not succeed. We probably should
> >>> plan another meeting for tomorrow to do all the video recording, etc.
> >>>
> >>> Does anybody have an idea on what needs to be done to get the control
> >>> center data passing functional?
> >>>
> >>> Thanks,
> >>>
> >>> Kenny
> >>>
> >>>
> >>
> >>
> >>
> >
> >
>
>
>
Other related posts:
