|Sunday, January 13th, 2008|
Pico-ITX based Johnny 5 Robot using Roborealm
Figured I would stop by and share a project I have been working on, it is a Johnny 5 type robot equipped with a full functioning onboard PC, using the pico-ITX formfactor from VIA. I'm using Roborealm (vision processing software) for it's autonomous behavior. Here is a short clip of some of my progress so far:
I have a few other videos posted on my Youtube account, and for more info on the entire project, visit my project page @ Trossen Robotics
|Saturday, January 5th, 2008|
I'm pretty new to LJ and this community, and was curious to see what you were all working on or why you are interested in image processing?
I'm a graduate student working on making a computer model of what draws human attention in images and videos (aka the field of visual saliency). I have many side interests in computer vision wrt super-resolution, graphical models (MRFs in particular), machine learning, segmentation, and clustering.
I'd love to hear from all of you! Current Mood: excited
|Saturday, December 1st, 2007|
|Wednesday, April 11th, 2007|
region-based segmentation code?
I was wondering if you guys know of any image segmentation code out there? I'm trying to do region-based segmentation, where each pixel begins as being a member of its own 'cluster' and you aggregate pixels together based on some criterion (distance formula, etc). I tried developing my own, but it doesn't work as well as I wish it would. I tried searching on the internet for one, but was unable to find anything. Anyone know of any code available online for this?
I'm currently working in matlab, but I'll take code in any format (C/C++, OpenCV, Matlab, etc). Thanks for any help!
|Thursday, December 14th, 2006|
How should I approach the following problem:
I have a signal with a certain S/N ratio. I want to digitize it to create an image. I will use this image for pattern recognition and display.
Question: what number of bits per pixel should I use.
|Thursday, March 30th, 2006|
Steve Eddings from Mathworks has a blog called Steve on Image Processing
. If you use MATLAB for your homework and/or research, but you're a MATLAB n00b like me, then you might be interested. I created mwsoip_rss
to syndicate his feed.
|Thursday, March 16th, 2006|
Journal of Vision
If you read ARVO
's Journal of Vision
, I threw up a syndicated account for its feed @ jov_rss
. Seems to be oriented towards biological vision, but since that's the standard we seek to emulate, maybe it's relevant to your interests...
|Friday, January 20th, 2006|
If you need to take a short break from your research, then check out this excerpt in New Scientist, The robots are coming!
It's from a book called "How to Survive a Robot Uprising" by Daniel H. Wilson
. I hope you find the excerpt is relevant to our interests...
|Wednesday, December 14th, 2005|
My name is Omlette, I'm currently studying Computer Vision at the University of Central Florida.
My current area of research is Automatic Target Recognition using Multi-View Morphing. I'm not really a big fan of military stuff, but until I can show off at least one project successfully completed, I'm just going to go w/ the flow... Once I can prove how awesome I am, I will switch to Medical Imaging or something equally harmless.
Is anyone else here doing object recognition/registration? If so, then what papers did you find useful? I saw the link to "Is bottom-up attention useful for object recognition?" and I added that to my reading list, what else? Thanks in advance
|Monday, December 12th, 2005|
Excuse for my English. It's not my native language, but I'm really in need of help.
I'm using MATLAB and I have some question.
There is, for example, a numerical variable a = 1000
and it is necessary to convert it into a string variable, i.e. a = '1000'
I don't know how to do that. Maybe you know, don't you?
And other question.
create text object in current axes, but I have to write text (x,y,'string')
. And if I have a numerical variable e.g. a = 1000
I cannot write it in such way, I have to convert it into a string format but I don't know how to do that. I'm sure there is another way to write text in current axes. And if you know this way please help me.
|Tuesday, July 5th, 2005|
I'm using MATLB7 and I have some trouble in MATLAB GUI programming, and there is no answer for it in Google, so I would be very grateful for your help.
I have made GuidemoMY.fig figure file and its corresponding GuidemoMY.m file was created.
GuidemoMY.fig contains several edit boxes (text fields for user input) which are initialized to some values, and one RUN pushbutton, that has callback function myRun(...) which has to retrieve those edit boxes values.
The problem is, that myRun(..) cannot read those values from 'handles' structure passed to it.
I don't know why, but before the apperance of callback function declaration and body, the 'handles' structure IS recognized, and the command
a = get(handles.figure1Probab,'String');
is successfully evaluated and returns string value of edit box with 'figure1Pribab' tag.
BUT same line INSIDE body of myRun(..) callback (which is successfully called after pressing the RUN button on GUI figure) -- is not evaluated,
and I get "invalid handle object" error, although 'handles' IS NOT EMPTY (I checked it in DEBUG mode).
What the hell is happenning and how can I retrieve the edit box values inside my callback?
Thank you very much!!
|Saturday, April 16th, 2005|
move that colorbar! and give me two axes that i can tailor
Hi. i am not sure how often people check this community but here goes my question:Query 1
: i am working on a plot with five parts to it but four subplots. I cannot figure out for the life of me to get the color bar to situate itself on the plot without rescaling one of the three lower subplots, so i moved it to the xyy plot at the top, but it cuts off the x-axis.
any ideas?Query 2
: the first subplot is a xyy plot and i need to set the values of the two y-axes as constant and not actually input data. i just need a blank plot with preset axes. is that possible?
I put the bits of code I am refering to in the following lj-cut. thanks a lot (in advance) ( subplot partyCollapse ) Current Mood: working
|Sunday, March 13th, 2005|
|Thursday, March 3rd, 2005|
anyone knows a working MATLAB code for calculating Area under ROC curve ? Had a code sometime back and am unable to get it working now :-(
|Saturday, December 11th, 2004|
|Thursday, November 25th, 2004|
|Sunday, November 7th, 2004|
I'm having a lot of problems finding ALL graphic file formats supported by the MatLab 'imread' function. Does anyone know all graphic file formats supported by MatLab?
|Friday, November 5th, 2004|
I am unsure if people actually even read this community and since there aren't any specific MATLAB communities (well none that I could find) I thought I might get a response here.
My question is:
Are there any ways to get MATLAB to save a figure (specifically a histogram) to file without viewing it first. I ask this because the version of Matlab I am using (R14 student edition) under Windows 2000 seems very unstable and often crashes as well as taking up vast amounts of memory when viewing a figure and hence I would like to bi-pass this step and merely save the file as a jpg directly.
Thanks in advance.
|Wednesday, September 1st, 2004|
Fourier Analysis for Dummies?
Could anyone please point me to a resource that explains Fourier analysis in words an idiot could understand? I have read so many things on it, and it still makes absolutely no sense to me. I understand it's a graph of the sinusoidal frequency...but the frequency of what, and how is this measured? And how are the points on the plot decided? I'm sure there's a simple answer for this, but it's just completely going over my head.
Thanks for your help!
|Wednesday, July 21st, 2004|
Hello to everybody!
As I see, the community isn't too active, though I thought that somebody may read it.
I have a problem of detection circles in the image.
As I know, the effective way to make this task is Hough transform.
May be anyone knows, where I can get sources that perform such a transfrom or something similar?
Thanks a lot!