My presentation on DMMS

Here is my presentation on distributed Multimedia systems . I didn't mention the references in my slides. So, if you need them please let me know :)

my LinkedIn and Twitter account

I added my LinkedIn and Twitter account to the right side of the blog . I use these services a lot , so you can be connected with me more within this way ...

my sight into CS world was worng ?

I didn't blog recently . The main reason was due to the context switching from doing pure CS foundation to more real-world-based applications . If some of the best criteria (based on Joel Spolsky's essay ) for being highly productive programmer (i.e. hacker in my term) are to be  :

1. Being Smart
2. Get things done.

I definitely did wrong at the second part , and I don't like it .

Something should change , not in word but in action ; Those changes has already begun  ...

What Uppsala taught me

The first year at Uppsala University has finished, and like always, I started to think what was the gain of this year .
I think the most important aspect of studying in Uppsala was that they taught me HOW TO PROGRAM MATHEMATICALLY !! This is not a simple matter , specially to a student who had Electrical Engineering background and not Computer Science . Of course, they didn't teach me step by step on how to do it, but they provide some proper infrastructure to achieve this goal .
Unfortunately , I learned it a bit late (mid of May was the "Aha" moment to understand the whole concept ), but when I look back , I'm satisfied with my progress . Besides of some difficulties that this major (Scientific Computing) has , it is very worthwhile for people who want to mix their science/engineering capabilities with high-level mathematical programming . I don't know how I wanted to learn these stuff if I didn't pursue this major , but I know that it changed my perceptive to science itself.  For instance , I always wondered how I can implement a very long formula which expresses a physical rule ! Now , I not only know how to do that , but I can observe the most computationally consuming parts that affect the whole relation .
If you are interested in these stuff , I can explain it in future posts ...

Artificial Life

YES !!
At last , Craig Venter and his team made it after 10 years .
I am so so excited , I even can not type these words ! Craig had a lecture in my university (Uppsala Uni.) this winter and gave a long lecture about their discoveries .
WOW ... this is amazing ; What other things creationists need to believe in evolution ??
I love science , I do love it  . It moves humbly without any exaggeration ...  

Knowledge and Simplicity

Everything should be made as simple as possible, but not simpler.
Albert Einstein

The Most important thing , I've learned here is to simplify stuff as much as you can "understand" every single detail of it . That's it , no more ....

My conclusion is so close to what Albert Einstein has told before !! But I didn't understand it till I figured out a concept in "Reinforcement Learning" during "Machine Learning" course . I'm used to give up or neglect very tricky and hard problems before , but for the first time in my life (yes,it's sad) , I decided to change my mind.
And guess what ?? By only spending one hour on one page , the result was awesome and I did understand the whole chapter afterwards .

The moral result : NOTHING is hard ; Some stuff just need more focus and dedication ...

That's it friends :)

Fuzzy image Dilation and Erosion

One of the generic approaches to fuzzy Morphology is the Fuzzy Dilation and Fuzzy Erosion . Based on an trouble that I faced with during the Fuzzy image Analysis course , I decided to publish the Matlab code , based on the paper by "Bloch and Maitre" on "Fuzzy mathematical morphologies: A comparative study" . This is just the implementation of Definition 1 (page 1344 ). Needless to say that next definition are the same to Def 1 . Just play with the parameters :)

Here is the code :

%{
        Fuzzy Dilation and Erosion
        
        Copyright 2010 Vahid Rafiei .        
        This program is free software; you can redistribute it and/or modify
        it under the terms of the GNU General Public License as published by
        the Free Software Foundation; either version 2 of the License, or
        (at your option) any later version.
        
        This program is distributed in the hope that it will be useful,
        but WITHOUT ANY WARRANTY; without even the implied warranty of
        MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
        GNU General Public License for more details.
        
        You should have received a copy of the GNU General Public License
        along with this program; if not, write to the Free Software
        Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston,
        MA 02110-1301, USA.
%}


% Settings
len=512;
useNoise=0;
strel=1;

%% Define v
if strel==1
x=linspace(-10,10,len);
v=2-(x).^2;
v(v<0)=0;
end

v=v/max(v);
plot(v,'--');


%% define mu
x=linspace(-4,4,len);
mu=3-x.^4+4*x.^2;
mu(mu<0)=0;
mu=mu/max(mu);
if(useNoise)
    mu=mu+0.1*(rand(size(mu))-0.5);
end

hold on
plot(mu,'-','Color','red')


legend({'V','mu'})

% mu and v is ready

%% Definition 1

% Dilation

nalpha=10;
alph=linspace(0.0001,1,nalpha);
deltaAlpha=1/nalpha;

Dmu=zeros(size(mu));

tic
for a = alph
   valpha=v>=a;     
  for x=1:numel(mu)    
      mask=zeros(size(mu));
      mask(x)=1;
      mask=convn(mask,valpha,'same');
      Dmu(x)=Dmu(x)+max(mask.*mu)*deltaAlpha;      
  end  
end
toc

figure(2)
plot(mu,'Color', 'red');
hold on
plot(Dmu,'Color','blue');
legend({'\mu','D\mu'})
title('Dilation (Def. 1)');

% Erosion
Emu=zeros(size(mu));

tic
for a = alph
   valpha=v>=a;     
  for x=1:numel(mu)    
      mask=zeros(size(mu));
      mask(x)=1;
      mask=convn(mask,valpha,'same');
      mask(mask==0)=NaN;
      Emu(x)=Emu(x)+min(mask.*mu)*deltaAlpha;   
  end  
end
toc

figure(3)
plot(mu,'Color', 'red');
hold on
plot(Emu,'Color','blue');
legend({'\mu','E\mu'})
title('Erosion (Def. 1)');

Cheers ,

Meditate in Science

After years of thinking about what's exactly the difference between super succeed persons (Science in my concern) and ordinary people , I've come up with this :
"You have to Meditate in Science .... There is no other way".
The set point was an article in Newyork Times ; This article is about the importance of Practise and hard working which is not new of course . The difference starts by this sentence :

Mozart played a lot of piano at a very young age, so he got his 10,000 hours of practice in early and then he built from there.

 This is exactly what I want to talk about : 10,000 hours ! If you spend 3 hours a day on an specific work , it takes about 10 years and if you devote more ,say 6 hours per day , it takes 5 years to be master in your stuff .

This kind of devotion needs a self-disciplined mind and a powerful motivation , and that's why I say you have to meditate in science . In meditation you almost cut your communications with around world and just concentrate on your breath, for example . But if you extend this mindset to your specific interest (any interest , from music to science ,etc.) and start to meditate on it , you'll get there so sooner as you imagine .
Again , it needs hard self-discipline, but as soon as you start it, you will get used to it little by little . Inertia occurs when you want to start something, but after that stage, another inertia needs to stop you ; and in your situation you don't need second type of it anymore :)

Good luck on your journey ...

New mapping technology from Microsoft.

This is what that changed my viewpoint to Microsoft Research for ever !!
This fantastic brand-new technology from Microsoft is exactly what I though about in Visualization course . I wanted to mix the same idea with Google Map , but I didn't know how to start :)
Anyway , Microsoft implemented it comprehensively , and I'm really excited about it . This is a glamorous technology that delighted me deeply .

Well done :)

Where is the privacy Google BUZZ ?

Lately , the Google company have presented their new brand product , Google Buzz . Though, it seems that Buzz is an interesting technology , the level of privacy is awful ! As soon as I award this problem, I turned it off . Mail address is a key in privacy and personal stuff and Google should think about it more conservative . I Hope they fix this problem soon ....

Update 1: This is the first attempt of Google to fix the problem .

So long till now

OOPS ! It seems that I'm not an active blogger anymore !! I think it's because of social networks like Facebook and Twitter . They are really addictive .

Not really important .... I'll be active more by the end of February and the nature of posts tends to be more technical and less personal . 

TODO list for 2010

For coming year, I want to do some interesting works and I publish them here to reinforce my motivation !

1- I've decided to open source some of my projects . I'll do it little-by-little . I think they are 7-8 projects . I want to actively contribute to open source world. I don't want to shame when I face the sentence "Speak is chip, Show me the code " anymore !!

2- I've decided to switch to "Front-End" world (mostly JavaScript) for fun , beside of my main area of interest (Computer Vision).

3- This year is a so critical year or me . Lots of works are waiting to get done. Fortunately , I've become more self-disciplined nowadays, and this rises my expectations ....



** Any suggestion guys ??

a one-day trip to Malmo

Here is some pics of my one-day trip to Malmo city , south of Sweden .

You can see the pics in Flicker .

IEEE and LinkedIn activities

I updated my IEEE and LinkedIn accounts !
IEEE has always been a must for me , and on the other hand , I've recently decided to involve in LinkedIn activities more .
Honestly , I didn't know that LinkedIn is social network (embarrassing , yes ) !!
If you are a "Techie" and you want to connect with me , please drop me a line .You can find me easily by googling me ...

A seminar at Scientific computing

Today , Dr. Frank Ham from Center for Turbulence Research, Stanford University attended a seminar in Uppsala University and gave a lecture on "Large Eddy Simulation on Unstructured Grids" . You can find the abstract of it here .

There is no doubt at all, that Computational Fluid Dynamics (CFD) is one the most complicated science that we have ever made ! And maybe that's why many of the most brilliant minds are working in this or related fields .
I was interested in two aspect of this lecture : the Visualization stage and the backbone GRID infrastructure of the their project .
Surprisingly , they need the technology 100 times faster than what they have now (Nov 2009) by the end of 2017 !! It means an enormous progress in both hardware and software . Can they achieve this ?? I don't know , but I hope so ....

The second period

For the second period , I chose two courses :
"Scientific Visualization" and "Computer-intensive Statistics and Data Mining".

In Scientific Visualization course, they focus on VTK library and consider this huge library as a high level abstract one . I mean they don't concentrate on "under the hood" computer graphics basis; instead, they train students how to think from a level upper to determine the problem specifications and apply visualization techniques to demonstrate the results.
Another point about the course is that the formal programming language for this course is Python !! Oh my goodness ! It was one of the best news I had ever heard :)

But the second course is more interesting ; it's about "Statistical Pattern Recognition" ; thought, the primitive 8 lecture (out of 21 lectures) is about "random number generation, Monte Carlo , and Bootstrap techniques" . A nice point about this course is that We have to do our project in "R programming language" , which is an amazing language with exceptional capabilities for Statistical Programming .

Nevertheless , I have to try hard to dominate all of them . It takes 99% perspiration :)

Favorite picture

Licensed by NASA

This is the most inspirational picture I've ever seen !!
It will (maybe) become my permanent Desktop background . I changed my Facebook's profile picture too !!
As you can see it's so mysterious , weird, lovely.... Something about our pure curiously to unknown areas (I don't mean Theology bullshits, I mean absolutely Nature) . Something which rises from your inner whispers, Something that forced our ancestors to move out from their caves ....

About me these days

1- My classes have started since Aug 31. Here , the educational system is very packed . I mean each semester is divided into about two "2 months period", and normally you pick up two course per period.
Well , in my opinion , it's a good method for hacker-minded people (or self-taughts ones). You just focus on two topics and finish them , that's all ! Like a hacker who wants to penetrate a new stuff !!

2- My research interests are shifting to "Data mining / Visualisation", but it's not clear at the moment... I'll write about it later.

3- Human beings are the same any where :) I knew this, but I can see it here more ...

A Heavenly Voice

oh my goodness !

I've just discovered this heavenly voice by surfing the web during
last weekend !! Her name is "Jessica Mauboy". She is Australian and I'm her fan from now !! You can see one of her nice music videos :

Facebook

I joined the Facebook today !!
I want to become more familiar with web2 technology ...

You can find my profile here :)

Update :
First Impression : Facebook is so amazing !!

Uppsala Pics

Here is the first pictures that I took on my arrival to Uppsala city .

Please see them at Flickr !!

UPPSALA

I'm in Uppsala , Sweden at the moment .


Uppsala is a very lovely city; Lots of bicycles, nice and neat weather and guess what ! Full of historical buildings ....

UPDATE: There is a revision in my post ,as may be you can see :/

Huge setp to vision

At last , I got the Sweden student visa. No news from Australia,
but I'm not sad anymore !
After a long thoughts about my future, I believe that the Uppsala University is a unique and extra ordinary opportunity to go further into Computer Vision.
Frankly, I felt disappointment when I faced new mathematical algorithm during my work on vision problems. For instance, in Optimization techniques I suffered from lack of advanced math courses and someone who can debug my problems !! In some cases, I had to re-invent the wheel again and again !! But , it seems that conditions are changing.
I really love collage life, because I don't have to invest entire of my time on , for example, dealing with customers rather than focusing on technical problems. I don't want to say that customers always suck! But, It's clear that working with potential customers in countries like middle-east is too different from a country like United States. Things are not classified here, and I do hate disorderliness ...
long story short, the most prominent aspect of Computational Science at Uppsala University is that it reinforces my math basis. Look the first quarter lessons : "Scientific Computing, bridging course" and "Optimization" ! The first concentrates on some subjects like Linear Algebra , Numerical Analysis, etc. and second one has no need to explain ...
On the other hand, let's not forget some other technical areas like "High Performance Computing" which Uppsala is very well-known for it.
As you see, this Master program is really GREAT.

WOW ! What a challenging and active researches I will face :)

Missing Link Found

"Scientists working in Africa have discovered a Stone Age skull that could be a link between the extinct Homo erectus species and modern humans"
--National Geographic

As a someone who believes in Darwin's theory, I'm so happy to hear this :)
Though, regarding to the nature of the science , any new discovery needs lots of work and examines to prove ...

OOPS ..... UPPSALA ?

I got another admission from Uppsala University , Uppsala, Sweden .
I admitted at IT department, Computational Science Division.
As you can see at it's syllabus page, it has lots of Math based courses which is necessary for high-tech areas like Computer vision.
On the other hand, the Uppsala's ranking is so amazing :
1st in Sweden and 63th in the world !

What a weird admission. I was preparing myself for QUT, Australia !!!
But after 4.5 month there is no news from Australia Embassy :(
I'm so confused ! I love Australia and its hyperactive atmosphere. But I think that Uppsala provides me challenging projects as well ; I mean the program is an exciting challenge itself too :)

Hope I make the right decision ...

An amazing Flight attendant

Just see how much creative this guy is !! He's a staff member of Southwest Airlines ...

Why I love FOSS - 2

As you can see in Eric's essay, he strongly recommend to start programming by Python. I was familiar with programming languages before; I had passed the programming course ( C language) a year before, so I was familiar with some basic skills of programming (loops, I/O, etc).
But in my opinion , python was absolutely different from C.
As you can guess, the most visible aspect was indentation! I didn't use to it and I had lots of struggles with it.
Another part was classes. The "__init__" syntax was so weird to me and I was more comfortable with something like "this" in Java.
Anyway, I had to fight with these initial set points, and I did. I started with python tutorial in its homepage, but it was(is?) so fuzzy. Unfortunately, there were and are too rare python developer in Iran and my English literature was not so good to start reading a book from the beginning to end.
Yes ! I confess that it was not a smooth path, but i do believe that it was worthwhile to try...
On the other hand, Inasmuch as I was studying Electrical Engineering (power) and the nature of lessons were so different from the subjects like O.S. and web programming, this problem got bewildered my mind.
Because of lots of input data and lack of proper concentration, my productivity was too low. Essentially, take this as an experimental result :
"Focus on just one subject or go to the hell !!"
If you want to be a genius in a major, you have to focus on it days and nights in an uninterpretable form. Only this can show you the magical doors. That's it , That's all.
As I mentioned before, the lack of focus on one subject decreased my productivity enormously and the biggest result of it was the decline of self confidence. You know, lack of self confidence is the mother of any defeat....
If I invest my time and energy on just hacking rather than work on some stuff like H.V. structures and Linux kernel processes at the same time, I had been a hacker till now...
Long story short, after graduation, I decided to put entire of my time on cutting-edge software developing; nevertheless, at this period of time my country is not a high-tech one. I couldn't find even one company that works on Computer Vision. Hence, my friend and I found a new one, but it's not so promising.
I really intend to immigrate from Iran to to some where else that I can work on my area of interest and have active connections with international community ...

Why I love FOSS

I started my Hacking Path [1] with Eric Raymond's legendary essay (How to become a hacker) in 2002 .
Though, I'm not a hacker, I really love it's culture and communications.
I knew nothing about computers and honestly I scared of them before.
You know why ? Because my first contact with computers was through Microsoft Windows some years before(win95 I think) .
It was all wizard (type of installation of software,etc)... When I was turning on the PC, the magic was starting, I was seeing a monocolor
screen and then a nice background . "How does windows do it ?" I asked this question numerous times from myself, and you know,
no one knew it. Everybody was conquered by its wizards.
Damn you Microshit with your damn wizards.
But,then, a flickering light enlightened my brain suddenly. I read Eric's essay, and I think you can guess what's happened after it .
I started to convey the linux and re-assembled it from scratch several times.There was no magic anymore. Every thing was absolutely clear , numerous papers were in the queue to read and lots of dedicated developers were ready to answer the questions.
My life and mindset changed completely. When I say my whole life, I'm not kidding. Because I have been an absolute freedom fan, I found
the liberality spirit in FOSS culture. Linux has never been just a simple O.S for me , but it's been a new viewpoint to the world to me.
Thanks Eric for Your legendary essay. Thanks Richard Stallman,the great, for GNU Idea and your dedicated works to support it.
I will talk about my first impression when I saw python soon ...


[1] Please attention, I mean the Hacker in free/open source software (FOSS) term .
The distance between Hacking and Cracking is the same as the distance between Heaven and Hell !!

ClusterMaps New policy

I just noticed that the ClusterMaps window on the right side of my weblog has changed !
They emailed me at the same time that they have decided to make annual archive of locations !
Hey, Come On guys ! We live in the 21th century and we have a term call opinion poll :/

My Admission

After receiving some offers of admission, I plan to accept the offer of "Queensland University of Technology" (QUT) , Brisbane, Australia .
I found them so active in my area of research ( Computer Vision), so I think If I can get the Australia's student VISA too (this is a so molesting procedure) a new season of my life will begin.
I can work on what I do love in a dedicated form. I will have lots of talented colleagues and we can work in exclusive teams. I really like team-work projects and I believe that it increases my productivity enormously. Two brain are always more productive that one!
I've never been sure about the future , but I've decided to work hard and keep my hopes. That's every thing I can do :)

Update 1: Humm .... maybe I get another offer from different university !
It is Curtin University of Technology , and has a VERY active lab in computer vision area. Its projects make any geek crazy :)
It's a hard dilemma ! Any suggestion ???

Update 2: At last , I decided to choose QUT !
I didn't know that QUT has very strong relations with industry !
Here are some of their Areas of Research and collaboration :

Image and Video Research Lab
Speech and Audio Research Lab
Australian Research Center for Aerospace Automation
CRC for Spatial Information

Seems so promising, isn't it ?