When I grow up, I’m gonna be a Colour scientist?

So I really shouldn’t be doing this right now, and I should be intense studying for my exams, but I’m about to go insane from studying and figure this might be a good release. I got fed up with trying to understand particles and statistical mechanics and what do I do? I decide to read up on my latest guilty pleasure in the world of science, and that would be colour and colour theory. It sounds kind of mundane at first (colour naming is a big part of these researcher’s jobs, they even have colour thesaurus!) but it just doesn’t make sense when you throw a bunch of post-docs with computer science and physics degrees into colour naming. Upon further reading, I’ve discovered that there’s actually a plethora of complicated physics involved unbeknownst to the world. So what better do I have to do at 4 am in the morning than to give a job profile of the lovely underrepresented colour theorists.

Research is colour science currently is two-fold. First and foremost is skirting close to God himself: healing the blind by creating synthetic eyeballs. The relationship between how the eye sees colour and how the brain processes it is still basically a complete mystery to us mere mortals. We understand how the eye functions itself, and we can transmit the information to the brain, but the processing of the data in the brain remains a very vague field. If we can understand this, we can create much more efficient imaging processing algorithms, used in places like CT Scans and MRI machines. The second big research field is perfect imaging capture. In many cases, pictures taken with a regular digital camera turns out completely misrepresentative of how the scene actually is. Each imaging input and output region filters the picture to less and less of what it should be. In the example of the camera, the digital camera first captures the light in a slightly distorted manner. Then comes the problem of displaying it in a combination of red, green, and blue pixels that attempt to describe the already-distorted picture. The only true non-distorted filter is the one through our own eyes, which only through understanding how the brain works with processing the information for what image we truly see. Applications to this minimization of distortion include virtual archives of museum artifacts, and more detailed medical imaging.

Colour science actually gets pretty involved. I would just like to note the difference when the filters I was talking about before are applied. First picture is a comparison between the RGB (red-green-blue) system to the CMYK (cyan-magenta-yellow-black) system. This is a typical conversion between the display on a monitor over the actual printed image.

The next picture is a display of the colour systems. It shows a graph (with no axis — what blasphemous science is this?!?) with wavelength as the horizontal axis and the intensity/amplitude as the vertical. As expected, it is a nice distribution of colours and progresses through the spectrum of the rainbow. Left of this graph would be ultraviolet light, and right of it would be infrared, both not visible to the human eye and not relevant. The full distribution of light is what our eyes can detect, and the shapes enclosing a certain area describes the limitation of what each imaging technology can display.

To improve the colour representation this is where the math comes in. Surprisingly, it involves calculus, and even more surprising is that it involves a lot of linear algebra. Like all of the abstract stuff like vector spaces and stuff. I mean, who even does that anymore? In any case, it turns out you can define a vector space of colour, a colour space, that describes all the colours with the 3 fundamental colour vectors, red, green, and blue. Alternatively, you can describe all the colours in greater precision and crispness with vectors of only hue, saturation, and lightness. And whaddya know, to get from one colour space to the other, we apply a coordinate transformation! If you’re not getting much of this, it’s okay. I don’t really either. Man, I should have paid attention in linear algebra.

Turns out, with these coordinate spaces, you can define coordinate systems and apply calculus to it to find the change in hue/saturation/brightness in a certain direction in an image, and then make modifications to that function calculated to adjust its image clarity. Below is just a generic coordinate system (cylindrical coordinates!) of HSL.

That’s all from here. I think, just maybe, it’s time for me to go to bed.

Technicolor girls….

In the midst of hell, there is fire. Who would’ve guessed?

So amidst the recent hell from school, (2 all-nighters in a row a couple weeks ago; a new record, but that’s besides the point), and the same setup of lab reports and midterms coming up next week, I am taking my leisure in napping, slacking off, and just generally basking in the calm before the storm. The real storm is exam time, when 12.5 hours of examination are worth 50-60% of my semester’s mark. Fun fun fun.

So about 3 weeks ago, I skipped a couple of lectures to go to a series of Capstone lectures by several students. For those of you not well acquainted with the U of T engineering system, Capstone is basically a half -year thesis course, seen more in the engineering and law sectors. This gives people a big project, but small enough to diversify. Anyways, each pair had to give a 20 minute presentation on their topic while being graded. One of particular which I found rather interesting was on the topic of fire tornadoes. Come on, with a topic like fire tornadoes, you can’t go wrong.

Here’s a video of the highlight of the presentation. Kudos to Song for taping the thing. Albeit slightly unclear, you can see “fire grooves” when spun, which is essentially the spiralling tornado. What strikes me as amazing is the fact that you essentially have fire, virtually impossible to model given its chaos, apply some controlled energy to the system, and suddenly you have a very simple way of modelling the properties of this fire. In fact, the fire is governed only on 2 concepts, Kelvin’s circulation theorem and Bernoulli’s equation. Kelvin’s circulation theorem says that if you have a fluid that is incompressible and a low viscosity (not like molasses; water and air are good examples of such fluids) then there is a conservation of angular momentum. Imagine the skater spinning slow with its arms out, and then faster when they pull their arms in, except it is something like water in a container rotating around the central axis. The speeds will be faster near the edges, and slower near the axis (centre).  Bernoulli’s equation says that if you have a fluid moving from a lower velocity to a higher one the pressure will decrease, maintaining a conservation of momentum.  The higher speed at which the outer column rotates at, the narrower and more spiralling the fire tornado will be. The difference in pressure and velocity based on these laws make air flow in such a way that the fire that requires the oxygen to burn moves like a tornado. This is another video I found on youtube, except much cornier.

I guess in the midst of my stress and despising of the work I forgot the beauty of physics and why I went into this in the first place. It is for this very reason, at which everything can be explained with simple equations and concepts, and in the end all that it’s doing is it’s confirming your intuition. That’s the beauty of the simplicity of math, however complicated it might be.

Who said alcohol and science don’t mix?

Most of you should know what that is. Felt burning down the throats of many, underage or not, then followed swiftly with a hard bite down on a slice of lime, this is the oh-so-famous tequila. It has been long established as a recipe for disaster for those who have this as their drink of choice for the night.

In the university scene, it is very well known that science majors are not exactly the partygoers. We have been known to stay in our rooms on a Friday studying for that midterm 2 weeks from now (wow, who does that?). But today, we strike back! A group of scientists have bravely taken up tequila and made diamonds out of them! Wait, what? Yeah, you heard me. They made diamonds with tequila.

Like alchemists trying to turn lead into gold not-so-many years ago, scientists now attempt to create diamonds from organic compounds. After experimenting with acetone (nail polish remover) and just normal ethanol, they noticed a 40:60 ratio of organic compound to water was perfect.

40% ethanol. Where have I heard that number before? Oh right, hard liquor! After experimenting with various alcohol distillation types, they have found tequila to be most successful. As quoted from the scientist in the article, “There is no doubt; tequila has the exact proportion of carbon, hydrogen and oxygen atoms necessary to form diamonds.”

Many long nights (and wasted tequila bottles) later,  an established methodology has been laid down as the groundwork for possibly further production in diamonds. The tequila is first heated to 280 ºC to turn it into a vapour. The gas is further heated to 800ºC under high pressure to break down the molecular structure, resulting in diamond crystals. The diamond crystals were gathered on a silicon tray to make a thin diamond film.

This diamond film is hard and heat resistant, more so than other diamonds made with different organic compounds. Useful applications of this could include high power semiconductors, found in your computer, radiation detectors, or high accuracy optical devices. But for now, scientists will first experiment with the different brands and their effects on the diamond film.

That’s all from me for now. I am currently in a gathering data mode for something closer-to-home, so stay tuned!

First Past the Post

So upon writing my first blog post on Thursday I was gonna do another but then I got distracted and then I was from henceforth in lockdown mode in my room listening to some Rachmaninoff doing something beautiful, otherwise phrased as vector calculus. But alas, the midterm is over, the textbook is completely finished, and I am free of midterms for the next 2 weeks.

And with no midterms, comes gratuitous naps and t3h internetz! Since I got home from dinner with my roommates (Aside: we ate at a restaurant called People’s Eatery. It had several signs that had a touch of communist propoganda but the food was great! A typical asian restaurant, in the heart of Chinatown, but well presented in its own manner.) I have been clicking reddit like no there is no tomorrow.

Article of interest today is on the mysterious world of space. A new galaxy has been discovered, except there’s a catch. It’s 10.1 billion light years away. The light we see coming from this galaxy, this image, is 10.1 billion years old. If that galaxy decided to suddenly disappear from the spacetime continuum we would still see it for another 10.1 billion years before it disappears from our eyes. In a perspective of magnitudes, the sun is roughly 7 light minutes away from Earth. Furthermore, it should be important to note that the universe itself is roughly 13.5 to 14 billion years old. The image we see now is of when the universe was just a touch less than a quarter of its current age. I don’t know about you, but I think this is completely phenomenal, and I’m just jumping in circles at the idea of how great this find is. With this data, scientists can determine what the heck was going on with the universe in its humble beginnings, and how it evolved to how it is today. Now if only we had such ridiculous breakthroughs on Dark matter and energy. Proposed in 1933, the term was tossed around for 70 years, and still no one knows what the hell it is. Oh science, how you rock my socks.

That’s all for tonight. Now that midterms are over, I have no excuse to clean my borderline gross apartment. At least there’s no guests. For now. Starting tomorrow, I will make an effort to bring out my camera so it won’t be boring reviews on reddit every time.

Here’s something to finish with. Found this on reddit and thought it was the cutest thing ever.

Start Again Fresh

Alright, as inspired by Mr. Keaner after a solid 5 months of break, I resolve to resume blogging. To give a proper introduction to myself, I am Ivan, and I am a second year engineering student. I like math, science, technology, and ultimate frisbee. If you didn’t know that, you’re probably in the wrong spot and will not find much interest in this. This blog had its humble beginnings in an engineering design portfolio assignment, and if you see my previous posts you will see my feeble half-assed rants on the odd techy article I read while browsing reddit.

After engineering design was over I partied like a mofo and forgot about this for the whole summer. Okay, I lied. I actually didn’t party all day, I just sorta left this here. But I shall attempt to take back the habit of blogging after some consideration that I actually liked critiquing random technological advances. Furthermore, I  have found interest in science and writing (explaining science to the public, etc.) and at some point in my future life hope to something that blends writing and science and public (writing to a magazine, ideally).

Now for the your job.  You really don’t have much of a job, except just read if you feel like it. If you felt like reading, and you feel like commenting, please do so! I would love to hear from you. Most of you readers have me on Facebook, so I’m totally up for a discussion if you wish so. So without further ado, let the blogging begin!

Showcase: the Pain in Presenting, and the Epiphany

Showcase is done! And so this will be. This is now the last post of the blog. Just a reflection of showcase, I was sort of looking back and somehow, by some miracle, I was reminded of a lecture slide in praxis. But not just any lecture. The first lecture. In first semester. Shocking I know. I was just conveniently reminded of the comparison between a student and a graduating engineer thinking process, and the analysis/comm. etc. graph. Hm, I guess I’ll take the effort to find it and paste it here. Anyways, I realized that what this whole project does is a piece of that. The RFP provides a problem, and 20 groups think of a solution. Each idea has its own pros and cons, which are in essence an abstraction of ideas. Not gonna lie, the abstraction made in praxis did not contain much difference. When we arrive at the end of it, this presentation is the show of result of our multiple-except-not-really abstractions. Each of our own ideas had its own pros and cons, and with a little adjusting, we can combine a hefty amount of our ideas together to create a super-idea, which is contains as many pros and as least cons as possible. Now when we step back and look at the professional environment, engineers seek to create something that seeks to take in the greatest goods and the least amount of cons (that’s one aspect of engineering, at least). The engineer is basically doing what we all did, except he/she is one person and we are 250-ish. That’s beast.

With regards to how we did, I think we did pretty well. Although we didn’t stick to our pitch much except for the one time we were marked, I would say it would be pretty good. It’s good we got practice on some upper year engsci students. They poked holes at us first thing. By the time we got to marking along with other people of higher stature listening to our presentation it was pretty smooth. A lot of people liked our decision, and it felt like a job well done.
In terms of what we all did, through discussion we came up with the idea of modular layouts. From there, we split up the work. Thariq proceeded to do the 3D modelling and oversaw parts through the project, while Rishi worked on the brochure and poster, while Song worked on the design sketches and myself the program.

And that brings a conclusion to that.

Further notes on the Ever-Painful Programming

I am now almost officially done the program. Approximately 40 hrs has been put into it, and now I am almost done. The numbers are up but in separate components, and all that’s left is to combine and formalize the results. That’s 40 hrs of blood, sweat, tears, and (mostly) anger. I seriously cannot believe I made it through all that.
Debugging aside through, I found it was not the programming that was the hard part, but the numbers. The simulation is supposed to produce a final number that quantifies comfort rating of each seat. In essence, I basically create a metric system of which units I notoriously dub “comfort points.” The comfort points are the basis of the program.
To define comfort, I based it on two things: getting in the bus, and getting out. A person getting in the bus would have a comfort rating that is a function of the distance of people around you, whether it was a seat or space to stand, and the distance from the exit.
The getting out part was a pathfinding algorithm which navigated the person through other people and out of the bus. Comfort points would be deducted if you had to squeeze through 2 people to get out, or if you had to get out from the inside of a seat.
Now comes the tricky part. Determining the numbers. When someone tells me so-and-so is 6’4” I will immediately think, that person is tall. Because I know the metric system, and have experience with it. This is a different story. This metric system has been created for only this program, and will not be used anywhere else. So first comes the problem of throwing ballpark values. The seat/standing rating and the distance from other people rating I simply gave arbitrary numbers, but the distance from the exit was a tricky function. It required us just going on a bus and watching people move around and about for us to actually realize it. The distance from the exit had a more exponential effect as you went farther away from it. Furthermore, the relationship between the number of people and the decrease in comfort as you went away from the exit was exponential as well. To determine coefficients for these numbers, I determined them at their worst case scenarios. Being close to the exit with lots of people is worse than being far away from the exit and very few people. It is, after all, the long ride that counts. With that, I set the coefficients to have a greater weight on the number of people than the distance.
I went through similar justification and discussion with my group members in praxis for other aspects of the simulation. At many times we disputed, but I realized how I was unable to quantify anything, and I was just throwing in numbers, and simply hoped for the best. This has always been unsatisfactory to me, as I have always loved math and sought to calculate out numbers and such. However, we are given what we are given. I do not have any data or papers on the analysis of flow through a bus, nor will my work be anything cited. It is in essence, to prove our gut feeling, and it will do so in whatever means necessary.
I shall end on this note by stating the moral of the story: Praxis is bad for your health and will result in high blood pressure.

Seriously guys, I’m not Crying Wolf this Time

When the robot uprising comes around this time, I’m not saving anyone. Well I probably won’t be able to save anyone because of this, because I’ll be broke and homeless.
According to the news article, scientists have reached one step closer to developing AI. You know, like the ones in I, Robot. The robot named Adam, armed with reasoning and deduction algorithms, was given only the data of the movement of a double pendulum, and was able to deduce Newton’s Laws, along with the differential equation relationship with it. It did what scientists did in an instant.
It gets more interesting. Another news article surfaces (not gonna lie, I found this on reddit) on the infamous Adam. It was now given data on a yeast cell’s DNA, and it was able to determine which genes were responsible for which diseases. For reference, the human genome project took roughly 7 years of thousands of scientists working on it around the world to find out which genes did what. This is a phenomenal result in the world of AI.
However, it will soon be obsolete. This was merely a test subject. Its much more powerful counterpart, ironically named Eve, will take over Adam’s post and be put into searching for new medicines.
As much as this makes me feel completely obsolete and just raises the competition for a job, I’m happy for the advancement of science. The application to save lives with the use of technology is astounding.
It seriously kills me on what the reasoning algorithm is though. This is one of those ridiculously abstract programs that no one except the programmer has a hope of understanding. Just HOW?!

It has now officially become my life goal to own a Steinway Lyngdorf

I have just seen the craziest thing ever. While procrastinating and an interesting talk with Song on music, he showed me this link on the Steinway Lyngdorf speaker system. Here’s the Youtube video:
http://www.youtube.com/watch?v=Fz4XDU9NTp4 (fast forward to about 1:55)
“Where distinctive design meets intuitive functionality.”

No kidding. At a whopping 140 grand, you really pay for what you get. With more aluminum than a small sized car, and standing at 7 feet, it is a work of exceptional craftsmanship. However, what wow-ed me was not the range of the sound (20 – 22k Hz), but was their “Room-Perfect Technology.” Originally a 3D imaging software, it was tweaked and placed into the speaker system to dynamically the sound for that specific room. Essentially each speaker detects itself relative to the other one and to the room, and emits a sound wave in conjunction with the other to give it the best possible sound.
I’ve always been a big fan of sound. I have a 3-piece Altec Lansing speaker set here at my dorm, costing me about $160. When I went back home for Christmas break I almost died living with just my laptop speakers. A good investment in a nice full sound is always good. But anyways, back at my dorm, there’s this sweet spot right by my seat where the combination of bass and the 2 treble speakers are the best. See the picture. Anywhere else, however, and the base gets obnoxiously dominant. And if it’s a really bass-y song, then sometimes it gives me a headache from just listening to it. To have that eliminated through the detection of the room is pretty amazing, I’m not gonna lie.
It really stands by what it quotes. Distinctive design, intuitive functionality. We all know we intuitively want to eliminate the bass overdrive, and this design is just unforgettable. I really do wonder on how they came up with such an idea. Let’s take some image analysis software, and apply it to sounds. And oh, let’s also use sound to obtain the image. If you check out their pdf under innovation describing their research and what it does, it is simply stunning.
As far as this application goes for anything else, I don’t think it has any other application aside for top-of-the-line listening. Research says you can’t tell between 128 and 328 kbps music. With this, I am totally convinced you can. The engineering design is just innovative, as well as how it is presented in such a mouth-watering way. Definitely an investment of mine when I am not so broke.
http://www.steinwaylyngdorf.com/

Google: when in doubt, brute force

Once again, I was browsing reddit, and I happened to come across this. The blogger’s name is Douglas Bowman, a designer for Google, now leaving Google for Twitter. He describes the Google designer team. Google, the prestigious software company that owns like half the interwebz. And he’s leaving it. What is he, on crack? But seriously, when I read the inside of Google, I was shocked. Being the most prestigious company, they got the most prestigious people. But no engineers. Every design decision to them was horror. They would implement every subject and test it before making a permanent decision. One of the decisions was whether a border should be 3, 4, or 5 pixels. The implementation on that was horrid, and the justification on that would have been dead simple and one would have been simply chosen. Google, however powerful and effective the strategy is, takes no risk in any of its design decisions, which costs time in which its designers could be better off doing other things.
Bowman even states, that no one at the helm of the design team understands the principles and elements of design (he even has a Wikipedia link to the principles and elements of design). Google doesn’t trust their instinct, and they resort to boring mundane tasks which take up time better spent doing other things.
Bowman goes on to state some interesting things though. A quote:
“When a company is filled with engineers, it turns to engineering to solve problems. Reduce each decision to a simple logic problem. Remove all subjectivity and just look at the data. Data in your favor? Ok, launch it. Data shows negative effects? Back to the drawing board.”
Wow, engineering is really just that. And that’s so convenient. I think I lost interest in trying for a co-op job for Google.