Andrey Vystavkin Blog

Recently, got myself into Usability fighting with the redesign of some web applications. The more advanced the project is becoming, the harder it comes to handle usability issues. Finally, it ended up with the full re-design having in mind key concepts from different web sources.

1. Proper Search

If the navigation sucks, it might be the only way how the user can find the required content. Statistically, the user uses search maximum twice, then he leaves the site if he didn’t find anything.

2.Design your Links

It’s quite hard to navigate over the complex website and afterword find yourself on the main page without a guess where you’ve been. So, do mark links as visited!

3. Banner site

With overwhelming number of banners, or any image/text advertisement of different features of the site/sites, do keep “an air” between the blocks, if you want to have lots of info on the front page, otherwise –   think again about your content placement. It’s really hard to read something when the rest of the content blinks/flashes/jumps/etc.  However, if you have to keep these jumping/flashing block – try to keep them far away from the content and each other.

4. Image data over Text and vice versa

The link is worth of a thousand words

5. And finally, don’t do this!

Yeah, this!

Design Design, Usability

There are a lot stereotypes about Soviet periods of Russia (and of course those republics as a part of Soviet Union), but what’s really hilarious and comes to a standstill when you see the architecture of that non-existent country.

Just check out some photos of 50’s and present time of Moscow State University, which was built in 1954. I’ve resized some of them via Flickr account, however you can find out more pictures here. The author tries to compare the composition of MSU environment planned in 50’s with chaotic ensemble of New York skyscrapers. You can use Google Translator to check the post.

IMHO,looks gorgeous!

Design, Links Architecture, Russia, Thoughts

Seems that a company leader for image graphics software – Adobe, is about to release a new language which will enhance the capacities of image filtering on the Web. Language’s codename is Hydra, and you can find it at Adobe Labs section. T

That’s what John Nack said about nice advanced technology:

“Hydra is tuned to run ridiculously fast on modern graphics cards (GPUs), and it’ll be tuned for multi-core CPUs as well…Here’s a key point, though: the same Hydra technology is being used to power the fast filters in After Effects CS3. Therefore an AE plug-in developer could effectively also develop runtime effects for Flash, while a Flash developer could leverage her work inside AE.”

Here are some benefits of Hydra language:

• Familiar syntax that is based on GLSL, which is C-based
• Allows the same filter to run efficiently on different GPU and CPU architectures, including multi-core and multiprocessor systems in a future update
• Abstracts out the complexity of executing on heterogeneous hardware
• Supports 3rd party creation and sharing of filters and effects
• Delivers excellent image processing performance in Adobe products

Some of the filtering examples you can find at Kevin Goldsmith’s blog.

Design, Gadgets Adobe, Hydra language, Image Processing, Max conference

I haven’t thought that it’s going to be such a headache to find some good software for Video and Audio Editing on Linux platform! At this type of software fields, you realize how much work should be done in future for Unices to reach Mac/Win multimedia platforms.

Win/Mac haters can take a deep breath, because there are no good alternatives for:

• Abode Premier
• Sony Vegas
• Sound Forge

However, I’ve managed to find the combination of Kinovd and Cinelerra. The last one appeared to be a nice video editor, even though some minor bugs arise (which were solved right away with Cinelerra helper). Here are some Ubuntu repositories for Cinelerra editor.

Enjoy editing!

Design, Gadgets Downloads, IT, Linux, Software, Ubuntu, Video

Working on one of my projects, I had to create some UML diagrams, not only for submitting it to the supervisors, but also for tracing the process of software development, and analysis stages.

I was really amazed trying to install new UML Modeler Plug-in for NetBeans IDE and seeing a nice shiny window:

“You can install the packages, but you won’t activate them!”

Just because of some dependencies incompatibility, their official website offered you a partial product: “Watch, but don’t touch!”

Hopefully, the development team decided to include some of those necessary plug-ins in the main part of NetBeans 6.0 version, so there would be no need of installing it again. From now on, I’ll have to switch to ArgoUML to finish up the stuff and patiently waiting for bug fixes of module dependencies

Design, Gadgets Development, IT, Java, Netbeans, UML

The semester slowly moves to its logical ending – the summer, and the only question I have is, if I’m going to use Matlab after finishing Data Image Processing lectures or not.

But since, I’m still using it, here is going to be one of the last projects for DIP lesson – offset changes of an image, with later on processing of histograms.

Comparing Matlab erode and dilate functions – offset procedure doesn’t effect the neighbouring pixels – all the actions are performed on each pixel respectively.

For changing offset value of an image we need to:

1. Create a matrix of an initial image;
2. Create a colour array for mapping pixels of the same colours (will be used for histogram);
3. Accept the offset value from the user;
4. And, of course, recalculate it all!

Since we’re going to work with grey scale image, we have up to 255 colours from black to white. Later on, I’ll use the picture which was taken close to the institute.


%Reading the image
Ia = imread('a.jpg');
disp('..Image "A" is loaded')
%Initializing the colour array

arrA = zeros([1 255]);

Since the image has 683×1024 size, we’re going to use the nested loop to match all the colours and store its pixels into the array:

for m=1: 683
for n=1: 1024
%Storing repetetive color indexes in color array.
for i=1: 255
%Placing colors into color arrays for 3 images respectively.

if i == Ia(m,n)
arrA(1,i) = arrA(1,i) +1;
end
end
end
end

After eliminating the matrix picture, and entering the offset value, we can create an offset.m function file, where the rest of the calculations will take place. For it, we’ll need to initiate a new matrix, of the same size, and another colour array, which will hold transformed pixel colours, and their quantities, to show the difference in the histograms:


for i=1: 683
for j=1: 1024
for k=1: 255
if k == Mtrx(i,j)
newArr(1,k) = newArr(1,k) + 1;
end
end
end
end
%...

disp('Resulting images...')
figure,imshow(y); title('Initial Image');
figure,imshow(Mtrx); title('Result of Offset.jpg');
disp('Histograms...')
figure,plot(z);title('Initial Histogram');
figure,plot(newArr); title('Resulting Histogram');

I’ve mentioned the main states of the calculation, but if you need to get really into the sources, simply download the source files, where three images have been processed, and enhanced with scaling function source files. For more complicated offset functionality, you may be interested in the offset filter made in Boston University.

Design, Gadgets Frederick University, Matlab