Particle Swarm Optimization for Picture Analysis 90
Roland Piquepaille writes "Particle swarm optimization (PSO) is a computer algorithm based on a mathematical model of the social interactions of swarms which was first described in 1995. Now, researchers in the UK and Jordan have carried this swarm approach to photography to 'intelligently boost contrast and detail in an image without distorting the underlying features.' This looks like a clever concept even if I haven't seen any results. The researchers have developed an iterative process where a swarm of images are created by a computer. These images are 'graded relative to each other, the fittest end up at the front of the swarm until a single individual that is the most effectively enhanced.'"
The only problem... (Score:5, Informative)
P.S. IAAAIR (I am an AI researcher, albeit not in computer vision)
Re:The only problem... (Score:4, Informative)
Actually I think the biggest problem with any of these techniques is finding an algorithmic definition of 'fittest' and 'effectively', the rest can be solved by throwing money at the computation.
Not exactly comprehensive (Score:5, Informative)
Re:The only problem... (Score:3, Informative)
Nothing new to see here (Score:3, Informative)
They've reinvented genetic algorithms ?
Without seeing the details (read TFA but it's a summary and quite a bad one at that), I can't see why this would be better than a Bayesian optimisation with a photometric constraint. "The objective of the algorithm is to maximize the total number of pixels in the edges" sounds very, very simplified.
There are efficient ways of solving these things. Interesting that they invent an image processing algorithm but publish it in a non image processing journal - I wonder why that is ?
Re:The only problem... (Score:3, Informative)
Oh god, not another 'Bayesian methods for everything' guy..
Genetic algorithms have major advantages over other approaches. When designed well they are easy to code, and they can get tasks done as well as, if not better then the alternative techniques. I have a GA that can outperform a neural network on a particular task (not all tasks, just one very hard pattern recognition task, not going into it though, that would result in too long a post). It outperforms NN, and is so much simpler you wouldn't believe it. I was shocked to discover how much better it was, and I wrote it.
As for computational effort, well duh..
If it wasn't a task that needed a lot of computational effort, it would hardly be interesting, probably it would be in P or something.
ACO does tend to take a while, but in my experience, most really interesting GAs can take weeks to complete a single run. As a rule what your after is the finished result, and the time taken, provided it doesn't run for more than a few weeks, is usually not much of an issue.
Re:Bullshit FTA (Score:2, Informative)
Re:The only problem... (Score:3, Informative)
Mine first: you're right, GAs are easy to program, once you know the selection criteria. How do you have the computer select the best looking photo? Photoshop has for years had a feature where the computer will supply some altered images and let YOU pick the right one, but how do you give the computer a sense of esthetics?
Yours: GAs are great for finding finished products that you can then use. Both GAs and neural networks can turn up interesting algorithms etc. But they're not great algorithms for getting things done. I don't want to have to apply them to each photograph, I want to have them produce an algorithm or trained network or something of that nature that I can in turn apply to each photograph.