You are currently browsing the tag archive for the ‘Image Segmentation’ tag.

Portugal at World Expo 1998 by Vitorino Ramos

Images – Portugal (1A – top left, original input satellite image below), geodesically stretched by one of my Mathematical Morphology algorithms, in order to represent real travel times from each of the 18 regional districts in Portugal, to the rest of the territory.  From the 18, three capital districts are represented here. As departing from Lisbon (1B – top right), from Faro (1C – South of Portugal, bottom left), and from Bragança (1D – North-East region, bottom right). [World Exposition, Lisbon, Territory pavilion, 1998].

Recently one of my colleagues who knows I love maps, pointed me to an old TV show “Câmara Clara“, a cultural TV show by RTP2, at one of the main public Portuguese TV stations. Main reason for my interest was his current theme: Maps. My second reason was their guests: Joaquim Ferreira do Amaral (an ex-Minister with a passion for maps) and Manuel Lima, which wonderful work on information visualization I know for a long time (on one of my past posts I referred to one of his ongoing working sites: visualcomplexity).
 

For my complete and positive surprise, their interview ended with some new examples, being one of my old works referred (from 57m 12s up to 60m 26s on http://camaraclara.rtp.pt/#/arquivo/131 ). It’s a long story on how I ended doing these kind of maps. Part of it, it’s here. During 1998, the World Exposition was in Portugal, and I got invited to present a set of 18 different maps from the Portuguese territory. So I decided to geodesically stretch the travel distances from any of the 18 different capital districts, to the rest of the territory, in order to represent travel Time not Distance, or Distance as time. For that,  I have coded new algorithms based on Mathematical Morphology (MM), taking in account every road (from main roads to regional, check some images below), from which I applied different MM operators.

Unfortunately, many of those maps are now lost. I did tried hard to find them from my old digital archives, but only found those above, which represent the departure from Lisbon (the Capital), Faro and Bragança. So, if by any reason you happen to have some photos from the 1998’s World Exposition in Lisbon, inside the Territory pavilion, I would love to receive them.

Os Portugueses e a Arte dos Mapas - Câmara Clara 131 - Maio 10 2009Video (LINK) – “Câmara Clara” TV show by journalist Paula Moura Pinheiro dedicated to maps (nº 131), at one of the main public Portuguese TV stations (RTP2), broadcasted on May 3 2009, in Portuguese.

A sketchy summary of this TV program went on something like this (the poor translation is mine): At the year Google promises to launch his first and exhaustive world-wide open-access digital cartography of the African continent, Joaquim Ferreira do Amaral, passioned by the Portuguese World Discover History and collector of historical maps, joins as guest with Manuel Lima, the Portuguese information designer that recently Creativity magazine has considered one of the top bright minds along with Google and Amazon founders, debating the importance of “navigating” reality with a map. From the Portuguese cartographic history, know to be the best in the XV and XVI centuries, up to the actual state-of-the-art in this area, from which Manuel Lima is considered to be one of the top researchers at global scale.

Original + Layers Portugal at World Expo 1998 by Vitorino Ramos
Advertisements

Kitaoka colour illusion

Fig. – Illusion created by Prof. Akiyoshi Kitaoka (Dep. of Psychology, Ritsumeikan Univ., Kyoto, Japan). If you don’t see any illusion at all, don’t worry. That’s exactly why this optical illusion is so great. The illusion is not there, or is it?! Meanwhile over his page, Akiyoshi warns: This page contains some works of “anomalous motion illusion”, which might make sensitive observers dizzy or sick. Should you feel dizzy, you had better leave this page immediately (more).

Where’s the illusion, right? Well,… what if I just tell you that no blue at all is used over this picture! No matter how strongly you want to believe you are seeing blue and green spirals here, there is no blue color in this image. There is only green, red and orange. What you think is blue is actually green. Don’t worry, … you are not daltonic. I mean, I’m a little bit but, you could check this out through Paint Shop Pro or Photoshop, if you need an affirmation. Indeed, these are just “Vain speculation un­deceived by the senses” (1670’s Scilla’s treatise) .

In fact, Relations here, between different colors (green, red and orange), are more important than each color by itself. Relations plus context are the key (more here over Generative Art, and here over Swarm Intelligence based Pattern Recognition). Through these relations, much probably using Gestalt‘s principles (the German word Gestalt could be translated into “configuration or pattern”), here Akiyoshi manages to emerge us the blue color over our perception. This does not cheat a computer of course, however could cheat our own eyes. In other areas the opposite could also be found. For instance, Humans can easily recognize a car over background trees (segment it, in just tiny lapses of a second), while this natural task could be extremely painful for computers over some cases (here is one example).

Born in Prague (inspired by 1890’s works of Christian von Ehrenfels, Austrian philosopher), then later absorbed by a great and tremendous intellectual period occurred from Germany back to Austria (Bauhaus), the Gestalt Laws of Organization have guided the study of how people perceive visual components as organized patterns or wholes, instead of many different parts. I would say that most certainly some Wertheimer’s gestaltic principles were used in here: Figure and Ground, Similarity, Proximity or Contiguity, Continuity, Closure, Area, and Symmetry (check Gestalt Theory of Visual Perception). We could see this happening also in other areas, … in Music for instance:

[…] Gestalt theory first arose in 1890 as a reaction to the prevalent psychological theory of the time – atomism. Atomism examined parts of things with the idea that these parts could then be put back together to make wholes. Atomists believed the nature of things to be absolute and not dependent on context. Gestalt theorists, on the other hand, were intrigued by the way our mind perceives wholes out of incomplete elements [1, 2]. “To the Gestaltists, things are affected by where they are and by what surrounds them…so that things are better described as “more than the sum of their parts.” [1, p. 49]. Gestaltists believed that context was very important in perception. An essay by Christian von Ehrenfels discussed this belief using a musical example. Take a 12 note melody. Play it in one key, say the key of C. Now change to another key, say the key of A flat. There might not be any notes the same in the two songs, yet a person listening to it knows that it is the same tune. It is the relationships between the notes that give us the tune, the whole, not which notes make up the tune. […], from “Gestalt Principles of Perception“, Bonnie Skaalid, Univ. of Saskatchewan, Canada, 1999.

Care for an contemporary example? Well, … the first thing that comes to my mind is DUB music genre. In fact, I do have several albums from different musicians over my house. Dub music evolved in Jamaica (1968) from early rastafarian instrumental reggae music and versions that incorporated fairly primitive reverbs and echo sound effects, being found by accident (engineer Byron Smith left the vocal track out by accident). Over decades, it inspired immense groups of musicians from well-known bands such as The Police, The Clash, UB40 up to reputed musicians such as Bill Laswell. Of course !, it was not far from what John Cage have made for the solo piano Music of Changes, to determine which notes should be used and when they should sound. In the fifty’s, Cage start it to use the mechanism of the I Ching (Chinese “Book of Changes”) in the composition of his music in order to provide a framework for his uses of chance.

Other most recent bands include, Leftfield, Massive Attack, Bauhaus, The Beastie Boys, Asian Dub Foundation, Underworld, Thievery Corporation, Gorillaz, Kruder & Dorfmeister, and DJ Spooky. But what is then so special about Dub? Well, one of this genre’s most striking features is the fact that some if not all musical sentences are incomplete. Those special sentences (Gestaltic, let me add), are normally followed by a pause. The most amazing thing however, is that us, Humans could perceive the entire sentence being formed on the back of our minds! So the music is not there, and at the same time, we are listening to two adjacent simultaneous melodies, as we were a composer. By just using relations among a few notes, we soon start to emerge a perception for the whole sentence, as if they were self-organizing! Being it extremely rhythmic, this often could lead us to a sweet soft state of overwhelming emotion, or exalted organic feel to the music .

As you will probably know by now, the same could happen over misplaced letters over an entire phrase. Even if some letters are not at their right proper place, at each word, we could still perceive the whole sentence meaning. Up to your gestaltic neurons to decipher.

Next time you go to a rave party (I never did, neither pretend to), do think about the title of this post, the figure above, as well as on all those great past musicians, along with – unfortunately – awkward current DJ’s, who pass on for hours strident music mixes without knowing at all what Gestalt is all about! Oh, … by the way, should you feel extremely dizzy, do follow Akiyoshi’s advice: If you start feeling unwell when using this website (rave party), immediately cover one eye with your hand and then leave the page (leave the party). Do not close your both eyes because that can make the attack worse!

Ants_Movie

Transition behavior of one Artificial Ant Colony in presence of a sudden change in his artificial digital image Habitat, between two different Digital Grey Images (face of Einstein and a Map). Created with an Artificial Ant Colony, that uses images as Habitats, being sensible to their gray levels [in, V. Ramos, F. Almeida, “Artificial Ant Colonies in Digital Image Habitats – a mass behavior effect study on Pattern Recognition“, ANTS’00 Conf., Brussels, Belgium, 2000].

After “Einstein face” is injected as a substrate at t=0, 100 iterations occur. At this point you could recognize the face. Then, a new substrate (a new “environmental condition”) is imposed (Map image). The colony then adapts quickly to this new situation, losing their collective memory of past contours.

In white, the higher levels of pheromone (a chemical evaporative sugar substance used by swarms on their orientation trough out the trails). It’s exactly this artificial evaporation and the computational ant collective group synergy reallocating their upgrades of pheromone at interesting places, that allows for the emergence of adaptation and “perception” of new images. Only some of the 6000 iterations processed are represented. The system does not have any type of hierarchy, and ants communicate only in indirect forms, through out the successive alteration that they found on the Habitat. If you however, inject Einstein image again as a substrate, the whole ant society will converge again to it, but much faster than the first time, due to the residual memory distributed in the environment.

As a whole, the system is constantly trying to establish a proper compromise between memory (past solutions – via pheromone reinforcement) and novel ones in order to adapt (new conditions on the habitat, through pheromone evaporation). The right compromise, ables the system to tackle two contradictory situations: keeping some memory while learning something radically new. Antagonist features such as exploration and exploitation are tackled this way.

Image Classification of Shellfish Larvae Digital Images using Swarm Intelligence. On the left a compendium of 9 raw images (out of 20 samples) used in the present study. Respective segmented images on the rigth.

Image Classification of Shellfish Larvae Digital Images using Swarm Intelligence. On the left a compendium of 9 raw images (out of 20 samples) used in the present project. Respective segmented images on the rigth.

[] Vitorino Ramos, Jonathan Campbell, John Slater, John Gillespie, Ivan F. Bendezu and Fionn Murtagh, Swarming around Shellfish Larvae Images, in WCLC-05, 2nd World Congress on Lateral Computing, Bangalore, India, 16-18 Dec., 2005.

The collection of wild larvae seed as a source of raw material is a major sub industry of shellfish aquaculture. To predict when, where and in what quantities wild seed will be available, it is necessary to track the appearance and growth of planktonic larvae. One of the most difficult groups to identify, particularly at the species level are the Bivalvia. This difficulty arises from the fact that fundamentally all bivalve larvae have a similar shape and colour. Identification based on gross morphological appearance is limited by the time-consuming nature of the microscopic examination and by the limited availability of expertise in this field. Molecular and immunological methods are also being studied. We describe the application of computational pattern recognition methods to the automated identification and size analysis of scallop larvae. For identification, the shape features used are binary invariant moments; that is, the features are invariant to shift (position within the image), scale (induced either by growth or differential image magnification) and rotation. Images of a sample of scallop and non-scallop larvae covering a range of maturities have been analysed. In order to overcome the automatic identification, as well as to allow the system to receive new unknown samples at any moment, a self-organized and unsupervised ant-like clustering algorithm based on Swarm Intelligence is proposed, followed by simple k-NNR nearest neighbour classification on the final map. Results achieve a full recognition rate of 100% under several situations (k =1 or 3).

(to obtain the respective PDF file follow link above or visit chemoton.org)

[] Vitorino Ramos, Filipe Almeida, Artificial Ant Colonies in Digital Image Habitats – A Mass Behaviour Effect Study on Pattern Recognition, Proceedings of ANTS´2000 – 2nd International Workshop on Ant Algorithms (From Ant Colonies to Artificial Ants), Marco Dorigo, Martin Middendorf & Thomas Stüzle (Eds.), pp. 113-116, Brussels, Belgium, 7-9 Sep. 2000.

Figure - Transition behaviour of one Artificial Ant Colony in presence of a sudden change in his artificial digital image Habitat, between two different Digital Grey Images. Created with an Artificial Ant Colony, that uses images as Habitats, being sensible to their gray levels. At the second row, Kafka image is replaced as a substrate, by Red Ant. In black, the higher levels of pheromone (a chemical evaporative sugar substance used by swarms on their orientation trought out the trails). It’s exactly this artificial evaporation and the computational ant collective group sinergy realocating their upgrades of pheromone at interesting places, that allows for the emergence of adaptation and perception of new images. Only some of the 6000 iterations processed are represented. The system does not have any type of hierarchy, and ants communicate only in indirect forms, through out the sucessive alteration that they found on the Habitat.

Figure - Transition behaviour of one Artificial Ant Colony in presence of a sudden change in his artificial digital image Habitat, between two different Digital Grey Images. Created with an Artificial Ant Colony, that uses images as Habitats, being sensible to their gray levels. At the second row, "Kafka" image is replaced as a substrate, by "Red Ant". In black, the higher levels of pheromone (a chemical evaporative sugar substance used by swarms on their orientation trought out the trails). It’s exactly this artificial evaporation and the computational ant collective group sinergy realocating their upgrades of pheromone at interesting places, that allows for the emergence of adaptation and "perception" of new images. Only some of the 6000 iterations processed are represented. The system does not have any type of hierarchy, and ants communicate only in indirect forms, through out the sucessive alteration that they found on the Habitat.

Some recent studies have pointed that, the self-organization of neurons into brain-like structures, and the self-organization of ants into a swarm are similar in many respects. If possible to implement, these features could lead to important developments in pattern recognition systems, where perceptive capabilities can emerge and evolve from the interaction of many simple local rules. The principle of the method is inspired by the work of Chialvo and Millonas who developed the first numerical simulation in which swarm cognitive map formation could be explained. From this point, an extended model is presented in order to deal with digital image habitats, in which artificial ants could be able to react to the environment and perceive it. Evolution of pheromone fields point that artificial ant colonies could react and adapt appropriately to any type of digital habitat.
 

 

(to obtain the respective PDF file follow link above or visit chemoton.org)

 

 

 
 
 
 
 
 

 

[...] People should learn how to play Lego with their minds. Concepts are building bricks [...] V. Ramos, 2002.

@ViRAms on Twitter

Archives

Blog Stats

  • 245,965 hits