While this type of information is available to anyone who has access to the IP addresses of those that visit a site, I thought it was interesting enough to share these locations. The service appears to do a decent job at filtering out the IP addresses of known spiders and other automated agents, leaving behind mostly legitimate hits.

Many of my former colleagues from MLAB at FAU have been traveling abroad this summer, to places such as China (Beijing), Malaysia, Austria, Brazil (Curitiba), and Colombia. Hits from all of these locations show up. The largest concentration of hits is in the vicinity of Phoenix, AZ, although I’m not quite sure why (perhaps there’s an ISP based there that isn’t being filtered yet?).

It’s interesting to contrast the new ClustrMaps with the Guestmap I’ve had on my site for over six years (a link to the Guestmap is just below ClustrMaps in the sidebar). I only recently moved the Guestmap to the front page — it was previously buried on the About page. Unfortunately, the Guestmap only retains the most recent 100 entries, so many have been deleted. The Guestmap asks users to manually select their location and add an entry — only a fraction of visitors will do this. An example of a more modern implementation is Frappr.

A CAPTCHA is method of authenticating human users. A common CAPTCHA displays some text (usually distorted) and asks a human user to type the text. The intention is that while a human can easily recognize the letters, a computer would have trouble.

IMAGINATION is interesting in that it uses images, not text, to authenticate users. It exploits the semantic gap in order to differentiate humans from machines. The semantic gap is the disparity between a human’s description of a scene and the computational representation of the same scene. While a human may be able to look at an image and conclude that it’s a picture of “two happy people running on the beach”, this is far more difficult for a computer to accomplish.

An earlier effort to use images as CAPTCHAs is KittenAuth. This system presents the user with an array of 16 or so images from a limited number of categories. The user is asked to select images with a common theme (e.g. “select all images of cats”). However, this method has several weaknesses. First of all, the number of possible solutions is far less than those of a text CAPTCHA. Each image is a binary choice, rather than at least 26 choices for each text character (far more if case sensitivity and numerical digits are also includes). Also, the categories of the images used can be inferred by trial and error. In other words, it may be easier for an automated system to use brute force to compromise this type of CAPTCHA.

Microsoft’s Asirra project proposes a similar approach to KittenAuth, but with one notable difference – it includes a database of over three million images. The Asirra project was also inspired by HotCaptcha, which asks users to select three pictures of “hot” people, another task that is easier for humans than for computers, but may offend users.

The first proposed image CAPTCHA project was ESP-PIX from CMU. The system presents the user four images and asks the user to select a single word from a list that describes the images. Again, there is a possibility that a brute force attack would succeed.

The key challenge to designing an effective CAPTCHA is that while the test must be difficult for a computer to solve, it should be easy for a human to handle. Unfortunately, as OCR and other techniques to interpret text-based CAPTCHAs improve, the text becomes more distorted and more difficult for humans to interpret. Perhaps image interpretation tasks offer hope, if the number of possible solutions to an image CAPTCHA can be expanded without frustrating the user.

Image-based CAPTCHAs could also be used to improve image retrieval. The image CAPTCHAs already ask users to annotate or categorize images, but compare their results to complete ground truth. One possibility is to randomly impute uncategorized/unlabeled images into the system (without notifying the user). These images would not affect the CAPTCHA, but they would be categorized or annotated by human users. Once enough concurrent information is collected, the image could be added to the database.

The Perceptually-Relevant Image Search Machine (PRISM) is a new way to retrieve images. PRISM learns as you (and others) use the system, improving itself over time. The PRISM Game is a research experiment which transforms the PRISM interface into a fun game where players compete for the highest score by organizing related images.

You are invited to play this new game! The results of your play will allow us to thoroughly evaluate and improve the system. The PRISM Game will take approximately 15 minutes to complete. To play, go to:

http://mlab.fau.edu/prism/site/?c=showGame

Once the game is complete your score will be displayed and you will be asked to complete a very important survey on your experiences. Completing the survey is required to have your results included in the research.

PRISM is part of a dissertation underway at Florida Atlantic University. You are welcome to pass the message along and invite others to play the PRISM Game.

Thank you for supporting this research!

• Blackfield - “Blackfield II“: Not only one of my favorite albums of 2007, Blackfield II is perhaps one of the best CDs I have heard (quite a claim). Steven Wilson (Porcupine Tree) and Aviv Geffen return with a consistent disc. The music is fairly straightforward, highlighting piano, guitar, string arrangements, and a few synths. My favorite track is “My Gift of Silence”. The band also released “Blackfield - NYC“, a DVD of a live show in New York which is definitely worth checking out. The live performance’s variations on the songs make them more powerful than the studio versions.
• Porcupine Tree - “Fear of a Blank Planet“: This is the second release from a Steve Wilson on this list. The album is a bit less experimental than previous Porcupine Tree releases, but it is still a solid listen. The epic song “Anesthetize” (over 17 minutes) highlights the disc.
• Dream Theater - “Systematic Chaos“: I was let down by Dream Theater’s previous outings, “Train of Thought” and “Octavarium“, but “Systematic Chaos” is a great release. The album is bookended by two parts of “In the Presence of Enemies”, an interesting epic in Dream Theater’s now familiar style.
• Glass Hammer - “Culture of Ascent“: I was very pleasantly surprised by this album. All of Glass Hammer’s previous releases have been solid and I really enjoyed “The Inconsolable Secret“, but “Culture of Ascent” introduces new elements to the group’s sound. The album is heavier than previous releases with more prominent guitars. “South Side of the Sky” is an excellent cover of the classic Yes song. “Ember without Name” kept my interest for its entire duration (over 16 minutes) with a sounds that were unexpected but contribute to the song.
• Jimmy Eat World - “Chase This Light“: This was another surprising album in a genre I usually don’t listen to. The songs are generally consistent and upbeat. “Gotta Be Somebody’s Blues” is a fantastic song that is quite different from the others on the disc.
• Jordan Rudess - “The Road Home“: Rudess is the keyboardist for Dream Theater. This solo album consists of classic prog rock songs redone in Rudess’ style. Prominent prog musicians feature in most of the tracks. The highlights are Sound Chaser and Tarkus.
• Riverside - “Rapid Eye Movement“: Riverside is a prog rock band from Poland. “Rapid Eye Movement” is their third album and is a continuation of the themes introduced in the previous releases. The sound is heavier than in previous releases. 02 Panic Room stands out from the other tracks.

Honorable mentions go to Hayehudim - “Forte” and Symphony X - “Paradise Lost“.

FAU Salient image database (303MB .zip)
FAU Salient image database (303MB .tar.gz)

If you use this database in your work please cite the following article:

Marques, O.; Mayron, L.M.; Borba, G.B. & Gamba, H.R., “An attention-driven model for grouping similar images with image retrieval applications“, EURASIP Journal on Advances in Signal Processing, Special Issue on Image Perception, 2007, Article ID 43450, 17 pages.

@ARTICLE{marques:eurasip2007,
author = {Marques, Oge and Mayron, Liam M. and Borba, Gustavo B. and Gamba, Humberto R.},
title = {An Attention-Driven Model for Grouping Similar Images with Image Retrieval Applications},
journal = {EURASIP Journal on Advances in Signal Processing},
year = {2007},
volume = {2007},
pages = {Article ID 43450, 17 pages},
note = {doi:10.1155/2007/43450}
}