 |
|
Machines can't replicate human image recognition, yet
September 10, 2009While computers can replicate many aspects of human behavior, they do not possess our ability to recognize distorted images, according to a team of Penn State researchers.
"Our goal is to seek a better understanding of the fundamental differences between humans and machines and utilize this in developing automated methods for distinguishing humans and robotic programs," said James Z. Wang, associate professor in Penn State's College of Information Sciences and Technology.
Wang, along with Ritendra Datta, a Penn State Ph.D. recipient, and Jia Li, associate professor of statistics at Penn State, explored the difference in human and machine recognition of visual concepts under various image distortions.
The researchers used those differences to design image-based CAPTCHAs (Completely Automated Public Turing Test to Tell Computers and Humans Apart), visual devices used to prevent automated network attacks.
Many e-commerce web sites use CAPTCHAs, which are randomly generated sets of words that a user types in a box provided in order to complete a registration or purchasing process. This is done to verify that the user is human and not a robotic program.
In Wang's study, a demonstration program with an image-based CAPTCHA called IMAGINATION was presented on imagination.alipr.com. Both humans and robotic programs were observed using the CAPTCHA.
Although the scope of the human users was limited, the results, presented in the September issue of IEEE Transactions on Information Forensics and Security, proved that robotic programs were not able to recognize distorted images. In other words, a computer recognition program had to rely on an accurate picture, while humans were able to tell what the picture was even though it was distorted.
Wang said he hopes to work with developers in the future to make IMAGINATION a CAPTCHA program that Web sites can use to strengthen the prevention of automated network attacks.
Even though machine recognizability does not exceed human recognizability at this time, Wang says that there is a possibility that it will in the future.
"We are seeing more intelligently designed computer programs that can harness a large volume of online data, much more than a typical human can experience in a lifetime, for knowledge generation and automatic recognition," said Wang. "If certain obstacles, which many believe to be insurmountable, such as scalability and image representation, can be overcome, it is possible that one day machine recognizability can reach that of humans."
Penn State
The researchers used those differences to design image-based CAPTCHAs (Completely Automated Public Turing Test to Tell Computers and Humans Apart), visual devices used to prevent automated network attacks.
Many e-commerce web sites use CAPTCHAs, which are randomly generated sets of words that a user types in a box provided in order to complete a registration or purchasing process. This is done to verify that the user is human and not a robotic program.
In Wang's study, a demonstration program with an image-based CAPTCHA called IMAGINATION was presented on imagination.alipr.com. Both humans and robotic programs were observed using the CAPTCHA.
Although the scope of the human users was limited, the results, presented in the September issue of IEEE Transactions on Information Forensics and Security, proved that robotic programs were not able to recognize distorted images. In other words, a computer recognition program had to rely on an accurate picture, while humans were able to tell what the picture was even though it was distorted.
Wang said he hopes to work with developers in the future to make IMAGINATION a CAPTCHA program that Web sites can use to strengthen the prevention of automated network attacks.
Even though machine recognizability does not exceed human recognizability at this time, Wang says that there is a possibility that it will in the future.
"We are seeing more intelligently designed computer programs that can harness a large volume of online data, much more than a typical human can experience in a lifetime, for knowledge generation and automatic recognition," said Wang. "If certain obstacles, which many believe to be insurmountable, such as scalability and image representation, can be overcome, it is possible that one day machine recognizability can reach that of humans."
Penn State
'Thirst for knowledge' may be opium craving
Neuroscientists have proposed a simple explanation for the pleasure of grasping a new concept: The brain is getting its fix.
Optical Breakthrough Enables Vehicle Occupancy Monitoring To Ease Jams
Infra-red cameras that automatically count people in cars could soon be a feature on the UK's motorways, making it easier to enforce priority lanes for car sharing to ease congestion and cut journey times. The unique patented technology to detect human faces in moving cars without distracting drivers was developed by Laser Optical Engineering (LOE), a spin out company from Loughborough University. Together with commercial, research and civic partners, it has developed a prototype camera system with Department for Transport and Engineering and Physical Sciences Research Council funds. The need for such a system arose following the launch in 1998 by Leeds City Council of Britain's first prior
Fingerprint recognition gets true `Fingerspitzengefühl`
Will we pay using our fingerprint, or enter a building just touching a sensor? Does our mobile phone recognize our fingerprint? It is possible, as far as Dutch PhD student Asker Bazen is concerned. He has improved the verification techniques, resulting in a better result even for deformed and damaged prints. Together with a higher speed, the new methods can take away existing reserves for implementing fingerprint verification. Bazen is finishing his PhD research at the faculty of Electrical Engineering of the University of Twente, on September 18. Of all biometric methods, fingerprints are the most practical, is the conviction of Bazen: it can be done with a simple sensor that is coupled in
More Image Recognition Current Events and Image Recognition News Articles
 |
Image Recognition and Classification (Optical Engineering, 78) by Bahram Javidi (Author) Presents an overview of current advances in the field and details the latest image processing algorithms and imaging systems for image recognition with diverse applications to the military, the transportation, aerospace, information security, and biomedical industries, radar systems, and image tracking systems. |
 |
Mathematics of Digital Images: Creation, Compression, Restoration, Recognition by S. G. Hoggar (Author) This is a major revision of the author's successful book Mathematics of Computer Graphics. It still focuses on foundations and proofs, but now exhibits a shift towards digital image compression, restoration, and recognition. Topology is replaced by Probability and Information Theory (with Shannon's source and channel encoding Theorems) which are used throughout. Several fractal methods are given in the service of Compression, along with linear transforms (hence FFT, DCT, JPEG, wavelets etc), recent neural methods, and the ubiquitous vector quantisation. Optimising for the Human Visual System is a subtheme. The superiority of Pyramid methods with respect to entropy is proved. Restoration offers convolution/deconvolution against noise and blurr. Recognition explores not only the Hough and... |
 |
Rockin' the Corps Starring: Louie Anderson, Cedric the Entertainer, Godsmack, The Hooter Girls, Randy Jackson Directed By: Daniel E. Catullo, Lawrence Jordan Also With: Daniel E. Catullo (Producer), A.J. Dickerson (Editor), Ashley Hughes (Editor), Jack Gulick (Producer), Spencer Proffer (Producer), Tima Surmelioglu (Producer) America's top rock 'n' rollers, comedians, celebrities and cheerleaders join together to pay tribute to the U.S. Marines returning from Iraq during this fun-filled event featuring music, laughs and countless stirring moments! Over 50,000 Marines, sailors and family members attended this once-in-a-lifetime celebration at Del Mar Beach, Camp Pendleton. Track Listing: Star Spangled Banner - Ted Nugent Awake - Godsmack Whatever - Godsmack I Stand Alone - Godsmack Cat Scratch Fever - Ted Nugent w/ Godsmack Stranglehold - Ted Nugent w/Godsmack Livin' On A Prayer - Richie Sambora It's My Life - Richie Sambora Wanted Dead Or Alive - Richie Sambora Only Wanna Be With You - Hootie & The Blowfish Time - Hootie & The Blowfish Hold My Hand - Hootie & The Blowfish Soldier... |
 |
Sony Cyber-shot DSC-T900 12.1 MP Digital Camera with 4x Optical Zoom and Super Steady Shot Image Stabilization (Black) by Sony "The Sony Cyber-shot DSC-T900 camera combines style with power and intelligence. At just over 5/8” (16.3mm) thin, the T900 comes in 4 sophisticated colors and has a stylish look. With the T900 taking great pictures is now easier than ever. The iAuto mode goes beyond the traditional auto mode, thinking and working for you; recognizing scenes, lighting conditions, faces, and automatically adjusts camera settings resulting in clearer images, and faces with more natural skin tones and less blur. T900 also features HD Movie mode, recording in 720p and stereo sound for capturing your precious moments on video. Compose, review, and share your memories on the dazzling 3.5” high-resolution (921k pixels) touch-screen LCD. The DSC-T900 offers a step up in features and capabilities from the T90... |
 |
Mega Brain Boost by Majesco Sales Inc. Now you can boost your brain power with 3 full games on 1 cartridge! The brain produces five wavelengths: alpha, beta,delta, gamma, and theta waves. In Mega Brain Boost you get both Brain Boost Gamma Wave and Brain Boost Beta Wave games plus an entirely new game! All three games utilize Makoto Shichida's "right brain development theory" to train your brain. |
|
ESKER FAX WORKGROUP-IMAGE RECOGNITION MOD ( 15614 ) by Esker Inc | |
 |
Historic Print (S): The first recognition of the American flag by a foreign government by Olde Yankee Map and Photo Shoppe This is a museum quality, reproduction print on premium paper with archival/UV resistant inks. The framed work is single matted (ivory), under acrylic glass, with a hanging wire. Date: c1916. Subject: Notes: Halftone repro. of painting by Edward Moran. Publ. Ind.; B.I.; U.S. Hist.--1778; Ships; Flags; Navy, U.S. Format: SOURCE: Library of Congress |
 |
Recognition Ctrl (Primary Contributor) |
 |
Destiny Starring: Rudolf Klein-Rogge, Paul Biensfeldt, Lil Dagover Directed By: Fritz Lang |
 |
Practical Algorithms for Image Analysis with CD-ROM by Lawrence O'Gorman (Author), Michael J. Sammon (Author), Michael Seul (Author) In classic "cookbook style," this book offers guided access for researchers and practitioners to techniques for the digital manipulation and analysis of images, ranging from the simplest steps to advanced functions. Drawing on their long experience as users and developers of image analysis algorithms and software, the authors present a description and implementation of the most suitable procedures in easy-to-use form. Each self-contained section treats a single operation, describing typical situations requiring that operation and discussing the algorithm and implementation. Sections start with a "before" and "after" pictorial example and a ready-reference listing typical applications, keywords, and related procedures. This new edition has additional sections on Gabor filtering and on... |
| © 2009 BrightSurf.com |