Hand Geometric Points Detection Competition


NEWS
9th Feb. 2011
- Publication of the submitted algorithms ranking.
9th Feb. 2011
- Testing data set and respective annotation file were published.
10th Jan. 2011
- Despite of today's deadline, you can still register and participate until the 4th of February.
10th Jan. 2011
- Publication of the list with the pre-registered competitors.
27th Dez. 2010
- Update of training data set and respective annotation file were published.
25th Dez. 2010
- Training data set and respective annotation file were published.

We would like to invite you to the first competition on Hand Geometric Points Detection.

foto      Hand recognition systems are among the oldest biometric tools used for automatic person authentication. Access control devices have been manufactured and commercialized since the late 1970s. Several patents have already been issued for hand recognition devices and live applications have been launched and used in nuclear plants, airports and hotels over the last 30 years. Palmprint recognition devices were used for security purposes in the Department of Energy, U.S. Naval Intelligence in the 1970s. However, hand biometry has become increasingly interesting in academic circles, mostly with the progress of computer vision research, only in the last decade.

Hand-based person recognition is reliable, low-cost and user-friendly. All in all, it is a viable solution for a range of access control applications. Other "nearest competitor" modalities are face, iris, fingerprint and retinal biometry. The face recognition alternative is another low-cost solution for access control. In contrast with these techniques, hand biometry offers some advantages: first, data acquisition is economical via commercial low-resolution scanners or cameras, and its processing is relatively simple; second, according to two public surveys, people like hand-based access systems, as they do not consider hand information to be as private as iris or fingerprints in daily applications. Hence, they find hand biometry to be less invasive and more convenient than other biometric modalities. Third, hand-based access systems are very suitable for indoor and outdoor usage and can work well in extreme weather and illumination conditions. Fourth, the hand features of adults are more stable over time and are not susceptible to major changes, except for injury or arthritis-based deformations. Finally, hand-based biometric information has proved to be very reliable, successfully recognizing people among populations of the order of several hundreds. Palmprint recognition can also be used to enhance the security of e-commerce and banking applications via integration in the conventional systems using PIN codes and passwords.

The novel aspect of this competition relies on the absence of any restrictions during image acquisition, i.e., there are no pegs, markers or restricted hand positions. Therefore, we believe that the possible outcome of this competition can be determinant to the improvement of the state-of-the-art in this field.