PART B: VALORISATION PLAN
|Please provide a short description of the state-of-the-art and/or current trends in the field? How does the result fit into it?|
|Physical access by a person to a certain ara may be allowed depending on payment, authorization, etc. These can be enforced by personnel such as a border guard, a doorman, a ticket checker, etc., or with a device such as a gate. An alternative of access control is a system of checking authorized presence, see e.g. Ticket controller on a train. Most of modern access control are automated in many different ways. The technology is now allowing some sophisticated ways to e.g. unlock doors. Some of them are based on biometrics that are methods for uniquely recognizing humans based upon one or more intrinsic physical or behavioral traits. Some examples of traits used in biometric are fingerprint, face recognition, DNA, palm print, hand geometry, retina scan or voice analysis.
All methods have goods and bads. Face recognition presents many advantages: contact-less, fast recognition with hats and sunglasses simple Integration, hands free, etc.
Eye-D is in the field of face recognition for access grant/control.
|What is the problem/need/knowledge gap that the research result is responding to? How was it addressed before?|
|Keywods: automatic video surveillance systems, face recognition, biometry access control
Face recognition is a computer technology that determines the locations, sizes and facial traits of human faces in digital images. Recognized traits may be compared to traits present in a database to relate them to the identity of a person.
In the past some evolutions of keys like magnetic badges, RFIDs have been used and are still in use. The related security is not as high as using a biometric system. Some systems are based on fingerprint or retina scan, face recognition may overcome some drawbacks of contact based biometric systems in certain environments (e.g. a construction site).
Many products are available on the market of face recognition access systems, some of them are listed below.
FIMS Access ( http://www.animetrics.com/Products/AnimetricsFIMS.html ) allows companies to use face as an identifying factor in access control. FIMS Access provides companies with a cost-effective biometric access control system that is non intrusive and easy to use. Verification of an authorized person is accomplished with high accuracy because of the Animetrics 2D to 3D technology used on every photo or video frame to be verified against a person’s enrollment.
FaceFirst ( http://www.facefirst.com/overview.html ) FaceFirst was built with the ambition of achieving global scalability. For every security or customer-identification goal, FaceFirst gives accurate alerts in seconds. FaceFirst is military-grade facial identification that communicates instantly with a vast central database, configured according to the user requirements. FaceFirst utilizes proprietary, patent-pending software that is built on Cognitec’s award-winning face-matching algorithm.
The Broadway 3D ( http://www.artecid.com/?gclid=CIq-0MiGhK0CFQKEDgod4WL35A ) is capable of visually identifying a walking or even running person regardless of their age or height. Identification takes a split second. The person simply glance at the device and it will “remember” and recognize the identity of one of the tens of thousands of previously registered people. Unlike human vision, however, the device is able to differentiate nuanced geometry with accuracy of up to fractions of a millimeter, enabling it to tell apart identical twins.
Responsive ( http://www.facerec.com/index.html ) software “finds” the user’s face within a frame and starts making comparisons instantly. Sensors don’t need to be adjusted for height; spectacles and safety gloves don’t need to be removed; operatives don’t need to touch a scanner or present their face in a particular way. Verification takes approximately 1.5 seconds. Successfully deployed in challenging real-life environments, overcoming usual biometric obstacles such as dust, dirt, grease, variable lighting and user co-operation.
|What is the potential for further research?|
|The potential is represented by the discover of new cases in which current algorithms are showing poor performance in terms of false alarms or fast response. That cases are not easy to be built artificially and may come with the application of the algorithms on a large number of situations coming from the real usage.
|What is the proposed method of IPR-protection? (patent, license, trademark etc.)|
|The research result is a software and it is not a stand-alone product.
At the moment two possibilities are envisaged: copyright or patent. The patent will be considered if the result will become part of a complete “machine” usable as a stand-alone product for access control.
|What are the steps that need to be taken in order to secure the IPR-protection? What is the cost of IPR-protection?|
|About the protection of the algorithm, the copyright could be considered.
Copyright protection by the registration of the software on the Public Register for computer programs has a cost starting from 120 €.
A similar level of protection can be made by the researcher by sending the software description and the software itself on a CD in a closed envelop by certified mail to himself. The stamp certifies the date of release of the software and can be used against claims of originality or to claim originality.
If the patent issue has to be carefully investigated. Infact, lots of patents related to the project keywords are shown by common patent search engines. Some examples are reported at the following links.
Existing patents cover both methods for face recognition and machines to be used in face recognition e.g. to control the access or the presence of a person.
The patent costs in Italy are around 2500 € (up front) + increasing annual fees from 60€ up to 650€ in the 20 year period.
|What is you overall assessment of the scientific maturity of the research result?|
|At scientific level the research result is at good level of maturity. Some issues may arise with existing (and possibly patented) products or methods.
KEYWORDS QUANTITATIVE ASSESSMENT (0-5).
|Please put X as appropriate.||1||2||3||4||5|