Technical feasibility

PART B: VALORISATION PLAN

SECTION I: Testing

 

1

Has the R & D result been tested?

YES

X

NO

 

 

The following question is replied according to the reply in question 1

 

If yes

 

1a

In what mode has the result been tested?

•             Prototype

•             Pilot Application

•             Alpha/BETA testing

 

Face recognition is a branch of computer vision – the science and technology that allow machines to “see”.  The meaning of “see”, in this case, is that the machine is able to extract information from an image. That information is used to execute some task. The information is extracted from a digital image by the way of real-time and optimized algorithms that have to be extremely robust when used in context as e.g. access control.

Eye-D is a face recognition software specialized to the task of detecting the face of a person moving in a given environment and check if its identity correctly matches the biometric information stored in a database.

Eye-D grants access to protected areas to recognized personnel and logs enter/exit time thus providing an effective way of enforcing safety and security in sensible working environments (e.g. hazardous areas or machinery access, restricted areas monitoring, etc.).

At the current level of development the software is available on PC/Windows platform and has been tested in a laboratory context and during some “on the field” trials. The trials have been performed with the assistance of researchers.

This is the reason why the R&D result is considered at a prototype level even if it has been fully tested in a subset of environment conditions.

 

1b. Please describe and discuss the testing result
The software has been positively tested in a controlled environment: the laboratory and during field trials. The functionality of the algorithm is satisfactory when the recognition is limited to a limited set of possible faces (in the order of 30) and in controlled light conditions. If the set of faces present in the database increases, the time needed for recognition could result longer than expected in some applications.

The research result need to be tested in real operation conditions and in connection with a statistically significant number of events to characterize also the frequency of false alarms.

 

If no:

 

1c

Describe what type of testing does the R&D result need?
N/A

 

 

 

1d. What is the time needed for testing?
N/A
1e. What is the cost needed for testing?
N/A

 

 

 

 

 

 

SECTION 2: Current Stage of Development

 

2a

To what extent does the development team have technical resources for supporting the production of a new product? (Researchers, human resources, hardware, etc. )
In the case of Eye-D the product is an algorithm. The algorithm need to be integrated in an hardware dedicated to a certain application. The company is open to different options regarding that integration. The first option is to provide a software package to be installed in a commercial off the shelf (COTS) PC connected to COTS video cameras. In that case the final user is responsible for the hardware and the software should released with a compatibility list or a set of requirements for PC and cameras.

The second option is to provide the SW to a video camera manufacturer interested in the integration of the algorithm directly in the camera firmware embedding the face recognition in the camera hardware. In that case a process of porting and integration of the software should be planned in conjunction with the partner.

The third option is to integrate the algorithm in existing access or surveillance systems providing an augmented level of security.

About the first option, the convergence to a product need some engineering activities. The first of them is the characterization of the system with respect to different cameras, focusing more on low cost cameras (e.g. webcams) to make the product appealing from the costs viewpoint.  That activity will help to write the requirements for the HW platform to be used. Another issue to consider is the choice of the focal length to be used in the optical part of the camera, that length could be modified to optimize the performance of the algorithms in different situations. Eventually, to complete the performance picture, it is necessary to execute some some stress tests in function of light conditions, number of faces in the database, to allow the measure of some figures like time necessary for a positive recognition, probability of false recognition, probability of failed recognition.

Regarding the first option, the company envisages the need of one researcher/engineer dedicated to the activity over a 3-6-months period.

Currently that resources in not allocated on the project.

About the second and the third options, the activities have to be planned as a function of the partnership with third party already working in that market. At the moment, the related effort  in not predictable since there are no external partners already involved in the project.

 

 

2b

What are the technical issues that need to be tackled for full deployment, if needed?
From the software viewpoint, the main issues to be faced are related to the use of a different digital camera and to the different light conditions that may occur in the final application.

 

 

2c

What additional technical resources are needed for the production of this new product?
The main technical issue to be faced is the physical integration with the extra hardware  in a real environment. To perform that part of validation it is necessary to explore the possibility to operate in conjunction with a company that could provide a commercial solution for that hardware.

 

2d

Overall assessment of the current stage of technical development.
The steps needed to complete the development are mainly about some engineering activities followed by test and validation activities. The test and validation activity provides a feedback to developers for bug fixing, performance and robustness improvements.

Provided that the research result is a software that need some extra validation, and given the applicable resources and the experience of the team, it can be stated that the R&D result is in a advanced stage of development. The integration with a real system, considered necessary to complete the validation, has not been started yet.

 

SECTION 3: Deployment

3a

Define the demands for large scale production in terms of
  • Materials
Negligible for the research result, the application in a product has to be made in co-operation with a third party responsible for the production. The materials may be categorized as electronic equipments (mostly COTS PCs and cameras in some cases embedded computers + cameras or simply “advanced” cameras).

 

  • technologies, tools, machineries
As for the previous point.

 

 

 

  • Staff effort
Given the integration on a system made before the deployment, the staff may be involved in some “after sales” activities as bug fixing or some tuning on special cases not foreseen during validation.

 

 

 

SECTION 4: Overall Assessment

 

1

What is you overall assessment of the technical feasibility of the research result?
The challenge is complete the characterization in a real scenery, using COTS hardware, and tune the software in reasonably defined conditions.

In the case described at point 2a as “first option”, few man months are necessary to reach a commercial stage.

Different is the situation described at point 2a as “second” and “third option”. In that case the test environment is not available at the moment, the main dependency is that some third parties are still to be found out.

Provided the completion of the validation, the the level of confidence about feasibility is considered good.

 

KEYWORDS QUANTITATIVE ASSESSMENT (0-5).

 

Please put X as appropriate. 1 2 3 4 5
Adequacy of testing activity undertaken so far     X    
Adequacy and availability of technical resources of the development team       X  
Current development stage       X  
Overall technical feasibility       X  

 

Bookmark the permalink. Follow any comments here with the RSS feed for this post. Post a comment or leave a trackback: Trackback URL.

Post a Comment

You must be logged in to post a comment.

Request a proposal

Valorisation Plan Authors

Related Documents

There in no related documents

Visit the other applications of the INTERVALUE Platform: R&D Repository | IP Agreements

© 2009-2010 INTERVALUE Project