Entex Facial link (555 KB) 

 SVS FRS V1.rar  (70 kb)

 facial linked to hand scan CODE .rar (85 kb)

DEMO EQUIPMENT

I-CUBE Facial Identification FLASH Presentation zip (8 MB)

 Face Identification .zip  (17 MB)

FACIAL ID, VERIFICATION AND LPR SPEED / ACCESS CONTROL SYSTEM

FRS SDK OCX:

SDK PDF DOCUMENT (40KB)    

     SDK WORD DOCUMENT (150KB)

 

I-Cube provides two types of facial recognition systems, 2D facial ID and advanced 3D neural network facial verification

    DISCOVERY

Discovery gives you the opportunity to explore the full range of possibilities for biometric authentication within your organization without a major up-front investment in software and hardware peripherals. FRS Discovery provides biometric authentication, combining the advanced face recognition capabilities of the Face Recognition System (FRS) with facial recognition and verification on a standalone PC.

To help you in your proof-of-concept activities, the Discovery 2.0 package includes the ability to link into web or CCTV cameras, proximity card and fingerprint readers, relay output cards and demonstration applications. The system also provides a full OCX interface to allow you to integrate any component of the system into custom applications written in Microsoft Visual C++ or Visual Basic.

The FRS Discovery package gives you first hand experience of our leading-edge face recognition technology. Features include:

The tracking capability of FRS is simply the most advanced in the industry. Discovery demonstrates tracking of up to four faces simultaneously in a wide variety of conditions (indoor, outdoor and low-light). The system tracks faces in real time (30 FPS) and at head rotation angles of up to 90 degrees from the camera. The video display shows the tracking operation using color-coded overlays, and even illustrates the ability of the system to determine head orientation.

The system incorporates an interactive enrollment procedure in which the computer guides the user through the process verbally. The system is aware when the user enters and leaves the camera's field of view, when multiple users are attempting to enroll, and when the user is positioned too far from the camera, and responds appropriately with verbal instructions.

A video image is often all that is available to identify an individual. The system allows enrollment form a static image (JPEG). Although identification and verification will be less accurate than would be the case with live enrollment, Acsys FRS demonstrates exceptional authentication capabilities using even a single photograph.

The system can identify up to four individuals simultaneously from a live or recorded video feed. This capability is extremely important in surveillance mode (i.e. identifying individuals in a crowd). The system logs each identification event and stores an image of the identified individual for subsequent analysis and reporting.

Verification is performed on a single individual within the video frame (or static image). In the event that multiple individuals are present within the field of view, the system issues verbal warnings indicating that only one individual is permitted in from of the camera. The ability to distinguish multiple faces simultaneously is a key requirement for both physical and logical (computer) access control applications.

Database searches can be conducted using live- or static-enrollment generated templates. The demonstration program integrates six of the most popular OLEDB providers, including Microsoft Access, Microsoft SQL Server and Oracle 9i. The system allows customization of searches based on the number of closest matches and on confidence thresholds. The search speed provided within the demonstration program is 10,000 images per second.

The access control application demonstrates the following combinations of standard card token (proximity card), personal identification number (PIN) and biometric authentication: 

•Proximity card and facial recognition

• Enrolling subjects from static image (JPEG), and identifying multiple individuals from live or recorded video

• Searching the database for enrolled subjects (whether the templates were generated from video or static images)

Remote devices can be activated upon detection of faces. Also, remote notification and image transmission via the Web can be triggered by the following event types:

 

 

 

 

 

 

 

 

 

 

 

The Software Developers’ Kit (SDK) contains face finding and face recognition technologies designed for use in one-to-one verification scenarios, such as access control or time and attendance, and one-to-many identification scenarios, such as identification of faces in a crowd. Based on Distributed Component Object Model (DCOM) technology, FRS technology can track up to 16 people simultaneously in real time at 30 FPS (frames per second), recognize faces, and provide multimedia display and feedback (using both video and static images). It can be implemented in a stand-alone configuration, or within 3-tier architectures (FRS client/FRS server/central database). The SDK 2.2 release operates with both LAN and Web connectivity. The SDK Web service component, for fully Web-deployable solutions, has recently been released. Our SDK currently supports up to 4000 FRS client installations and over 1 million enrolled users. Advanced utilities provide automated and/or manual synchronization of clients with the central database. Client machines can perform all functions (face tracking, verification, identification and enrollment) while disconnected from the FRS server and central database. Data synchronization occurs automatically upon reconnection of the client to the FRS network.

Compatibility with Windows CE provides support for portable and mobile biometric solutions. All technology is encapsulated as DCOM modules operating from a standard ATL interface for compatibility with Microsoft Visual C++, Microsoft Visual Basic and most other Windows application development environments.

• Multimedia Control: DCOM support for image display/overlay using the latest DirectX 8.1 drivers and import of static images in most standard formats.

• Enrollment Control: enroll a person from video or static images (JPEG, BMP, etc.) and create a database of facial biometric templates for later use in verification (one-to-one

• Verification Control: finds faces within video input or static images (JPEG, BMP, etc.), and verifies them against a database of facial biometric templates.

• Classification Control: finds faces within video input or static images (JPEG, BMP, etc.), and generates a list of closest matches against a database of facial biometric templates. The classification operation (one-to-many identification) is performed against 25,000 individuals/second.

• Tracking Control: finds and tracks faces within video input, and tracks up to 16 individuals simultaneously at a full 30 FPS. Provides pan/zoom/tilt control for autoscanning systems, tracking intruders, etc.

• Database Control: provides all interfaces necessary to add, delete, and retrieve biometric templates, images and users from the central database. Provides the ability to import/export

• Communication Control: provides support for LAN, WAN or Web-based communication between FRS server and central database server and up to 4000 FRS clients.

• Demo Applications: sample enrollment, verification and classification application source code written in Visual C++.

SDK

I-Cube.   All rights reserved.  Revised: January 03, 2008 .