Latest News

DISTRIBUTORS  The following sell / support and integrate I-CUBE products:    FANG (Estate specialists);  BIAMIC ICT Solutions; 

Marshall International

 SECURITY  SOLUTIONS

MINING Consulting  services

Digital Video Solutions

 Intervid 

CAMSEC

If you wish to distribute I-Cube products APPLY HERE:

 License Plate Recognition Products

Product line

Seelane system

V-Metrics 

LPR downloads

SA Personalized LPR Plate Demo (Zipped file 1 MB)

Weigh Bridge integration

LPR for speeding

 TURNKEY DIGITAL SURVEILLANCE AND LAW ENFORCEMENT SYSTEM

 Facial Products

I-Cube Biometric Selector

Facial Solutions for MONTECASINO (pdf)

FACIAL VERIFICATION linked to a PIN Code

Facial FLASH Presentation (10 MB)

DISCOVERY PDF DOCUMENT (50KB)     DISCOVERY WORD DOCUMENT (200KB) 

FRS Discovery System OCX Control Reference Manual (100 KB PDF)

Installation and user manual (1.3MB PDF) 

Facial linked to LPR 

ESTATE Access control via LPR & Facial Verification

LPR / Facial Capture / ID or Verification LINKED to CARD / Finger Print readers and/or Car Colour:

Power Point Presentation (8 MB)

PDF (1 MB)

Facial Verification linked to License Plate Recognition (PDF)

SVS CASINO ID software (RAR) - Requires DISCOVERY CD

Facial linked to hand scan CODE (RAR) - Requires DISCOVERY CD

Web site is packed with information on our product lines. We invite you to explore the site and download the technical documentation, news items, photos, description of sample installations, system simulations and recognition demos. 

Ellips Rio High Performance Frame Grabber

Overview Ellips Rio High Performance Frame Grabber: Designed to get the most out of any S-Video/composite colour or standard black and white camera, the Ellips Rio is a high performance PCI frame grabber, targeted at real-time industrial image processing applications. The Rio comes in two versions: Basic version with one input module Full version with two input modules The Basic Rio features: 3rd generation PCI bus master with flexible transfer controller 6 (CVBS or monochrome) cameras or 3 Y/C cameras can be connected Black & white (grayscale) digitizing Colour digitizing, inputs Y/C or CVBS Simultaneous grabbing from two black and white cameras Supports progressive scan cameras The extra Full Rio features are: Independent processing of two video sources Fast switching between colour cameras Simultaneous grabbing from four black and white cameras High quality black and white input with programmable gain and offset Supports RGB and RGBA cameras   Features 3rd generation PCI bus master with flexible transfer controller Several PCI grabbers have been available on the market for some time, but, depending on their architecture, most of them are not suited for real-time imaging processing applications: Some grabbers are based on a PCI SVGA controller which have an image port that allows write access to the video memory from an external access bus (typical bandwidth 45MB/s). Although this method offers live video on the desktop without hogging the PCI bus, reading the image back from video memory to main memory suffers from slow memory access due to the nature of these SVGA controllers: typical write bandwidth from PCI to SVGA is up to 90MB/s, whereas read bandwidth from SVGA to PCI is limited to approximately 7MB/s. Therefore SVGA-controller based PCI grabbers are generally not suited for real-time imaging applications.   Some PCI grabbers employ efficient PCI burst transfers, but lack flexible transfer controller capabilities. Therefore, more data is sent over the PCI bus than is actually needed, thus slowing the system down. For example, when tracking moving objects, the region of interest changes for every video frame. This calls for changing the region of interest at field rate. The Ellips Rio offers PCI bus master DMA, allowing burst transfers of up to 132MB/s (depending on the quality of the main board) without tying up the main CPU, combined with a very flexible transfer controller. Using a Register Program Sequencer, input window, output destination, scaling, and clipping can be reprogrammed on a line-by-line basis without CPU intervention. The sequencer program can be changed every field to incorporate complex changes of the image parts transferred over the PCI bus to main memory, thus tracking moving objects. The PCI interface supports virtual memory addressing for operating systems running virtual demand paging (like Windows). The integrated Memory Management Unit (MMU) translates linear to physical addresses using a page table in system memory provided by the software driver. The MMU supports op to 4 Mbytes of virtual address space per DMA channel. 6 (CVBS or monochrome) cameras or 3 Y/C cameras can be connected Up to 6 CVBS or monochrome cameras or 3 Y/C cameras can be connected to a Rio board. Also CVBS or monochrome connections in combination with Y/C connections are allowed. Of course every Y/C camera needs to be connected to Y and C connectors on the Rio that belong together. Black & white (grayscale) digitizing Black and white camera standards CCIR (50 fields per second) and EIA (60 fields per second) are supported. Different output formats are possible (grayscale, binary). Colour digitizing, inputs Y/C or CVBS The NTSC (60 fields per second), PAL and SECAM (50 fields per second) colour standards are supported. A wide variety of output formats is supported. Simultaneous grabbing from two black and white cameras Although the basic version of the Rio only has one input module, it still is capable of simultaneous grabbing at 50 (CCIR) or 60 (EIA) fields per second with two genlocked monochrome cameras (stereo locked mode). However to obtain conventional output formats, postprocessing is needed, which will have an impact on the capture rate. Supports progressive scan cameras The Rio can be used to digitize from a progressive scan camera at 50 (CCIR) or 60 (EIA) frames per second, e.g. JAI CV-M10). Full frame camera input is supported for the Full Rio as well as for the Basic Rio. For the Basic Rio however postprocessing is needed to obtain conventional output formats. Additionally the Full Rio can also adjust brightness and contrast in full frame mode, or the Full Rio can be used to digitize from two progressive scan cameras simultaneously.   Independent processing of two video sources Because the full version of the Rio also has two DMA channels, two colour cameras each can be handled totally separately, so it is possible for example for one input module to be connected to a colour camera and one input module to a monochrome camera. Also simultaneous grabbing from two colour cameras is possible with the Full Rio.     Fast switching between colour cameras For many image processing applications evaluating outputs from multiple cameras is desirable. When allowable with respect to speed constraints, hardware costs can be reduced by fast switching between multiple cameras on a field by field basis, using a single frame grabber. With traditional grabbers, fast switching between colour cameras demands colour carrier sub-locking in addition to a conventional genlock. This often involves modifying camera electronics, thus increasing costs. The full version of our Ellips Rio frame grabber is capable of fast switching between two colour cameras without colour carrier sub-locking, because each of the two cameras can be connected to its own input module. In this mode, a field rate of 25 fields (PAL; 30 in NTSC) per second per camera can be achieved. When no RGB output is required, this mode can even be used at 50 (PAL/SECAM) or 60 (NTSC) fields per second per camera.   Simultaneous grabbing from four black and white cameras With the Full Rio, because it has two input modules, both modules can be used in stereo locked mode, making simultaneous grabbing at 50 (CCIR) or 60 (EIA) fields per second with four genlocked monochrome cameras possible. However to obtain conventional output formats, postprocessing is needed, which will have an impact on the capture rate.

High quality black and white input with programmable gain and offset For high quality black and white measurements, the full version of the Rio has a separate converter stage with programmable gain and offset. Its input can be switched at field rate between genlocked cameras. In contrast to the other modules (which only have an output range of 192 values, which will result in missing values when mapped to a 256 value scale), the high quality black and white module has a full output range of 256 values. Therefore for serious black and white measurements, the full version of the Rio is recommended. Supports RGB and RGBA cameras With the Full Rio, one RGB(A) camera can be connected. Post-processing is needed to obtain one of the output formats that are supported (including planar output).   Architectural Overview On a Basic Rio, only input module 0 is available without the High Quality Black & White functionality. Connected to the input modules are the two scalers; the High Performance Scaler (HPS) that consists of three parts (downscaler, color space converter and pixel formatter) and the more basic Binary Ratio Scaler. Each input module can be connected to either scaler. The scalers are connected to the DMA control (two DMA channels), which on its turn is connected to the PCI local bus.       Specifications Host Computer Interface High speed DMA burst transfer (up to 132MB/s) Programmable transfer controller Support for multiple boards in one system Acquisition RS-170, CCIR, NTSC, PAL, SECAM, Y/C or monochrome inputs Resolutions: any size up to: PAL, SECAM, CCIR native: 768x576 NTSC, EIA native: 640x480 Output formats: RGB32 (aRGB) RGB24 (RGBR...) RGB16 (5:6:5) RGB15 (a:5:5:5 or 5:5:a:5) RGB8 (3:3:2) YUV16 (CCIR) Y8 (256 grayscale) Y2 (2 bit grayscale) Y1 (B/W) Compensation of Gamma Pre-Correction is supported on RGB modes. Sample rate: 14.75 / 7.38 MHz for PAL / SECAM; 12.27 / 6.13 MHz for NTSC. Digital PLL provides stable synchronization even when using still video cameras and VCRs Support for any combination of up to three Y/C video sources and six composite or monochrome video sources 6 SMB connectors for six monochrome, CVBS, three Y/C pairs (S-Video) or one RGB(A) camera. Software support Portable operating system independent C library. Supported platforms: Dos (32-bit) (including full source code): Watcom C/C++ (11.0) + Pharlap TNT DOS-extender Windows 9x/NT/2000/XP (32-bit): Microsoft Visual C 4.x / 5.x / 6.x Microsoft Visual Basic 6 VxD/device driver and DLL for C-library and Visual Basic Real-time operating system VxWorks WindRiver Tornado 2.x Capture and sample applications for DOS and Windows. Support for Image-Pro Plus.(development tool)  3rd party support by ILIB. Specifications subject to change without notice. For more information, feel free to contact Ellips.   Sidebar: PCI Bus Issues Burst transfers To understand PCI transfer rates, some background information on the PCI bus protocol may come in handy. A PCI bus can be populated by several masters (or initiators) and targets. Ownership of the bus is arbitrated between masters. A master that owns the bus can transfer data to and from targets. PCI bus transfers occur in bursts: an address exchange followed by one or more data exchanges. The more data is exchanged after a single address, the higher the transfer rate. The following table lists PCI bus efficiency versus burst length (for single clock cycle arbitration latency and two clock cycle target latency, using a PCI clock frequency of 33 MHz): Burst Size/transfer Maximum Available Bandwidth PCI Bus Efficiency 1 Dword (4 bytes) 33 MB/s 25 % 2 Dwords (8 bytes) 53 MB/s 40 % 4 Dwords (16 bytes) 75 MB/s 57 % 8 Dwords (32 bytes) 96 MB/s 73 % 16 Dwords (64 bytes) 111 MB/s 84 % 32 Dwords (128 bytes) 121 MB/s 91 % 64 Dwords (256 bytes) 126 MB/s 96 % As arbitration latency and target latency increase, bandwidth and efficiency decrease, especially with small bursts. Bus master DMA Transfers can be carried out either by the main CPU, or by the PCI board, acting as bus master DMA controller. The former is more simple for PCI hardware developers, since they only have to implement a PCI target interface, which is freely available for the users of most types of FPGAs. A target-only implementation, however, ties down the CPU during transfers. A better alternative is a bus master DMA controller, allowing the CPU to continue program execution. FIFO instead of frame memory Given the fact that nowadays most PCI main board chipsets support transfer rates of at least 65 MB/s, on-board frame memory is unnecessary. A small FIFO can accomodate for bus latency. Using a FIFO instead of frame memory decreases image latency and hardware costs. The danger of FIFO overruns can be eliminated by transfering only the parts of the image actually needed, using a flexible transfer controller. Real-time response Due to the nature of burst transfers, access to the PCI bus becomes available within a guaranteed, small, time frame. This ensures that access to a new frame is available before the FIFO is completely filled. It also ensures that during image transfer the bus is freed periodically to allow access to memory or e.g. a host bus adapter. Bus arbitration ensures a fair time slot allocation, making the bus suitable for real-time applications.

License Plate Recognition

License Plate Recognition for a wide range of applications including Parking, Access Control, Logging all vehicles & alarm when Wanted Vehicles detected.   

Facial Identification & Verification Solutions   

CALL Barry on +27 31 764 3077  or   + 27 (0) 82-562-8225  or E-Mail NOW (infoAT I-Cube DOT co DOT za)  OR Fax Number : 0866539659 OR Contact one of our DISTRIBUTORS  

Please download (ZIP / Word) the I-Cube Company Profile and Products for more information

I-Cube.   All rights reserved.  Revised: January 05, 2008 .                           BTD (QG) Quotes