top logo
Insights

Bottom logo
www.ptgrey.com

Did you know PGR's Flea™ and Scorpion™ cameras support ROI (Region of Interest) & Binning?



January 2005
 
 
Scorpion

PGR has announced the addition of the SCOR-14SO model to the Scorpion™ line of cameras.  Using Sony's ICX267 CCD sensor, this camera streams 1.4 mega-pixel images at 19 FPS.

SPECIFICATIONS:
Sensor
Sony ICX267
color & BW
Resolution
1360x1024
Frame Rate
19 FPS
SNR
(Signal-Noise Ratio)
>50 dB

As with the other Scorpion models, additional features include:
  • Custom image modes
  • Automatic synchronization
  • General purpose I/O
  • Built-in RS232 serial port
Production units will be available February 2005.  For pricing information or to place an order for the full development kit, contact sales@ptgrey.com.

Each full development kit contains:
  • Scorpion camera
  • 4.5m, 6pin, IEEE-1394 cable
  • IEEE-1394 OHCI PCI Host Adapter 3 port-400 mbps/sec card
  • C/CS mount lens holder
  • 12 pin Hirose connector
  • PGR FlyCapture C/C++ API and device driver
  • C/C++ Image acquisition library
  • Example source code for quick start within the programming environment


For more image acquisition flexibility, PGR's Flea™ and Scorpion™ cameras offer both Region of Interest (sub-window) and Binning (pixel merge) readout modes.  Both methods reduce the output resolution of the camera, but do so in different ways that have different advantages and characteristics.

ROI - Region of Interest

Region of Interest (ROI) readout defines an output image area that is smaller than the full resolution of the camera.  For example, a 640x480 camera can define an output area of 320x240 or 640x64.    The output is a sub-window with a reduced field of view.

Examples of ROI and the increase in FPS:
Camera model
ROI
Frame Rate
FLEA-BW 640x480 320x240 99 FPS
FLEA-HIBW 1024x768 320x240
65 FPS
SCOR-03SO 640x480 320x240 80 FPS
SCOR-14SO 1360x1024 320x240
30 FPS
SCOR-20SO 1600x1200 320x240 24 FPS

Binning

Binning readout merges pixels together on the sensor before readout and digitization in the camera.  The Flea™ and Scorpion™ support 1x2 and 2x2 binning.  Binning 2x2 merges together pixel pairs horizontally (rows) and vertically (columns) to create "super pixels".  Since the pixels values are added together on the sensor, any color information is lost so output images are always monochrome.  For example, a 640x480 sensor would be reduced to 320x240 with 2x2 binning or 640x240 with 1x2 binning.

Examples of 2x2 Binning and the increase in FPS:
Camera model
Resolution
Frame rate
FLEA-BW 640x480
320x240
103 FPS
FLEA-HIBW 1024x768
512x384
51 FPS
SCOR-03SO 640x480
320x240
109 FPS
SCOR-14SO 1360x1024
680x512
28 FPS
SCOR-20SO 1600x1200
800x600
26 FPS

The following chart compares the pros and cons of these features:

Feature
ROI   
Binning
Faster frame rates when number of rows reduced
Yes
Yes
Maintains field of view at the lower resolution
No
Yes
Significant improvements in sensitivity & signal-to-noise
No
Yes
Same pixel size
Yes
No
Maintains color information
Yes
No

Region of Interest and Binning can also be used together to further fit the field of view and resolution to your imaging needs.  The reduced resolution can be used to increase the frame rate, get more cameras on the 1394 bus simultaneously, or just reduce the processing load.

For more information on the above cameras and features, contact sales@ptgrey.com.

Comments, Queries, Applications?

We encourage you to email this newsletter to a friend or colleague.  PGR is always interested in pursuing new projects and welcomes your comments and ideas. For information on Point Grey Research and our products, please visit our website www.ptgrey.com

Cheers,
The staff at PGR
 
 
HandVu features PGR's Firefly® extended camera

Wearable Firefly<sup>®</sup>

The HandVu project developed by a group at the University of California, Santa Barbara, features PGR's Firefly® extended camera. 

Firefly<sup>®</sup> extended
Firefly® extended camera
head piece

With the camera as a head-worn piece, the HandVu software detects the hand in a standard posture, then tracks it and recognizes key postures - all in real-time and without the need for camera or user calibration.

HandVu implements a complete, vision based human-computer interface that allows interaction with non-traditional computing environments such as wearable computers and augmented reality.  It comes with several applications that demonstrate video capture with OpenCV's highgui, DirectShow, and the Artoolkit. 
To download the library or for more information, please visit the main HandVu web site.

Check out the following images captured using HandVu.


Virtual object tracking
Interaction with virtual objects

gesturing with hand and mouse
Mouse and button interface
with hand gestures

For more information on the Firefly® camera or any other PGR Imaging Product, contact sales@ptgrey.com.









Order Now!
If you are interested in purchasing one of our products or would like a price quote, please contact us via email at sales@ptgrey.com or by phone (604)730-9937.  For a more information on our products and services, visit us at  www.ptgrey.com
We are happy to inform you that PGR will be releasing newsletters periodically.  If you want to unsubscribe, reply to this message with "unsubscribe" in the subject line.  If you have any comments about our newsletter, we would love to hear them: sales@ptgrey.com.