1. No category

DATA SHEET
InView Technology Corporation Brings Compressive Sensing to SWIR Imaging and Analytics
InView210 High-Resolution SWIR Camera
TM
For applications in microscopy, scientific imaging, and image analytics
The InView210M™ is a high-resolution, shortwave infrared (SWIR) camera for microscopy and scientific
imaging applications. The camera is a computational imaging and processing platform based on a new
sampling architecture that provides researchers a way
to access high resolution information at reasonable
prices. With the InView210™, InView Technology
Corporation has developed the world’s first camera
based on Compressive Sensing. Compressive Sensing is
a new data sampling strategy whose mathematical
foundation was developed only within the last decade.
From this foundation, InView has developed a
computational imaging platform protected by over 25
patents that constructs high-resolution images from
low-resolution sensors. InView has proven the concept
by building XGA resolution SWIR cameras from a single
InGaAs detector. At 1024 x 768 pixels, XGA represents
nearly twice the pixel count of typical SWIR cameras.
Additionally, InView has developed a multi-diode
camera design that uses a small number of diodes to
significantly reduce the data collection time per frame,
allowing video rate operation and hosting a number of
enhanced processing algorithms.
Figure 1 – InView210™ camera mounted on an
ordinary optical microscope
Figure 2 – InView210™ microscope images, SWIR, 1024 x 768
InView210M SWIR Camera Data Sheet
Figure 3 – InView210™ outdoor images, SWIR, 256x192
SWIR imaging applications include surveillance and security, semiconductor and solar panel inspection,
material science, food inspection, machine vision, and process monitoring, among many others.
Figure 4 – InView210™ (top) mounted on an Olympus BX50
microscope
InView210M SWIR Camera Data Sheet
Compressive Sensing and the Single-Pixel Camera
The computational architecture of the InView camera represents a significant departure from traditional
digital cameras. The InView210™ is the world’s first embodiment of the innovative sampling theory
called Compressive Sensing that has been developed only in the last decade. Compressive Sensing
exploits the property that images are often sparse in some transform basis to recover such images from
a small number of measurements. InView has leveraged this theory to reduce the cost of SWIR imaging
while doubling resolution compared to current typical cameras.
Steps in the Compressive Sensing imaging architecture are enumerated in Fig. 5 below.
Figure 5 - How Compressive Sensing Cameras Work
The architecture implements the Compressive Sensing sampling strategy with Texas Instruments’ digital
micromirror technology (TI DLP™).  Step 1: In contrast to traditional cameras that image directly onto
a grid-like focal plane array (FPA) of small pixels, in the Compressive Sensing architecture the scene is
imaged onto the TI DLP™.  Step 2: Modulation patterns designed using proprietary algorithms are
combined with optical images using the micro-mirrors, and  Step 3: the encoded scene is focused
onto one or a small number of detectors.  Step 4: For each modulation pattern, one measurement is
taken and added to a data vector. Although the data vector is much shorter than data streams read out
from pixel arrays of traditional cameras, it contains all the information of the scene plus the information
InView210M SWIR Camera Data Sheet
encoded in the patterns.  Step 5: This data vector can be used to construct high resolution images or
undergo other processing for target detection, tracking or classification applications.
Compressive Camera Design
The computational architecture of InView’s compressive cameras is a unique combination of optical and
computational components. The camera system consists of four printed circuit boards that perform the
following functions:
1. System Controller Board: provides services to the other functional modules including start up
control, power, analog-to-digital converter (ADC) components, communication timing control,
and communication interface with external human machine interface (HMI) devices.
2. DMD Pattern Generation Board: carries a field-programmable gate array (FPGA) to generate
patterns and deliver them to the DMD for embedded CS schema. It also has a high-bandwidth
data pathway (PCIe) to deliver externally generated mirror patterns to the DMD, facilitating the
development of novel patterns and algorithms.
3. Analog Front End (AFE) Board: supports the optical detector and its thermal electric cooling.
4. DMD Carrier Board: supports the TI 1024×768 pixel (XGA) DMD.
Data is acquired and pre-processed on-board the camera. Image reconstruction is performed on a
tethered PC.
InView210M SWIR Camera Data Sheet
Figure 6 - Opto-mechanical CS camera design showing (a) enclosed system; (b) Opto-electronics; (c) detector front end in place;
(d) removable detector alignment fixture
The system controller CPU manages all aspects of the camera including updating embedded firmware,
performing device power-on, self-tests, and managing modulator and detector calibrations. An image
capture starts when an external Personal Computer (PC) sends the acquisition parameters to the camera
CPU. The CPU then parses this information and configures the system accordingly. Once the
parameters are loaded into the FPGAs the acquisition is started. Patterns are either generated internally
in the pattern generator FPGA, or are received directly via the PCIe bus from software running on the
PC. As the patterns are displayed on the modulator, each detector directly measures the optical signal
representing the modulated scene. These measurements are tagged with information from the pattern
generator indicating its position in time with respect to the pattern on the modulator. Measurement
data can be passed unaltered directly to the reconstruction software on the PC, or it can be processed
and filtered by the signal capture and processing FPGA and then sent to the PC.
Modulator
Pattern
Generator
FPGA
Detector
System
Controller
CPU
Detector
Electronics
ADCs
Signal
Capture &
Processing
FPGA
Camera
PCIe
Ethernet
or USB
PC
User Interface
&
Reconstruction
Figure 7 - CS camera operation flow chart and photograph of internal boards and optics.
InView210M SWIR Camera Data Sheet
User Interface: CompressView 1.0
TM
CompressView™ software is an easy to use interface developed in-house exclusively for use with
InView’s Compressive Sensing-based cameras. The application is designed to manage camera setup,
data capture and processing, and image reconstruction, display and saving. With this interface the user
can connect the camera to a network or automatically find any camera already on the network. Once
found, the camera is ready to use. Default parameters for acquisition, reconstruction and timing can be
easily changed using scroll down menus. The software comes included with every SWIR camera and is an
easy to use application compatible with virtually any Windows based PC. CompressView™ also contains
the camera’s API and is scalable and configurable for the development of application specific software
packages.
Figure 8 – CompressView™ user interface
InView210M SWIR Camera Data Sheet
InView210™ FEATURES
Camera Front
ETHE
P WE SWITCH
L
P WE C
Back Panel Connections
InGaAs Detector Spectral Responsivity
Responsivity (A/W)
TEST
ET
Output Image Format - 1024 x 768 or 256 x 192
Sensor Image Area - 14mm x 10.5mm (17.5mm Diag)
Output Frame Rate - ~1/3 FPS (256 x 192)
~ 1.0 FPM (1024 x 768)
Quantum Efficiency - >65% from 0.9 to 1.7um
Spectral Response – InGaAs, 0.9um to 1.7um
Dynamic Range - 5000:1
Lens Mount - M42 x 1mm
Camera Dimensions - 6.75” x 6.75” x 6.75” in W H D
Camera DC Input Voltage - 5V
Typical Power Consumption - <15W
Host Data Connection - Ethernet
Software - CompressView™ Microsoft Windows
Application
Operating Temperature -10oC to 50oC
800
900 1000 1100 1200 1300 1400 1500 1600 1700 1800
Wavelength (nm)
InView210M SWIR Camera Data Sheet
CAMERA OPERATION
As shown in Figure 1, the Camera incorporates:




Optics, XGA TI DMD (1024x768,) TIR prism, and internal condensing lens.
FPGA-based DMD controller circuitry that loads mirror patterns into the DMD
A InGaAs detector followed by a transimpedance amplifier and a high-resolution, analog to-digital
converter (ADC)
GigE Ethernet digital interface allowing for Camera control and image data acquisition by the host PC.
Incoming
Image
DMD
Pattern
Generation
Board
#2
DMD
Carrier
Board
#1
Reflected
Image
Analog Front
End Board with
photo diode &
amplifier
#3
Analog
Signal
Figure 2 – Camera Internal Component Layout
OPTICAL SYSTEM DESCRIPTION
The optical system focuses the target image from
the microscope onto a DMD (Digital MIcromirror
Device). A Total Internal Reflection (TIR) prism allows
the light reflected off the DMD to travel at a right
angle to the incoming light, allowing a compact
system. The light leaving the TIR is condensed and
focused onto the Photo Diode where it is digitized
and then processed.
Figure 2 - Optical System Diagram
ADC & System
Controller
Board
#4
Ethernet
Host
PC
Lenore McMackin, PhD, President/CTO, InView Technology Corporation, 6201 E. Oltorf St., Ste.400, Austin TX 78701
TECHNICAL SPECIFICATIONS
MECHANICAL
IMAGING
Resolution
XGA, 1024 x 768
Lens Mount
M42
Frame Rate
1 FPM
Sensor Dimensions
6.75 x 6.75 x 6.75 in WHD
Weight
<9 lbs
SENSOR
Type
InGaAs photo diode,
Spectral Response
0.9um to 1.7um
Quantum Efficiency
>65% from 0.9 to 1.7 m
INTERFACE
Ethernet
10/100
POWER
AC Adapter Input Voltage
120/240 VAC
DC Voltage
5V
Power Consumption
<15W
COMPRESSIVE SENSING RESEARCH AT INVIEW
Through a continuous program of funded research and development InView is enhancing the
computational capabilities of its compressive sensing platform, reducing the number of measurements
needed for imaging and target detection, and increasing frame rates with innovations to its algorithms
and architecture. InView continues to seek partners for development and commercialization of its
technology.
First LWIR CS Images and
Hyperspectral Data Cube
Video Compressive Sensing
Algorithms
Air Force
Research Lab
Solar Exclusion
8.7um 10 um 10.3 um 10.6 um 10.9 um 11.5 um
InView210 White Paper, Rev 1.0
Copyright © 2015, InView Technology Corporation
January, 2015
9
Preliminary Document: This document contains advanced information on a product under development. Specifications are subject to change without notice.
Lenore McMackin, PhD, President/CTO, InView Technology Corporation, 6201 E. Oltorf St., Ste.400, Austin TX 78701
SWIR-VIS
multi-color camera
Zeroth order
detector
Imaging Lens
Spectral
features
Faster Multi-detector
CS Camera
Tracking
High-Speed Objects
with a CS Imager
0th order
Focusing Lens
Diffraction
grating
TIR prism
DMD
Figure 9 – Past and current research partners and programs performed at InView

InView has demonstrated an adaptive imager that can dynamically perform pixel
exclusion or region-of-interest aggregation to eliminate background signal and to
improve overall signal-to-noise performance for a target of interest.

InView is also developing a multi-spectral camera capable of imaging throughout the
visible to SWIR wavelength range

In addition to the single-diode camera, InView has developed a multi-diode camera that
uses a small number of diodes to significantly reduce the data collection time per frame,
allowing video rate operation.

InView has also developed a set of algorithms making data acquisition several times
more efficient, and provided new strategies for detecting events of interest in the data.
These algorithms make InView’s compressive sensing architecture an intelligent
platform for quickly and automatically extracting information from a scene without
image data overload.
Patents
InView has exclusive license and sub-licensing rights to ice University’s foundational Intellectual
Property (IP) which fundamentally defines the concept of CS imaging. That IP was developed and
demonstrated at Rice with over $10M in government funding. The Rice University inventing professors
are co-founders, advisors and consultants of InView. InView has developed 24 additional patents on the
design and operation of compressive sensing imagers and data processors.
InView is seeking to license or sell their technology and patents. The Rice University patents, as noted in
the patent list, if purchased, would have to be negotiated separately.
InView210 White Paper, Rev 1.0
Copyright © 2015, InView Technology Corporation
January, 2015
10
Preliminary Document: This document contains advanced information on a product under development. Specifications are subject to change without notice.
Lenore McMackin, PhD, President/CTO, InView Technology Corporation, 6201 E. Oltorf St., Ste.400, Austin TX 78701
The following US patents are assigned or exclusively licensed to InView:
Patent No
Title
US20130002715
Image Sequence Reconstruction based on Overlapping Measurement
Subsets
Mechanisms for Conserving Power in a Compressive Imaging System
6/28/11
Publish/
Grant
Date
1/3/13
6/28/11
1/3/13
User Control of the Visual Performance of a Compressive Imaging
System
HIGH-SPEED EVENT DETECTION USING A COMPRESSIVE-SENSING
HYPERSPECTRAL-IMAGING ARCHITECTURE
Dedicated Power Meter to Measure Background Light Level in
Compressive Imaging System
OVERLAP PATTERNS AND IMAGE STITCHING FOR MULTIPLE-DETECTOR
COMPRESSIVE-SENSING CAMERA
Generating Modulation Patterns for the Acquisition of Multiscale
Information in Received Signals
Sensing Signals with Affine-Harmonically Related Rows of KroneckerProduct Matrices
Efficient Transforms and Efficient Row Generation for Kronecker
Products of Hadamard Matrices
SIGNAL RECONSTRUCTION USING TOTAL-VARIATION PRIMAL-DUAL
HYBRID GRADIENT (TV-PDHG) ALGORITHM
*Method and apparatus for compressive imaging device
6/28/11
1/3/13
6/28/11
5/23/13
8/11/10
11/11/14
12/14/12
6/19/14
1/16/13
7/17/14
1/31/13
7/31/14
1/31/13
7/31/14
12/21/12
10/2/14
4/21/05
6/12/12
10/25/06
7/9/2013
5/10/06
9/18/07
8/8/07
3/31/09
US20130002858
US20130002968
US20130128042
US8885073
US20140168482
US20140198236
US20140211000
US20140211039
US20140297703
US8199244
US8483492
Priority
Date
US7271747
*Method and apparatus for sparse signal detection, classification and
estimation from compressive Measurements
* Method and apparatus for distributed compressed sensing
US7511643
* Method and apparatus for distributed compressed sensing
US8570405
Determining light level variation in compressive imaging by injecting
calibration patterns into pattern sequence
Low-pass filtering of compressive imaging measurements to infer light
level variation
Dynamic range optimization in a compressive imaging system
8/11/10
10/29/13
8/11/10
10/29/13
8/11/10
1/21/14
Adaptively filtering compressive imaging measurements to attenuate
noise
Dual-port measurements of light reflected from micromirror array
8/11/10
5/6/14
8/11/10
5/6/14
TIR prism to separate incident light and modulated light in compressive
imaging device
Focusing mechanisms for compressive imaging device
8/11/10
5/6/14
8/11/10
5/6/14
Adaptive search for atypical regions in incident light field and spectral
classification of light in the atypical regions
Compensation of compressive imaging measurements based on
measurements from power meter
Decreasing image acquisition time for compressive imaging devices
9/30/11
5/6/14
8/11/10
6/24/14
8/11/10
10/14/14
Dedicated power meter to measure background light level in
compressive imaging system
Hot spot correction in a compressive imaging system
8/11/10
11/11/14
8/10/11
12/30/14
US8570406
US8634009
US8717463
US8717466
US8717484
US8717492
US8717551
US8760542
US8860835
US8885073
US8922688
*Owned by Rice University and licensed to InView; InView has sub-licensing rights
InView210 White Paper, Rev 1.0
Copyright © 2015, InView Technology Corporation
January, 2015
11
Preliminary Document: This document contains advanced information on a product under development. Specifications are subject to change without notice.
Lenore McMackin, PhD, President/CTO, InView Technology Corporation, 6201 E. Oltorf St., Ste.400, Austin TX 78701
REFERENCES AND FURTHER READING
1. Lenore McMackin ; Matthew A. Herman ; Bill Chatterjee and Matt Weldon,
"A high-resolution SWIR camera via compressed sensing", Proc. SPIE 8353, Infrared Technology and
Applications XXXVIII, 835303 (May 1, 2012); http://dx.doi.org/10.1117/12.920050
2. Matthew A. Herman ; James Tidman ; Donna Hewitt ; Tyler Weston and Lenore McMackin
" A higher-speed compressive sensing camera through multi-diode design ",Proc. SPIE 8717,
Compressive Sensing II, 871706 (May 31, 2013); http://dx.doi.org/10.1117/12.2015745
3. James Tidman ; Tyler Weston ; Donna Hewitt ; Matthew A. Herman and Lenore McMackin
" Compact opto-electronic engine for high-speed compressive sensing ", Proc. SPIE 8856, Applications
of Digital Image Processing XXXVI, 885616 (September 26, 2013); http://dx.doi.org/10.1117/12.2024148
4. M. Herman, "Compressive Sensing with Partial-Complete, Multiscale Hadamard Waveforms," in
Imaging and Applied Optics, OSA Technical Digest (online) (Optical Society of America, 2013), paper
CM4C.3.
http://www.opticsinfobase.org/abstract.cfm?URI=COSI-2013-CM4C.3
5. L. McMackin, M. A. Herman, D. Hewitt, and T. Weston, "Low-cost, High-resolution Shortwave
Infrared Microscope Camera Based on Compressive Sensing," in Optics in the Life Sciences, OSA
Technical Digest
(online)
(Optical Society of America, 2013),
paper
NTh2B.4.
http://www.opticsinfobase.org/abstract.cfm?URI=NTM-2013-NTh2B.4
6. Chengbo Li; Ting Sun; Kelly, K.F.; Yin Zhang, "A Compressive Sensing and Unmixing Scheme for
Hyperspectral Data Processing," Image Processing, IEEE Transactions on , vol.21, no.3, pp.1200,1210,
2012. http://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=6016235&isnumber=6151934
7.
Thomas
A.
Russell ;
Lenore
McMackin ;
Bob
Bridge and
Richard
Baraniuk
" Compressive hyperspectral sensor for LWIR gas detection ", Proc. SPIE 8365, Compressive Sensing,
83650C (June 8, 2012); http://dx.doi.org/10.1117/12.919522
8. Dharmpal Takhar ; Jason N. Laska ; Michael B. Wakin ; Marco F. Duarte ; Dror Baron, et al.
"A new compressive imaging camera architecture using optical-domain compression", Proc. SPIE 6065,
Computational Imaging IV, 606509 (February 02, 2006); http://dx.doi.org/10.1117/12.659602
9. Mark A. Davenport ; Marco F. Duarte ; Michael B. Wakin ; Jason N. Laska ; Dharmpal Takhar, et al.
"The smashed filter for compressive classification and target recognition",Proc. SPIE 6498,
Computational Imaging V, 64980H (February 28, 2007); http://dx.doi.org/10.1117/12.714460
10. Baraniuk, Richard G. "Compressive sensing." IEEE signal processing magazine 24.4 (2007).
11. Baraniuk, Richard G. "Single-pixel imaging via compressive sampling." IEEE Signal Processing
Magazine (2008).
Compressive Sensing: The Big Picture - https://sites.google.com/site/igorcarron2/cs
Compressed sensing - http://en.wikipedia.org/wiki/Compressed_sensing
InView210 White Paper, Rev 1.0
Copyright © 2015, InView Technology Corporation
January, 2015
12
Preliminary Document: This document contains advanced information on a product under development. Specifications are subject to change without notice.
Lenore McMackin, PhD, President/CTO, InView Technology Corporation, 6201 E. Oltorf St., Ste.400, Austin TX 78701
About the Company
The InView engineering and business team has deep experience in CS mathematics and algorithms,
opto-mechanical system design, hardware and software design, manufacturing, and business
development, and has filed additional IP patents which define techniques required to build practical,
high-volume CS cameras.
InView’s work in designing and manufacturing standard products has been well funded by investors
including Cottonwood Capital Partners, In-Q-Tel and the State of Texas Emerging Technology Fund.
THE FINE PRINT
The information contained in this document has been carefully checked and is believed to be entirely reliable.
However, no responsibility is assumed for inaccuracies. Furthermore, InView Technology Corporation reserves the
right to change this document and product without notice and to make improvements in reliability, function and
design without notice. InView Technology Corporation neither assumes any liability arising out of the application
or use of any product, software or circuit described herein, nor does it convey any license under its right or the
rights of others. No part of this document may be reproduced or transmitted in any form or by any means,
electronic, mechanical, for any purpose, without the express written permission of InView Technology
Corporation.
InView Technology Corporation
6201 E. Oltorf Street
Suite 400
Austin, TX 78757
Tel: (512) 243-8751 x105
[email protected]
www.inviewcorp.com
© 2015 InView Technology Corporation. All rights reserved
InView210 White Paper, Rev 1.0
Copyright © 2015, InView Technology Corporation
January, 2015
13
Preliminary Document: This document contains advanced information on a product under development. Specifications are subject to change without notice.