infrared technology - seeing the invisible (part two: camera technology)
DESCRIPTION
Through specific applications examples with sample images, this presentation introduces you to the basics of infrared (IR) imaging technology. You will learn that in the IR-world things look different and that you can visualize with an IR camera things which you cannot see with your own eyes. To understand “the why”, we touch on some basics about IR radiation and corresponding imaging sensor technologies.TRANSCRIPT
Infrared Imaging: Seeing the Invisible
Part Two:
Camera Technology
Sensor incl. digitization
Optics &
Filters
Sensor Cooling (optional)
Gain/Offset
Correction
(NUC)
Defect
Pixel
Correction
Background
Correction
Temp.
Calibration
via
LUT
Drift
Compen-
sation
Firmware • Feature Control • Image Correction • Temperature
Calibration
Interface and I/O Control
Structure of an Infrared Camera
Optics & Filters
SWIR optimized lens Non-optimized lens
Image with / without SWIR Lens
MWIR and LWIR Optics
• For wavelengths > 2.5 µm that glass would block
• Special & costly optics: germanium and silicon
• Further materials available for high transmittance
• No standard mounts
Filters for SWIR Wavelengths
• Filters are used to increase contrast
• They often correspond to the absorption spectra of specific substances.
Example: Water filter 1450 nm
without filter with filter
IR SWIR (InGaAs)
narrow bandpass (1450nm)
Visible light
• Filters are used to increase contrast
• They often correspond to the absorption spectra of specific substances.
black
dark
clear
Water color
How the Water Filter Works
Sensor Technology
Quantum vs. Thermal Detectors
• Quantum Detectors
• Sensitivity dependent on wavelength
• Require cooling to improve S/N ratio especially for wavelengths beyond 1µm
• High detection performance and fast response
• Thermal Detectors
• Detect IR energy as heat
• In general do not require cooling
• Have a slow response time and detection capability
1 2 3 4 5 6 7 8 9 10 11 12 13 14 [µm]
InGaAs
InSb
µ-Bolometer
QWIP
MCT
Si-based
CCD/CMOS
Quantum
Detectors
Thermal
Detectors
LWIR N I R
SWIR MWIR V I S
Spectral Sensitivity
for Typical IR Detector Types
Infrared Detector Selection
Min. Object Temperature (self-emissive)
Sensor Type Sensor wavelength [µm]
Operating Temperature
800 °C CCD/CMOS [Si] < 1 300 K (27 °C)
250 °C SWIR [InGaAs] < 1.7 300 K (27 °C)
0 °C MWIR [InSb] < 6 77 K (-196 °C)
-70 °C LWIR [µBolometer]
< 14 300 K (27 °C)
-150 °C LWIR [MCT] < 20 77 K (-196 °C)
Reference temps: White hot steel ~1200 °C Melting point of aluminum 660 °C Water boils at 100 °C Uncooled camera at 38 °C Human body at 37 °C, radiates at ~ 10 μm Water freezes at 0 °C
Cooling Methods
• Cryogenic Cooling
– dry ice or liquid nitrogen
– mechanical cooling using Stirling elements
• Thermoelectric Cooling (TEC) using Peltier elements
– Lower cost
– Solid state – no vibration
SWIR Sensor Technology
• Quantum detector
Working principle: Absorption of photons that elevate the material’s electrons to a higher energy level, so that they can be counted
• Hybrid array: IR detector, Si readout
Indium bumps on each pixel of array and readout IC
• Thermal detector
Working principle: Detection of electrical resistance changes in a thermally insulated absorber material (VOx, a-Si)
• Hybrid array: IR detector, Si readout
Spectral range: 8 ..14 µm i.e. for LWIR
µBolometer Sensor Technology
Comparing Camera Performance
• Noise Equivalent Temperature Difference [NETD]: A measure of detector sensitivity; influences precision of temperature measurement – Measured in °C or K
– 10 mK – 200 mK typical
• Is equal to temperature difference which would produce given noise
NETD
f-number
thermal time constant
temperature
Influencing physical variables:
Various Heat Sources Cause Drift
• Heat comes from: – Scene / object of interest
– Lens
– Camera housing
– Sensor (FPA)
For temperature measurement, corrections for the undesired heat effects are essential
Heat can´t be “blocked” like visible light
Optical lens
FPA
Image Processing
Closer Look at SWIR Sensor Image
• Non-uniformities
• Defect Pixels
• Incorrect flip-chip bonding
1. Original image of an uncooled SWIR sensor
2. With Gain-Offset Nonuniformity Correction (aka NUC)
3. With Error Pixel Correction
How an Image is Processed
@20ms Exposure
@100ms Exposure after NUC
@40ms Exposure @100ms Exposure
Influence of Exposure Time
1. Sensor Temp. +40°C
@100ms Exposure
2. Sensor Temp. -11°C
@100ms Exposure
3. @800ms Exposure
4. Including NUC 5. Including Defect
Pixel Correction
Effect of Sensor Temperature
Allied Vision Technologies GmbH Taschenweg 2a 07646 Stadtroda, Germany Tel.: +49 36428 / 677-0 Fax: +49 36428 / 677-24
[email protected] www.alliedvisiontec.com
Follow us: