Thermographic camera

Infrared was discovered in 1800 by Sir William Herschel as a form of radiation beyond red light.[1] These "infrared rays" (infra is the Latin prefix for "below") were used mainly for thermal measurement.[2] There are four basic laws of IR radiation: Kirchhoff's law of thermal radiation, Stefan-Boltzmann law, Planck's law, and Wien's displacement law. The development of detectors was mainly focused on the use of thermometers and bolometers until World War I. A significant step in the development of detectors occurred in 1829, when Leopoldo Nobili, using the Seebeck effect, created the first known thermocouple, fabricating an improved thermometer, a crude thermopile. He described this instrument to Macedonio Melloni. Initially, they jointly developed a greatly improved instrument. Subsequently, Melloni worked alone, developing an instrument in 1833 (a multielement thermopile) that could detect a person 10 metres away.[3] The next significant step in improving detectors was the bolometer, invented in 1880 by Samuel Pierpont Langley.[4] Langley and his assistant Charles Greeley Abbot continued to make improvements in this instrument. By 1901, it had the capability to detect radiation from a cow from 400 metres away, and was sensitive to differences in temperature of one hundred thousandths of a degree Celsius.[5] The first commercial thermal imaging camera was sold in 1965 for high voltage power line inspections.

The first advanced application of IR technology in the civil section may have been a device to detect the presence of icebergs and steamships using a mirror and thermopile, patented in 1913.[6] This was soon outdone by the first true IR iceberg detector, which did not use thermopiles, patented in 1914 by R.D. Parker.[7] This was followed up by G.A. Barker's proposal to use the IR system to detect forest fires in 1934.[8] The technique was not truly industrialized until it was used in the analysis of heating uniformity in hot steel strips in 1935.[9]

In 1929, Hungarian physicist Kálmán Tihanyi invented the infrared-sensitive (night vision) electronic television camera for anti-aircraft defense in Britain.[10] The first American thermographic cameras developed were infrared line scanners. This was created by the US military and Texas Instruments in 1947[11] and took one hour to produce a single image. While several approaches were investigated to improve the speed and accuracy of the technology, one of the most crucial factors dealt with scanning an image, which the AGA company was able to commercialize using a cooled photoconductor.[12]

The first infrared linescan system was the British Yellow Duckling of the mid-1950s.[13] This used a continuously rotating mirror and detector, with Y axis scanning by the motion of the carrier aircraft. Although unsuccessful in its intended application of submarine tracking by wake detection, it was applied to land-based surveillance and became the foundation of military IR linescan.

This work was further developed at the Royal Signals and Radar Establishment in the UK when they discovered that mercury cadmium telluride was a photoconductor that required much less cooling. Honeywell in the United States also developed arrays of detectors which could cool at a lower temperature, but they scanned mechanically. This method had several disadvantages which could be overcome using an electronic scanning system. In 1969 Michael Francis Tompsett at English Electric Valve Company in the UK patented a camera which scanned pyro-electronically and which reached a high level of performance after several other breakthroughs throughout the 1970s.[14] Tompsett also proposed an idea for solid-state thermal-imaging arrays, which eventually led to modern hybridized single-crystal-slice imaging devices.[12]

One of the most important areas of development for security systems was for the ability to intelligently evaluate a signal, as well as warning of a threat's presence. Under the encouragement of the US Strategic Defense Initiative, "smart sensors" began to appear. These are sensors that could integrate sensing, signal extraction, processing, and comprehension.[15] There are two main types of Smart Sensors. One, similar to what is called a "vision chip" when used in the visible range, allow for preprocessing using Smart Sensing techniques due to the increase in growth of integrated microcircuitry.[16] The other technology is more oriented to a specific use and fulfills its preprocessing goal through its design and structure.[17]

Towards the end of the 1990s, the use of infrared was moving towards civil use. There was a dramatic lowering of costs for uncooled arrays, which along with the large increase in developments lead to a dual way use market between civil and military.[18] These uses include environmental control, building/art analysis, medical functional diagnostics, and car guidance and collision avoidance systems.[19][20][21][22][23][24]

A thermographic image of several lizards

Thermal imaging camera & screen, in an airport terminal in Greece. Thermal imaging can detect fever, one of the signs of infection.

Cooled detectors are typically contained in a vacuum-sealed case or Dewar and cryogenically cooled. The cooling is necessary for the operation of the semiconductor materials used. Typical operating temperatures range from 4 K to just below room temperature, depending on the detector technology. Most modern cooled detectors operate in the 60 K to 100 K range (-213 to -173 °C), depending on type and performance level.[26]

Without cooling, these sensors (which detect and convert light in much the same way as common digital cameras, but are made of different materials) would be 'blinded' or flooded by their own radiation. The drawbacks of cooled infrared cameras are that they are expensive both to produce and to run. Cooling is both energy-intensive and time-consuming.

The camera may need several minutes to cool down before it can begin working. The most commonly used cooling systems are peltier coolers which, although inefficient and limited in cooling capacity, are relatively simple and compact. To obtain better image quality or for imaging low temperature objects Stirling engine cryocoolers are needed. Although the cooling apparatus may be comparatively bulky and expensive, cooled infrared cameras provide greatly superior image quality compared to uncooled ones, particularly of objects near or below room temperature. Additionally, the greater sensitivity of cooled cameras also allow the use of higher F-number lenses, making high performance long focal length lenses both smaller and cheaper for cooled detectors.

An alternative to Stirling engine coolers is to use gases bottled at high pressure, nitrogen being a common choice. The pressurised gas is expanded via a micro-sized orifice and passed over a miniature heat exchanger resulting in regenerative cooling via the Joule–Thomson effect. For such systems the supply of pressurized gas is a logistical concern for field use.

Materials used for cooled infrared detection include photodetectors based on a wide range of narrow gap semiconductors including indium antimonide (3-5 μm), indium arsenide, mercury cadmium telluride (MCT) (1-2 μm, 3-5 μm, 8-12 μm), lead sulfide, and lead selenide

Infrared photodetectors can be created with structures of high bandgap semiconductors such as in Quantum well infrared photodetectors.

A number of superconducting and non-superconducting cooled bolometer technologies exist.

In principle, superconducting tunneling junction devices could be used as infrared sensors because of their very narrow gap. Small arrays have been demonstrated. They have not been broadly adopted for use because their high sensitivity requires careful shielding from the background radiation.

Superconducting detectors offer extreme sensitivity, with some able to register individual photons. For example, ESA's Superconducting camera (SCAM). However, they are not in regular use outside of scientific research.

Uncooled thermal cameras use a sensor operating at ambient temperature, or a sensor stabilized at a temperature close to ambient using small temperature control elements. Modern uncooled detectors all use sensors that work by the change of resistance, voltage or current when heated by infrared radiation. These changes are then measured and compared to the values at the operating temperature of the sensor.

Uncooled infrared sensors can be stabilized to an operating temperature to reduce image noise, but they are not cooled to low temperatures and do not require bulky, expensive, energy consuming cryogenic coolers. This makes infrared cameras smaller and less costly. However, their resolution and image quality tend to be lower than cooled detectors. This is due to differences in their fabrication processes, limited by currently available technology. An uncooled thermal camera also needs to deal with its own heat signature.

Uncooled detectors are mostly based on pyroelectric and ferroelectric materials or microbolometer technology.[27] The material are used to form pixels with highly temperature-dependent properties, which are thermally insulated from the environment and read electronically.

Thermal image of steam locomotive

Ferroelectric detectors operate close to phase transition temperature of the sensor material; the pixel temperature is read as the highly temperature-dependent polarization charge. The achieved NETD of ferroelectric detectors with f/1 optics and 320x240 sensors is 70-80 mK. A possible sensor assembly consists of barium strontium titanate bump-bonded by polyimide thermally insulated connection.

Silicon microbolometers can reach NETD down to 20 mK. They consist of a layer of amorphous silicon, or a thin film vanadium(V) oxide sensing element suspended on silicon nitride bridge above the silicon-based scanning electronics. The electric resistance of the sensing element is measured once per frame.

Current improvements of uncooled focal plane arrays (UFPA) are focused primarily on higher sensitivity and pixel density. In 2013 DARPA announced a five-micron LWIR camera that uses a 1280 x 720 focal plane array (FPA).[28] Some of the materials used for the sensor arrays are amorphous silicon (a-Si), vanadium(V) oxide (VOx),[29] lanthanum barium manganite (LBMO), lead zirconate titanate (PZT), lanthanum doped lead zirconate titanate (PLZT), lead scandium tantalate (PST), lead lanthanum titanate (PLT), lead titanate (PT), lead zinc niobate (PZN), lead strontium titanate (PSrT), barium strontium titanate (BST), barium titanate (BT), antimony sulfoiodide (SbSI), and polyvinylidene difluoride (PVDF).