New generation night vision devices..

New generation night vision devices..

NEW GENERATION NIGHT VISION DEVICES

Night vision devices (NVD) are designed for observation and aiming in the dark — at dusk and at night. NVD appeared at the end of World War II and have since undergone a complex path of development. They have found wide application in both military and civilian technology. Their development can be divided into a number of stages, which are associated with the emergence of certain generations of NVD. Each subsequent generation differed from the previous one by a greater range of vision, better image quality, reduced weight and dimensions, increased operating time, increased resistance to light interference and a number of other advantages. To date, three generations of NVD are known [1].

The main feature by which generations of night vision devices are distinguished is their main element – ​​an electron-optical converter (EOC), designed to convert an image invisible to the human eye into a visible one and to enhance its brightness.

The diagram of the most advanced third-generation image intensifier tube is shown in Fig. 1, where 1 is a photocathode with negative electron affinity (based on GaAs) applied to a glass input window; 2, 3 are a microchannel plate; 4 is a screen applied to an output glass window 5 (Fig. 1a) or to a fiber-optic plate 6 (Fig. 1b) without image wrapping for coupling with a CCD matrix of a television (TV) camera, or to a fiber-optic plate 7 (Fig. 1c) with image wrapping by 1800.

Fig. 1. Diagram of a third-generation image intensifier tube

Over the past decade, the development of new generation image intensifier tubes (and, accordingly, night vision devices) has been intensively conducted in the USA under the OMNIBUS program [2, 3]. This work is being conducted by ITT Defense and Litton Systems with the aim of increasing the integral sensitivity of the photocathode, the signal-to-noise ratio, and the resolving power of generation III image intensifier tubes (see Table 1). Work under the OMNIBUS III and IV programs has made it possible to shorten the technological cycle for creating generation III image intensifier tubes and develop new models of the OMNI series. The best models of these image intensifier tubes have achieved a resolving power of up to 84 lines/mm, a signal-to-noise ratio of over 23, and an integral sensitivity of the photocathode of over 2000 μA/lm. Accordingly, the viewing range in night vision devices has increased by 1.5 times compared to traditional generation III night vision devices. The new image intensifier tubes are called “highly informative generation III image intensifier tubes.” Despite their high cost (up to $10,000 per unit), all generation III image intensifier tubes have been replaced with highly informative image intensifier tubes in the USA since 1999. In some literature sources, these EOPs are even classified as generation IV. However, in fact, generation IV EOPs include products whose parameters are listed in the table, as well as the latest EOP samples from Litton: their signal-to-noise ratio exceeds 33, and the integral photocathode sensitivity is 2200 μA/lm.

Table 1.

Physically, these achievements are associated with the elimination of the ion-barrier film in the IV generation of the image intensifier tubes, which was applied to the microchannel plate (MCP) to protect the image intensifier tube photocathode from the effects of ions generated in this plate. However, this same film resulted in the reflection of the electron flow moving from the image intensifier tube photocathode to the MCP. This caused electron scattering, reduced the signal-to-noise ratio, worsened the image intensifier tube resolution, and limited the dynamic range of its operation (i.e., reduced the limits of the photocathode operating illumination). In addition, the IV generation of image intensifier tubes uses thin-film MCPs with a sharply reduced microchannel diameter of 6 μm instead of the traditional 12 μm. This led to an increase in the resolution to 64–84 lines/mm. Another distinctive feature of the IV generation of image intensifier tubes is the presence of a strobe high-voltage power source, which automatically changes the duty cycle of operation in the pulse mode in accordance with the external illumination [2]. The pulse mode not only expands the dynamic range of the image intensifier tube and, accordingly, the night vision device in a wide range of changes in external illumination, but also suppresses noise. The degree of its attenuation is proportional to the duty cycle of the high-voltage power source in the pulse mode. Due to this, the signal-to-noise ratio can reach 100 or more.

A specific example of a generation IV NVD is the AN/AVS-9 night vision goggles from ITT (USA) (photo 1) [4]. The NVD is equipped with a generation IV image intensifier. ITT has signed a $43 million contract with the US Navy to supply generation IV NVDs. The main parameters of the AN/AVS-9 device are: field of view angle of 400, weight without head mount of 540 g, with mount – 780 g, supply voltage of 3 V, operating temperature range of -32 – +52 0C [5]. The goggles are intended for piloting helicopters.

Photo 1. Night vision goggles
AN/AVS-9 from ITT , USA

The Joint Institute of Semiconductor Physics of the Siberian Branch of the Russian Academy of Sciences has also developed a fourth-generation IOP technology [6]. Instead of a traditional microchannel plate, it uses a semiconductor dynode that operates “on a shoot-through” beam of electrons from the IOP photocathode. According to the developers, this allows for a high signal-to-noise ratio and significant protection against light interference with an IOP service life of up to 50,000 hours.

Thus, at present there is no clear understanding of what should be considered an EOP and, accordingly, a 4th generation NVD. In this regard, it is necessary to select the most important criterion that allows classifying new types of NVDs from the point of view of their possible classification as 4th generation. It seems that such a criterion is the shift in photocathode sensitivity to the infrared (IR) region of the spectrum.

Currently, traditional GaAs-based photocathodes installed in III-generation image intensifier tubes operate in the 0.4–0.9 μm spectral range (Fig. 2, curve 1). However, in recent years, photocathodes with negative electron affinity have been created based on the InGaAs–InGaAsP structure, operating in the 0.4–1.1 μm spectral range. In particular, Varian Associates (USA) has developed three types of photocathodes: with a direct emitter based on p-type InGaAsP with a quantum yield of 2.7%, which is 10–20 times better than that of the S-1 oxygen-cesium photocathode, based on a p-type InP/InGaAsP hybrid heterojunction with a quantum yield of up to 10%, and based on a p-type InGaP/InGaAs heterojunction with a quantum yield of 20–30%. The greatest success was achieved by the Litton company, which created an image intensifier tube with a photocathode also based on InGaAs, but with a high level of indium doping – up to 55% [7]. Fig. 3 shows the spectral sensitivity curves of the Litton image intensifier tube photocathodes. It is evident from Fig. 3 that with indium doping up to 55%, the spectral sensitivity region of the image intensifier tube photocathode shifts to the spectral region above 1.6 μm. Such a shift in sensitivity to the near IR region of the spectrum makes it possible to observe in NVD the radiation of the most common laser target designators-rangefinders operating at a wavelength of 1.06 μm. But the main advantage is that in the near region of the spectrum the level of natural night illumination is significantly higher, the spectral distribution of which is presented by curve 1 in Fig. 2. Curve 2 of the sensitivity of the photocathode of the third generation image intensifier tube and curve 3 of the sensitivity of the image intensifier tube with a photocathode based on InGaAs are also shown here. The advantage of such an image intensifier is obvious. It is designated as generation III+” (Near IR Gen III). These image intensifiers are installed in the AN/AVS-6 (ANVIS) and AN/PVS-7B night vision goggles from Litton [8, 9], designed for night piloting of helicopters, as well as in the AN/PVS-13 night sight for the assault rifle from the same company (photo 2) [8]. But in the AN/PVS-14 night sight (photo 3) for individual weapons from ITT, an image intensifier of the IV generation is installed [10]. The sight has a magnification of 6X, a field of view of 5.630, a range of 700 m, a weight of 1.9 kg, overall dimensions of 317x107x107 mm, a supply voltage of 3 V, and an operating temperature range of -51 – +52 0C [8].

Fig. 2. Spectral sensitivity curves

Fig. 3. Spectral sensitivity curves of
photocathodes of the image intensifier tube from Litton, USA

Photo 2.
AN/PVS-13 night sight from Litton, USA

Photo 3. Night sight
AN/PVS-14 by ITT, USA

Since the III+ generation image intensifiers are completely interchangeable with the III and II (II+) generation image intensifiers, the weight and dimensions of the new generation NVDs do not differ from the III generation NVDs.

However, the most pressing issue is the creation of a generation of NVDs with a working spectral region shifted to the range of 1.4-1.8 µm. The creation of an image intensifier for such NVDs is provided by the OMNIBUS VI program [2]. Let us consider what advantages this spectral region offers for NVDs. For this, we will use the results of work [11].

The average value of natural night illumination on a moonless night for the spectral range of 0.4 – 0.9 μm (photocathode of EOP II, II+, III generations) reaches (1.5 – 3)x10-9 W/cm2μm, and in the spectral range of 1.4 – 1.8 μm – (1.5 – 2)x10-7 W/cm2μm, i.e. two orders of magnitude higher. In addition, the transparency of the atmosphere improves: with a meteorological visibility of 10 km, the transmittance of the atmosphere of 1 km at a wavelength of 600 nm is 0.72, and in the center of the spectral range of 1.4 1.8 μm – 0.93. At the same time, the brightness of the atmospheric haze decreases by more than an order of magnitude in the spectral range of 1.4 – 1.8 μm compared to the visible range of the spectrum. The contrast of the observed object with the background in this spectral region is more stable and 1.4 – 1.5 times higher than in the spectral region of 0.4 – 0.9 μm. In addition, if in this spectral region the illumination at night varies from 10-5 to 2.5×10-9 W/cm2, then in the region of 1.4 – 1.8 μm it varies from 1.6×10-4 to (3 – 4)x10-7 W/cm2 under the same illumination conditions, i.e. by almost two orders of magnitude. The percentage of illumination provision throughout the year for natural night illumination within 5×10-3 – 5×10-4 lux for the spectral region of 1.4 – 1.8 μm is also almost 2 times higher than for 0.4 – 0.9 μm [11].

In the spectrum range of 1.4 – 1.8 µm, it is possible to work to a certain extent in some smoke and dust, and also to visualize the radiation of modern laser target designators-rangefinders operating at a wavelength of 1.55 µm and 1.7 µm.

The use of NVDs operating in the 1.4–2.0 μm spectrum range is very effective for unmasking objects. Fig. 4 shows the reflectivity curves of a former USSR soldier’s uniform (curve 1), a US soldier’s uniform (curve 2) and natural vegetation (curve 3) [12]. It is evident from Fig. 4 that in the 1.4–2.0 μm spectrum range, the difference in the reflectivity of the uniform allows not only to detect a soldier against a green background, but also to distinguish one’s own from an enemy.

Fig. 4. Reflectivity curves of soldiers’ uniforms

It is known that camouflage allows various objects to be masked against the background of the surrounding space. However, camouflage developed for the visible spectrum may be ineffective for the spectrum range of 1.4 – 1.8 µm. For this range, the camouflage pattern disappears, and the silhouette of the camouflaged object is revealed [12].

In the 1.4 – 1.8 µm spectral range, you can see in fog, detect traces of ice on the roofs of airplanes at airports. This is similar to “black” ice on roads. It cannot be seen in the visible spectrum, but can be seen in the 1.4 – 1.8 µm spectrum. In this range, you can detect earlier paintings in paintings, hidden under a layer of oil paint. This is achieved due to the fact that many pigments of oil paints, which color light in the visible region of the spectrum, are transparent in the region of 1–2 μm [12].

Let us now consider the ways of practical implementation of new generation NVGs operating in the spectrum region of 1.4–1.8 μm.

One ​​of the ways, as already mentioned, is the creation of an image intensifier with a photocathode based on InGaAs with a high level of indium doping (see Fig. 3).

Another way is to create an image intensifier with a photocathode based on Schottky barriers, the so-called TER photocathode (TER – Transferred Electron Photocathode) [11]. Fig. 5 shows the spectral sensitivity curve of the TEP photocathode (curve 1) in comparison with the sensitivity curve of a conventional photocathode of a third-generation image intensifier [13]. Taking into account the lower sensitivity of the TER photocathode, it seems advisable to use it in a TV camera based on a CCD with electron bombardment [13]. Fig. 6 shows the design of such a TV camera, where 1 is a TER photocathode; 2 is an electron flow; 3 is an evacuated volume; 4 is a CCD matrix; 5 is a video amplifier; 6 is a liquid crystal (LCD) TV monitor. A TV camera with a 2/3-inch format, 768×244 pixels, and a frame rate of 60 Hz has been developed. The maximum resolution of the TV camera is 45 lines/mm. When the TV camera operates for 12,000 hours, the photocathode sensitivity drops by 50%. The TV camera has a strobe mode. This allows it to be used together with a pulse laser illuminator generating at a wavelength of 1.54 µm as an active-pulse TV system. The front and tail times of the strobe pulse do not exceed 50 ns. At a voltage of 2 kV, the camera gain is over 150. The TV camera does not have a microchannel plate or fiber-optic parts that reduce image quality, which are common for third-generation image intensifiers. The dimensions of the TV camera do not exceed W50x15 mm. The video signal from the CCD matrix output (4) is amplified in the video amplifier (5) and fed to the LCD TV monitor (6), on the screen of which a TV image is created.

Fig. 5. Spectral sensitivity curves

Fig. 6. Schematic diagram of a TV camera based on a TEP photocathode

The next version of the TV system operating in the spectral range of 1 – 1.8 µm is a TV camera based on an IR vidicon. In particular, the Hamamatsu company (Japan) has developed a compact TV camera R5509 [14]. Its spectral sensitivity is shown in Fig. 7 (curve 1). Also shown here is curve 2 of the spectral sensitivity of a similar TV camera (model 7869) from NPO Electron [15]. The TV camera is based on an IR vidicon with a target based on PbO/PbS (or Pb-O-S) with a resolution of 600 TV lines with a target working area size of 9.5 x 12.7 mm. In the USA, the Optical Systems Inc. company has developed a TV camera (model Find-R-Scope 85400/95345) (photo 4) operating in the spectral range up to 2.2 µm [16]. The TV camera weighs 1.587 kg, has dimensions of 305x89x114 mm, a resolution of 550 TV lines, and a current consumption of 1.3 A at a supply voltage of 24 V [14]. In addition to effective observation at low atmospheric transparency, the TV camera allows visualization of laser rangefinder radiation at wavelengths of 1.06, 1.55, 1.7, and 2.1 μm. This is not ensured by the other means mentioned above. Curves 3 and 4 of the spectral sensitivity of the TV camera are shown in Fig. 7. However, all these TV cameras provide operation only on a moonlit night and at twilight. Their common drawback is the presence of an evacuated volume.

Fig. 7. Spectral sensitivity
of a TV camera based on IR vidicon

Photo 4. TV camera Find-R-Scope 85400/95345
Optical Systems Inc., USA

In this regard, the greatest interest is presented by a TV camera using a matrix photodetector based on InGaAs with the number of elements 640×480 [17]. The company Emerging IR Technology (USA) created such a TV camera (model SU320-1,7RT) with the power consumption less than 0,1 W. The TV camera operates in the spectral region of 0,9 – 1,7 or 0,9 – 2,0 μm [12]. The dimensions of the camera do not exceed 158x103x103 mm. Its detection capacity is more than 1012 W-1Hz1/2cm. However, to ensure the advantage of the TV camera in comparison with the third generation image intensifier, its sensitivity should be more than 1014 W-1Hz1/2cm. To achieve such sensitivity, thermoelectric cooling to a temperature of about 230 — 250 K is necessary. However, work is currently actively continuing to reduce the level of dark current and increase the sensitivity of the TV camera. The dynamics of this work allows us to expect that the specified sensitivity will be achieved in the near future without the need for cooling. The spectral sensitivity curve of the TV camera is shown in Fig. 8, and its appearance is in Photo 5. Since the pixel size of the matrix photodetector is 25×25 μm, and in industrial TV cameras on CCD matrices this size reaches 7×7 μm [17], work is underway in the USA to create hyperspectral TV detectors. In this case, a combination of an IR matrix and a visible range matrix is ​​assumed in one chip [17]. Such systems operate in various spectral ranges, by varying which it is possible to select various objects by spectrum.

Fig. 8. Spectral sensitivity curve of a TV camera
using an InGaAs-based matrix photodetector

Photo 5. TV camera SU320-1.7RT
from Emerging IR Technology, USA

As is known, thermal imaging devices can be characterized by such a parameter as NETD. Modern thermal imaging devices based on HgCdTe and InSb focal matrices have NETD values ​​of about 0.01 0C [12]. However, there are a number of applications where significantly lower sensitivity is required, but small weight, dimensions and lower cost are more important. Such applications include observation of heated objects, both on the battlefield and in industrial conditions. For InGaAs-based cameras in the spectrum range of 0.9 – 1.7 μm, we have NETD of about 1 0C [10]. This is sufficient for observation of many heat-emitting objects. An example of such a camera is the Merlin-NEAR model (photo 6) from Indigo Systems (USA), operating in the spectrum range of 0.9 – 1.68 μm. It has an element size of 30×30 µm, a field of view of 22×160 or 11×80, a weight of 1.6 kg, and dimensions of 102x114x203 mm. The camera can measure temperatures in the range from 250 to 2000 0C [12]. The cost of InGaAs-based cameras does not exceed $10–15 thousand, and their price is expected to decrease to $5 thousand. At the same time, operating such cameras is no more difficult than using a household video camera.

Photo 6. Merlin-NEAR camera by Indigo Systems, USA

Such cameras use a TV monitor of the same type as the electronic viewfinder in consumer video cameras. However, significant progress has been made in the field of creating liquid crystal TV monitors in recent years. Therefore, the ideology of such cameras allows for the creation of a completely solid-state image converter (Fig. 9) containing a photodetector matrix (1), an electronic processing unit (2) and an LCD indicator (3). The latter can be a model of a miniature LCD indicator ProCam 1 from Rockwell Science Center (USA) with a pixel count of 1936×1088, with a power consumption of no more than 0.18 W when powered from = 1.5 V and a dynamic range of over 63 dB [18]. Similar parameters are also found in military-grade liquid crystal TV indicators [19]. The electronic processing unit ensures synchronous control of the photodetector matrices and the LCD indicator, as well as image processing in real time.

Fig. 9

Thus, the prospects for creating new generations of NVDs are associated with the use of fully solid-state image converters operating in the spectrum range of 0.9 – 2 μm, providing observation in a wide range of external conditions and new functions that are absent in NVDs based on image intensifier tubes.

Literature

1. Salikov V.L. Night Vision Devices: History of Generations//Special Equipment, 2000, No. 2, pp. 40–48.
2. Gourley S. See in the Night. Jane’s Defense Weekly, 2001, Vol. 21, No. 3, pp. 20–27.
3. Proceedings of the International Conference Night Vision, Great Britain, London, March 3–4, 1999
4. ITT Boosts Night For U.S. Navy Aviators. Defense News, 2001, March 19, p. 21.
5. AN/AVS-9 (F4949) Aviators Night Vision Imaging System. Brochure by ITT Defense and Electronics, USA, 2001
6. New generation image intensifier tube (IIT). 4th International Exhibition Science. Scientific Instruments-2000”, October 16-20, 2000, Russian Academy of Sciences. Developments of the Siberian Branch. Catalogue, Moscow, 2000, p. 25.
7. Estrera J.P., Sinor T.W., Passimoro K.T., Rector M. Development of Extended (1 – 1.3 mkm) Image Intensifiers. SPIE, 1995, Vol. 2551, pp. 135 – 144.
8. Catalogue of night vision devices by Litton, USA, 2000.
9. Salikov V.L. The Era of Night War//Special Equipment, 2000, No. 5, pp. 21 – 32.
10. Gen 3 Pinnacle Meets Gen 4. Armada International, 2001, No. 2, p. 48.
11. Koshavtsev N.F., Koshchavtsev A.N., Fedotova S.F. Analysis of Prospects for the Development of Night Vision Devices. Applied Physics, 1999, issue 3, pp. 66 – 69.
12. Ushakova M.B. InGaAs-Based Rulers and Matrices for the Spectral Range of 0.8 – 2.6 µm and Portable IR Cameras Using Them. Review of foreign literature for 1993 – 2000. ONTI GUP NPO “Orion”, RF, M., 2000.
13. Aebi V.W., Costello K.A., Edgecumbe J.P. Gallium Arsenide Electron Bombarded CCD Technology. SPIE, 1998, Vol. 3434, pp. 37 – 44.
14. Step up to photon counting at 1700 nm…for the very first time! Advertisement by Hamamatsu. Laser Focus World, 2000, Vol. 36, No. 6, p. 39.
15. Korshunova G.I., Stepanov R.M., Naumov Yu.V., Andreeva K.P. Television transmitting device for the spectral range of 0.4 – 2 µm. Applied Physics, 1999, Issue 3, pp. 96 –101.
16. Infrared Viewer Camera Find-R-Scope. Model No. 85400/85345. FJS Optical Systems Inc. prospectus, USA, 1999.
17. Volkov V.G. Small-sized television systems. Review of domestic and foreign press materials for 1980-2000, No. 5591, Moscow, NTC Informtekhnika, 2000.
18. CMOS sensor for industrial video cameras. Laser Focus World, 2001, Vol. 37, No. 6, p. 243.
19. Hercovitz D. A Sampling of Military Displays. Journal of Electronic Defense, 2001, Vol. 24, No. 1, pp. 61 – 65.