Ex) Article Title, Author, Keywords
Current Optics
and Photonics
Ex) Article Title, Author, Keywords
Curr. Opt. Photon. 2023; 7(4): 354-361
Published online August 25, 2023 https://doi.org/10.3807/COPP.2023.7.4.354
Copyright © Optical Society of Korea.
Kwang-Woo Park , Jonghwa Choi, Jian Kang
Corresponding author: *pkw@add.re.kr , ORCID 0000-0003-0354-0275
This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/4.0/) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.
We present the optical design and experimental verification of resolving performance of a 3× long wavelength infrared (LWIR) zoom camera for drones. The effective focal length of the system varies from 24.5 mm at the wide angle position to 75.1 mm at the telephoto position. The design specifications of the system were derived from ground resolved distance (GRD) to recognize 3 m × 6 m target at a distance of 1 km, at the telephoto position. To satisfy the system requirement, the aperture (f-number) of the system is taken as F/1.6 and the final modulation transfer function (MTF) should be higher than 0.1 (10%). The measured MTF in the laboratory was 0.127 (12.7%), exceeds the system requirement. Outdoor targets were used to verify the comprehensive performance of the system. The system resolved 4-bar targets corresponding to the spatial resolution at the distance of 1 km, 1.4 km and 2 km.
Keywords: Airborne camera, Ground resolve distance, Long wavelength infrared, Modulation transfer function, Signal noise ratio
OCIS codes: (040.2480) FLIR, forward-looking infrared; (040.3060) Infrared; (080.3620) Lens system design; (110.3080) Infrared imaging
With the advent of the Fourth Industrial Revolution, a wide variety of drones (unmanned aircraft), which combine information and communications technology (ICT) and airborne technology, are being used for various purposes in both military and civilian fields. The electro-optical tracking system (EOTS) is essential equipment for drones, which requires concentrated technologies in various fields such as electronics, optics, control, and communication. The use of drones is increasing in various tasks in the private sector as well, such as monitoring road conditions and providing driving information, conducting environmental surveys, responding to disasters and emergencies, environmental monitoring, facility inspection and management, and topographic intelligence investigation [1]. The specifications of the EOTS are determined based on the mission and overall performance requirements of the drone to be equipped. Depending on the mission of the drones, it can have in various forms ranging from high-altitude, long-endurance fixed-wing aircraft to low-altitude, small-sized multi-copter types. The electro-optical system mounted on a drone for image acquisition refers to all systems that convert photons into electrons and then convert them into electrical signals for use. Electro-optical/infrared (EO/IR) sensors convert the brightness difference of the image formed by the optical system into an electrical signal using a detector. This system then converts the signal back into a visible image that can be seen by the human eye [2]. The IR sensor used has advantages in observing and acquiring long-range targets. The IR sensor collects infrared energy emitted directly from the object to create an image. The common wavelength bands used are the mid wavelength infrared (MWIR) band with good atmospheric transmissions at 3 µm–5 µm and the long wavelength infrared (LWIR) band at 8 µm–12 µm. The LWIR sensor has less image blooming and sun glint effects, and better dust transmissions compared to the MWIR band [3–5]. Compared to MWIR sensors that use ultra-low temperature cooling systems, LWIR sensors are much cheaper and have the advantage of easy installation and operation of the sensor. As a result, LWIR sensors are widely used in industrial and civilian fields [6–8].
The electro-optical system described in this paper is based on the LWIR sensor. The performance analysis method of an IR camera involves analyzing and quantifying the characteristics of external factors that affect the image, such as the sensor’s own performance, target contrast, solar illumination conditions, atmospheric conditions, etc. To ultimately predict the resolution according to the distance of the sensor. The method of predicting the distance performance of an electro-optical system has various approaches, but the basic one is to quantitatively calculate the signal-to-noise ratio (SNR) and resolutions [9–15]. The SNR is the ratio of the signal difference between the target and the background to the noise in the image. The smaller mean in this value calls for a higher performance sensor. The SNR is affected by external factors of the sensor, such as the sensor’s performance, target, background, atmospheric conditions, etc. Once the specifications of the sensor are determined, its performance is determined by the level of noise in the detector output at the standard blackbody temperature. The spatial resolution of the sensor is the ability to distinguish how fine a detail can be separated when the SNR is sufficiently high. In this paper, the distance performance of an IR optical system was derived by analyzing the ground resolved distance (GRD). The GRD analysis refers to the distance that can be resolved from the actual images [16, 17].
In cases where the contrast of the target itself is sufficiently high and atmospheric effect is low, the GRD is mainly determined by the optical spatial resolution of the system. In the case of long-range target images with low target contrast or significant atmospheric attenuation, the measured GRD values due to the SNR of the image output signal vary significantly in addition to the optical spatial resolution of the system. For example, considering a 3-bar target with different spatial frequencies, a large target with a low spatial frequency can be clearly resolved even with a low SNR. As the spatial frequency increases, the difference between the target and the background becomes increasingly faint, and they become indistinguishable beyond a certain spatial frequency. Even if the target becomes faint, if there are three black bars, it can be counted, and the target that shows the entire length of the bars becomes the resolution in the image. In a 3-bar target, the distance from the center of one adjacent black bar to the center of the other becomes one cycle of the spatial frequency of the target. If the inverse is taken, it becomes spatial frequency in the cy/mm unit. By considering the parameters from the target to the camera, the system SNR defined by the signal difference between the target and background of the detector output (
By obtaining the SNR as a function of spatial frequency in this method, the maximum spatial frequency that satisfies the set SNR threshold value becomes the system’s resolution. The value obtained by converting one cycle of this spatial frequency to a value in the required distance is defined as the ground resolved distance (GRD) [18–20].
In this paper, we derive design specifications for a camera with the LWIR band for drone mounting through GRD analysis.
Based on the specifications derived through GRD analysis, we designed and manufactured the LWIR optical system. After the manufacturing process, experimental verification of the distance performance was conducted using a 4-bar target with a spatial frequency corresponding to the detection range. Finally, through field testing, we validated the successful resolution of the target placed at the detection distance using a developed camera.
The IR camera is mounted on a drone, which acquires signals of the LWIR band emitted from the target and converts them into XGA (1,024 × 768) images. The detection range performance of the IR camera applies Johnson’s criteria. The target must be resolvable within three cycles and must be recognizable at a distance of 1.0 km with a probability of 50%. Image acquisition conditions were 15 km of visibility, 1.0 km of altitude, 3 m × 6 m of target size and 3 K temperature difference of the target and background.
The system SNR and GRD analysis were performed to derive the optimized design parameters of the camera. In the used wavelength bands, the response of the IR detector is almost constant regardless of the wavelength. The main target of interest is a target with a slightly lower or higher temperature than the background temperature, so objects with a significant temperature difference can be excluded from the discussion. When calculating SNR, the total amount of noise at the given reference background temperature can be represented by the noise equivalent temperature difference (NETD). This is expressed in terms of temperature without considering incident radiation (including background radiation energy) and responsivity according to wavelength [21–23]. Under these assumptions, it can be defined as Eq. (2).
where, ∆
System SNR is defined by Eq. (1) MTF(
TABLE 1 System signal-to-noise ratio (SNR) and ground resolved distance (GRD) analysis conditions of the infared (IR) camera
Item | Value |
---|---|
Wavelength (μm) | 8–12 |
Target Size (m) | 3 × 6 |
Visibility (km) | 15 |
Atmospheric Transmittance (km−1) | 0.9303 |
Temperature Difference (Target & Background) (K) | 3 |
Reference SNR | 5 |
The recognition range performance of an IR camera, defined by the GRD values in Eq. (3), stipulates that the target size must be resolved into three cycles.
To resolve a size of 3 m × 6 m, of a 3-bar target at an altitude of 1.0 km and a distance of 1.0 km, the SNR analysis result at the spatial frequency must be greater than the reference SNR. The SNR threshold value of the human eye varies individually and is typically in the range of 2.5 to 5.0. The system SNR must be higher than this threshold value. The reference SNR for the IR camera has been set to 5 based on experience. Figure 1 shows the GRD analysis results according to the optical system aperture (F/#). The ideal F/# of the optical system is approximately 1.4 to 1.8. The infrared optical system in this study, which is mounted on a drone and operated, should be optimized considering limitations in weight and volume. The goal is to acquire high-resolution images by considering the weight of the optical system and the detection range for each F/#. The F/# of the optical system has been defined as 1.6 to satisfy the requirement.
The analysis of the GRD performance was conducted assuming that a blackbody was located in front of the IR camera. The analysis results in Fig. 2 show the MTF due to the optical system, detector, and aircraft vibration. The MTF value meets or exceeds 0.1 at the reference spatial frequency of 3.0 cy/mrad (41.6 cy/mm). The MTF of the IR camera system is derived from the optical MTF (0.16), detector MTF (0.7), and motion MTF (0.97). The optical MTF needs to be at least 0.16, a value that includes the effects of the optical design (0.2), manufacturing (0.9), assembly and alignment (0.9), and environment (0.97) at 3.0 cy/mrad. So, the optical design MTF of this system is required to be 0.2 or higher. The analysis results of the IR camera’s SNR based on Eq. (1) are shown in Fig. 3.
We can determine the spatial frequency at the threshold SNR value. The initial SNR is 14, and the spatial frequency at the required SNR of 5 is 1.4 cy/mrad. GRD is the value obtained by dividing the target distance by this spatial frequency [Eq. (4)]. Where,
The analysis results show that the GRD meets the required value satisfactorily.
Based on the findings from the previous section, the optical system’s F/# has been set to 1.6. Table 2 presents the design specifications of this lens system. Taking into account weight and volume, the optical design specifications have been defined with a lens diameter of 70 mm or less and a total track length of 100 mm or less. Ultimately, the optical system incorporates a 3× zoom function.
TABLE 2 Design specification of the 3× LWIR zoom system
Parameter | Value | |
---|---|---|
Wavelength (μm) | 8–12 | |
F/# (@ Tele) | 1.6 | |
Zoom Ratio | 3 | |
FOV (@ Tele) (°) | 9.3 × 7.0 (±0.3) | |
MTF (@ Nyquist) | Tele | 20% ↑ (@ 0–filed) |
Tele | ||
DST (Distortion) | Wide | ±5% ↓ |
Tele | ±10% ↓ | |
Diameter (mm) | 70 ↓ | |
OAL (Over All Length) (mm) | 100 ↓ | |
RI (Relative Illumination) (%) | 95 ↑ | |
Weight (g) | 170 ↓ |
When the lens is in telephoto position (narrow FOV), the MTF value at the image center has been set to be 20% or higher, and the distortion aberration to be within ±5%. This optical system employs a 3-group, 4-element zoom lens design. During the zooming process, G1 (L1) remains fixed while G2 (L2, L3) moves along the optical axis to change the lens magnification. G3 (L4) compensates for the camera’s image displacement caused by the movement of the G2. It functions as a focusing mechanism to keep the image fixed at the same position, while driving the varifocal zoom lens. The G1, G2 and G3 were arranged in a P-N-P configuration [24–26]. Here, P refers to the positive group and N refers to the negative group. If the G1 has a positive power, obtaining a wide field of view may not be advantageous, but it has the advantage of having a high zoom ratio and being beneficial for a compact size. A symmetric structure was achieved by placing an aperture between G1 and G2 to be effective for chromatic aberration correction. All lenses, L1 through L4, are made of Germanium. Especially in the telephoto position, to achieve maximum performance, the front surface of L1 was designed as an aspherical shape and the rear surface utilized both refraction and diffraction by adding a diffractive optical element (DOE) to the aspherical surface. The DOE surface is of the kinoform type with rotational symmetry. The diffraction order of the DOE is 1, the design wavelength is 10 μm, and the phase coefficient (C1) is –2.9355 × 10–5 mm. Taking into account manufacturability and scattering due to the diffraction pattern, the DOE was set with three diffraction zones (the number of rings is 3), and performance improvement was confirmed. The pitch of the diffraction zone is 2.5 mm, and there are no issues with manufacturability when fabricating using a diamond turning machine.
In the infrared images, a signal difference occurs between the center and edges of the detector surface during operation. The system is designed to perform non uniformity correction (NUC) by positioning the motor-driven shutter at the position where the diameter of the optical path is smallest, either in front of the detector or within the optical system. To perform NUC, the system is designed with a motor-driven shutter placed at the position where the diameter of the optical path is smallest at the front of the detector. The lens design was optimized by adjusting the curvature, refractive index, dispersion constant, and thickness as design variables to meet a zoom ratio of 3×. The optical design results (Figs. 4 and 5) of the infrared optical system are as follows: The maximum diameter of the system is 70 mm, the effective focal length is 75.1 mm – 24.5 mm, the wavelength range is 8 μm – 2 μm, the field of view is 9.4° × 7.0° – 28.2° × 21.3°, the system transmission rate is 0.5 based on a 1,024 × 768 uncooled staring focal plane array (FPA), the pixel size is 12 μm × 12 μm, the designed optical system MTF and distortion at the narrow field of view is 20.9% and +1.5%. The OAL is 95.3 mm. In all position, the RI shows performance of 95% or higher across all positions. Lastly, the lens weighs approximately 162 g (L1: 119 g, L2: 12 g, L3: 11.5 g, L4: 14.4 g, Filter: 3.6 g, Window: 1.3 g).
One of the important criteria for evaluating the performance of imaging system or its components is the MTF value, which is an international standard (ISO Standard 15529) [27].
In this section, we verified whether the optical system met the system MTF value of 0.1 (10%), which was predicted through GRD analysis in the previous Section 2.1. Figure 6 represents the configuration for measuring the MTF of the LWIR zoom camera. The infrared (IR) beam of the half-moon target, emitted through the collimator, is directed onto the zoom camera where it focuses. During the measurement, the temperature difference between the target and background is set to be 20 ℃ and the collimator focal length at 1,500 mm, the obtained MTF value of 0.127 (12.7%) met the desired performance criteria. The MTF measurement result is shown in Fig. 7.
The GRD performance measurement of the designed and manufactured IR camera is conducted through ground simulation tests in a laboratory environment before its installation on a drone for flight testing. Through performance analysis, the IR temperature difference and target size corresponding to the GRD for each distance are calculated based on altitude (1.0 km) for atmospheric effects on targets [28], as shown in Table 3 considering the target is at a long distance, an optical collimator is required to create a parallel beam. A black body, which acted as the IR target source was installed. In this experiment, the effects of sensor motion caused by drone flight were excluded.
TABLE 3 Equivalent condition of flight laboratory environment
Item | Flight Test Environment | Laboratory Environment |
---|---|---|
Temperature Difference (Target & Background) (K) | 3 | 2.79 |
Target Size (GRD) | 0.7 m | 1.06 mm |
Target Source | Target radiation | Black body |
Figure 8 compares the flight test environment and laboratory environment for the imaging resolution test, considering the equivalent factors in Table 3. For instance, if a ground bar target with a GRD of 0.7 m and a temperature difference of 3 K at a distance of 1 km from an IR camera at a designated altitude (1.0 km) is simulated in a laboratory environment, the bar target in the laboratory will have a GRD of 0.7 m and a temperature difference of 2.79 K.
The experimental setup for measuring the optical imaging resolution is shown in Fig. 8. We installed a bar target with a spatial frequency equivalent to a GRD of 0.7 m at a target distance of 1 km was installed on the optical collimator, and images were acquired after setting the temperature difference using a black body [29–31].
The focal length of the collimator is 1,500 mm. The size of the 4-bar target corresponding to John’s criterion is based on 3 cycles resolution, with
Three types of bar targets were installed. We installed three types of bar targets, equivalent to distances of 1.0 km, 1.4 km, and 2.0 km, on the optical collimator to acquire IR images. This is to verify the image resolution according to the distance. The results are shown in Fig. 9 and the target is being well resolved according to the target temperature. We can confirm that the conditions of GRD are sufficient, satisfied at a distance of 1.0 km. It can be confirmed that the 4-bar target can be clearly distinguished at distances of 1.4 km and 2.0 km. Based on the analysis of GRD performance in the previous section, the optical system specifications were determined, manufactured, and assembled. The resolution of the camera measured in the laboratory environment shows that it meets the required GRD.
We conducted an outdoor field test of the developed camera, verifying its basic operational functions, including control of focus, automatic exposure, and resolution capabilities as assessed in the lab. Real scenes (buildings) and 3-bar targets were selected for evaluation. In this session, we present the results obtained for the real scenes (buildings) only. The buildings selected for evaluation were located approximately 1 km away from the camera. The weather conditions during the test included a visibility range of 15 km, a temperature of 10.5 ℃, and a humidity of 55%. Figure 10 presents the images obtained from the telephoto position and wide angle position. These images demonstrate that effective focus adjustment was accomplished. The exposure time was 8 ms and the average image level was 5,543 gray level (33%). It serves as validation for the camera’s effective image resolution capabilities.
The required performance of a LWIR camera is that it should have a 50% probability of recognizing a target of size 3 m × 6 m at an altitude of 1 km and a target range of 1 km. Through GRD (0.7 m) analysis, we derived a design specification for a camera with the LWIR band suitable for drone mounting. Based on the specifications derived through GRD analysis, we designed 3× zoom lens system with a F/# 1.6, focal length range of 75.1 mm – 24.5 mm, and FOV of 9.4° × 7.0° – 28.2°× 21.3°.
We verified whether the optical system met the 3× zoom lens system MTF value of 0.1 (10%), which was predicted through GRD analysis. The measured MTF value in the laboratory was 0.127 (12.7%) and it satisfied the desired performance criterion of 0.1 (10%).
Additionally, the image resolution performance of the camera manufactured based on the analysis results was measured in a laboratory environment. A bar target with a spatial frequency corresponding to the GRD at a target distance of 1.0 km was installed on an optical collimator. To simulate the distance between the target and the camera, the temperature of the target (ΔT : 2.79 K) corresponding to the distance was set, and then the image was acquired. The measurement results confirmed that the GRD target corresponding to 1 km was well resolved.
Finally, we conducted an outdoor field test of the developed camera. The buildings (target) selected for evaluation were located approximately 1 km away from the camera. The field test results revealed that the developed camera exhibited excellent image resolution capability, enabling the clear distinction of window frames on buildings.
In the future, we plan to analyze various photographic images obtained through actual flight tests to get the reliability of the resolution measurement tests performed in the laboratory.
The authors declare no conflict of interest.
Data underlying the results presented in this paper are not publicly available at the time of publication, but may be obtained from the authors upon reasonable request.
We would like to express our thanks to Siyoun Choi and Kyounghoon Baek of LIG Nex1 for their support in manufacturing and measurement.
Agency for Defense Development Grant Funded by the Korean Government (912855101).
Curr. Opt. Photon. 2023; 7(4): 354-361
Published online August 25, 2023 https://doi.org/10.3807/COPP.2023.7.4.354
Copyright © Optical Society of Korea.
Kwang-Woo Park , Jonghwa Choi, Jian Kang
Agency for Defense Development, Daejeon 34186, Korea
Correspondence to:*pkw@add.re.kr , ORCID 0000-0003-0354-0275
This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/4.0/) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.
We present the optical design and experimental verification of resolving performance of a 3× long wavelength infrared (LWIR) zoom camera for drones. The effective focal length of the system varies from 24.5 mm at the wide angle position to 75.1 mm at the telephoto position. The design specifications of the system were derived from ground resolved distance (GRD) to recognize 3 m × 6 m target at a distance of 1 km, at the telephoto position. To satisfy the system requirement, the aperture (f-number) of the system is taken as F/1.6 and the final modulation transfer function (MTF) should be higher than 0.1 (10%). The measured MTF in the laboratory was 0.127 (12.7%), exceeds the system requirement. Outdoor targets were used to verify the comprehensive performance of the system. The system resolved 4-bar targets corresponding to the spatial resolution at the distance of 1 km, 1.4 km and 2 km.
Keywords: Airborne camera, Ground resolve distance, Long wavelength infrared, Modulation transfer function, Signal noise ratio
With the advent of the Fourth Industrial Revolution, a wide variety of drones (unmanned aircraft), which combine information and communications technology (ICT) and airborne technology, are being used for various purposes in both military and civilian fields. The electro-optical tracking system (EOTS) is essential equipment for drones, which requires concentrated technologies in various fields such as electronics, optics, control, and communication. The use of drones is increasing in various tasks in the private sector as well, such as monitoring road conditions and providing driving information, conducting environmental surveys, responding to disasters and emergencies, environmental monitoring, facility inspection and management, and topographic intelligence investigation [1]. The specifications of the EOTS are determined based on the mission and overall performance requirements of the drone to be equipped. Depending on the mission of the drones, it can have in various forms ranging from high-altitude, long-endurance fixed-wing aircraft to low-altitude, small-sized multi-copter types. The electro-optical system mounted on a drone for image acquisition refers to all systems that convert photons into electrons and then convert them into electrical signals for use. Electro-optical/infrared (EO/IR) sensors convert the brightness difference of the image formed by the optical system into an electrical signal using a detector. This system then converts the signal back into a visible image that can be seen by the human eye [2]. The IR sensor used has advantages in observing and acquiring long-range targets. The IR sensor collects infrared energy emitted directly from the object to create an image. The common wavelength bands used are the mid wavelength infrared (MWIR) band with good atmospheric transmissions at 3 µm–5 µm and the long wavelength infrared (LWIR) band at 8 µm–12 µm. The LWIR sensor has less image blooming and sun glint effects, and better dust transmissions compared to the MWIR band [3–5]. Compared to MWIR sensors that use ultra-low temperature cooling systems, LWIR sensors are much cheaper and have the advantage of easy installation and operation of the sensor. As a result, LWIR sensors are widely used in industrial and civilian fields [6–8].
The electro-optical system described in this paper is based on the LWIR sensor. The performance analysis method of an IR camera involves analyzing and quantifying the characteristics of external factors that affect the image, such as the sensor’s own performance, target contrast, solar illumination conditions, atmospheric conditions, etc. To ultimately predict the resolution according to the distance of the sensor. The method of predicting the distance performance of an electro-optical system has various approaches, but the basic one is to quantitatively calculate the signal-to-noise ratio (SNR) and resolutions [9–15]. The SNR is the ratio of the signal difference between the target and the background to the noise in the image. The smaller mean in this value calls for a higher performance sensor. The SNR is affected by external factors of the sensor, such as the sensor’s performance, target, background, atmospheric conditions, etc. Once the specifications of the sensor are determined, its performance is determined by the level of noise in the detector output at the standard blackbody temperature. The spatial resolution of the sensor is the ability to distinguish how fine a detail can be separated when the SNR is sufficiently high. In this paper, the distance performance of an IR optical system was derived by analyzing the ground resolved distance (GRD). The GRD analysis refers to the distance that can be resolved from the actual images [16, 17].
In cases where the contrast of the target itself is sufficiently high and atmospheric effect is low, the GRD is mainly determined by the optical spatial resolution of the system. In the case of long-range target images with low target contrast or significant atmospheric attenuation, the measured GRD values due to the SNR of the image output signal vary significantly in addition to the optical spatial resolution of the system. For example, considering a 3-bar target with different spatial frequencies, a large target with a low spatial frequency can be clearly resolved even with a low SNR. As the spatial frequency increases, the difference between the target and the background becomes increasingly faint, and they become indistinguishable beyond a certain spatial frequency. Even if the target becomes faint, if there are three black bars, it can be counted, and the target that shows the entire length of the bars becomes the resolution in the image. In a 3-bar target, the distance from the center of one adjacent black bar to the center of the other becomes one cycle of the spatial frequency of the target. If the inverse is taken, it becomes spatial frequency in the cy/mm unit. By considering the parameters from the target to the camera, the system SNR defined by the signal difference between the target and background of the detector output (
By obtaining the SNR as a function of spatial frequency in this method, the maximum spatial frequency that satisfies the set SNR threshold value becomes the system’s resolution. The value obtained by converting one cycle of this spatial frequency to a value in the required distance is defined as the ground resolved distance (GRD) [18–20].
In this paper, we derive design specifications for a camera with the LWIR band for drone mounting through GRD analysis.
Based on the specifications derived through GRD analysis, we designed and manufactured the LWIR optical system. After the manufacturing process, experimental verification of the distance performance was conducted using a 4-bar target with a spatial frequency corresponding to the detection range. Finally, through field testing, we validated the successful resolution of the target placed at the detection distance using a developed camera.
The IR camera is mounted on a drone, which acquires signals of the LWIR band emitted from the target and converts them into XGA (1,024 × 768) images. The detection range performance of the IR camera applies Johnson’s criteria. The target must be resolvable within three cycles and must be recognizable at a distance of 1.0 km with a probability of 50%. Image acquisition conditions were 15 km of visibility, 1.0 km of altitude, 3 m × 6 m of target size and 3 K temperature difference of the target and background.
The system SNR and GRD analysis were performed to derive the optimized design parameters of the camera. In the used wavelength bands, the response of the IR detector is almost constant regardless of the wavelength. The main target of interest is a target with a slightly lower or higher temperature than the background temperature, so objects with a significant temperature difference can be excluded from the discussion. When calculating SNR, the total amount of noise at the given reference background temperature can be represented by the noise equivalent temperature difference (NETD). This is expressed in terms of temperature without considering incident radiation (including background radiation energy) and responsivity according to wavelength [21–23]. Under these assumptions, it can be defined as Eq. (2).
where, ∆
System SNR is defined by Eq. (1) MTF(
TABLE 1. System signal-to-noise ratio (SNR) and ground resolved distance (GRD) analysis conditions of the infared (IR) camera.
Item | Value |
---|---|
Wavelength (μm) | 8–12 |
Target Size (m) | 3 × 6 |
Visibility (km) | 15 |
Atmospheric Transmittance (km−1) | 0.9303 |
Temperature Difference (Target & Background) (K) | 3 |
Reference SNR | 5 |
The recognition range performance of an IR camera, defined by the GRD values in Eq. (3), stipulates that the target size must be resolved into three cycles.
To resolve a size of 3 m × 6 m, of a 3-bar target at an altitude of 1.0 km and a distance of 1.0 km, the SNR analysis result at the spatial frequency must be greater than the reference SNR. The SNR threshold value of the human eye varies individually and is typically in the range of 2.5 to 5.0. The system SNR must be higher than this threshold value. The reference SNR for the IR camera has been set to 5 based on experience. Figure 1 shows the GRD analysis results according to the optical system aperture (F/#). The ideal F/# of the optical system is approximately 1.4 to 1.8. The infrared optical system in this study, which is mounted on a drone and operated, should be optimized considering limitations in weight and volume. The goal is to acquire high-resolution images by considering the weight of the optical system and the detection range for each F/#. The F/# of the optical system has been defined as 1.6 to satisfy the requirement.
The analysis of the GRD performance was conducted assuming that a blackbody was located in front of the IR camera. The analysis results in Fig. 2 show the MTF due to the optical system, detector, and aircraft vibration. The MTF value meets or exceeds 0.1 at the reference spatial frequency of 3.0 cy/mrad (41.6 cy/mm). The MTF of the IR camera system is derived from the optical MTF (0.16), detector MTF (0.7), and motion MTF (0.97). The optical MTF needs to be at least 0.16, a value that includes the effects of the optical design (0.2), manufacturing (0.9), assembly and alignment (0.9), and environment (0.97) at 3.0 cy/mrad. So, the optical design MTF of this system is required to be 0.2 or higher. The analysis results of the IR camera’s SNR based on Eq. (1) are shown in Fig. 3.
We can determine the spatial frequency at the threshold SNR value. The initial SNR is 14, and the spatial frequency at the required SNR of 5 is 1.4 cy/mrad. GRD is the value obtained by dividing the target distance by this spatial frequency [Eq. (4)]. Where,
The analysis results show that the GRD meets the required value satisfactorily.
Based on the findings from the previous section, the optical system’s F/# has been set to 1.6. Table 2 presents the design specifications of this lens system. Taking into account weight and volume, the optical design specifications have been defined with a lens diameter of 70 mm or less and a total track length of 100 mm or less. Ultimately, the optical system incorporates a 3× zoom function.
TABLE 2. Design specification of the 3× LWIR zoom system.
Parameter | Value | |
---|---|---|
Wavelength (μm) | 8–12 | |
F/# (@ Tele) | 1.6 | |
Zoom Ratio | 3 | |
FOV (@ Tele) (°) | 9.3 × 7.0 (±0.3) | |
MTF (@ Nyquist) | Tele | 20% ↑ (@ 0–filed) |
Tele | ||
DST (Distortion) | Wide | ±5% ↓ |
Tele | ±10% ↓ | |
Diameter (mm) | 70 ↓ | |
OAL (Over All Length) (mm) | 100 ↓ | |
RI (Relative Illumination) (%) | 95 ↑ | |
Weight (g) | 170 ↓ |
When the lens is in telephoto position (narrow FOV), the MTF value at the image center has been set to be 20% or higher, and the distortion aberration to be within ±5%. This optical system employs a 3-group, 4-element zoom lens design. During the zooming process, G1 (L1) remains fixed while G2 (L2, L3) moves along the optical axis to change the lens magnification. G3 (L4) compensates for the camera’s image displacement caused by the movement of the G2. It functions as a focusing mechanism to keep the image fixed at the same position, while driving the varifocal zoom lens. The G1, G2 and G3 were arranged in a P-N-P configuration [24–26]. Here, P refers to the positive group and N refers to the negative group. If the G1 has a positive power, obtaining a wide field of view may not be advantageous, but it has the advantage of having a high zoom ratio and being beneficial for a compact size. A symmetric structure was achieved by placing an aperture between G1 and G2 to be effective for chromatic aberration correction. All lenses, L1 through L4, are made of Germanium. Especially in the telephoto position, to achieve maximum performance, the front surface of L1 was designed as an aspherical shape and the rear surface utilized both refraction and diffraction by adding a diffractive optical element (DOE) to the aspherical surface. The DOE surface is of the kinoform type with rotational symmetry. The diffraction order of the DOE is 1, the design wavelength is 10 μm, and the phase coefficient (C1) is –2.9355 × 10–5 mm. Taking into account manufacturability and scattering due to the diffraction pattern, the DOE was set with three diffraction zones (the number of rings is 3), and performance improvement was confirmed. The pitch of the diffraction zone is 2.5 mm, and there are no issues with manufacturability when fabricating using a diamond turning machine.
In the infrared images, a signal difference occurs between the center and edges of the detector surface during operation. The system is designed to perform non uniformity correction (NUC) by positioning the motor-driven shutter at the position where the diameter of the optical path is smallest, either in front of the detector or within the optical system. To perform NUC, the system is designed with a motor-driven shutter placed at the position where the diameter of the optical path is smallest at the front of the detector. The lens design was optimized by adjusting the curvature, refractive index, dispersion constant, and thickness as design variables to meet a zoom ratio of 3×. The optical design results (Figs. 4 and 5) of the infrared optical system are as follows: The maximum diameter of the system is 70 mm, the effective focal length is 75.1 mm – 24.5 mm, the wavelength range is 8 μm – 2 μm, the field of view is 9.4° × 7.0° – 28.2° × 21.3°, the system transmission rate is 0.5 based on a 1,024 × 768 uncooled staring focal plane array (FPA), the pixel size is 12 μm × 12 μm, the designed optical system MTF and distortion at the narrow field of view is 20.9% and +1.5%. The OAL is 95.3 mm. In all position, the RI shows performance of 95% or higher across all positions. Lastly, the lens weighs approximately 162 g (L1: 119 g, L2: 12 g, L3: 11.5 g, L4: 14.4 g, Filter: 3.6 g, Window: 1.3 g).
One of the important criteria for evaluating the performance of imaging system or its components is the MTF value, which is an international standard (ISO Standard 15529) [27].
In this section, we verified whether the optical system met the system MTF value of 0.1 (10%), which was predicted through GRD analysis in the previous Section 2.1. Figure 6 represents the configuration for measuring the MTF of the LWIR zoom camera. The infrared (IR) beam of the half-moon target, emitted through the collimator, is directed onto the zoom camera where it focuses. During the measurement, the temperature difference between the target and background is set to be 20 ℃ and the collimator focal length at 1,500 mm, the obtained MTF value of 0.127 (12.7%) met the desired performance criteria. The MTF measurement result is shown in Fig. 7.
The GRD performance measurement of the designed and manufactured IR camera is conducted through ground simulation tests in a laboratory environment before its installation on a drone for flight testing. Through performance analysis, the IR temperature difference and target size corresponding to the GRD for each distance are calculated based on altitude (1.0 km) for atmospheric effects on targets [28], as shown in Table 3 considering the target is at a long distance, an optical collimator is required to create a parallel beam. A black body, which acted as the IR target source was installed. In this experiment, the effects of sensor motion caused by drone flight were excluded.
TABLE 3. Equivalent condition of flight laboratory environment.
Item | Flight Test Environment | Laboratory Environment |
---|---|---|
Temperature Difference (Target & Background) (K) | 3 | 2.79 |
Target Size (GRD) | 0.7 m | 1.06 mm |
Target Source | Target radiation | Black body |
Figure 8 compares the flight test environment and laboratory environment for the imaging resolution test, considering the equivalent factors in Table 3. For instance, if a ground bar target with a GRD of 0.7 m and a temperature difference of 3 K at a distance of 1 km from an IR camera at a designated altitude (1.0 km) is simulated in a laboratory environment, the bar target in the laboratory will have a GRD of 0.7 m and a temperature difference of 2.79 K.
The experimental setup for measuring the optical imaging resolution is shown in Fig. 8. We installed a bar target with a spatial frequency equivalent to a GRD of 0.7 m at a target distance of 1 km was installed on the optical collimator, and images were acquired after setting the temperature difference using a black body [29–31].
The focal length of the collimator is 1,500 mm. The size of the 4-bar target corresponding to John’s criterion is based on 3 cycles resolution, with
Three types of bar targets were installed. We installed three types of bar targets, equivalent to distances of 1.0 km, 1.4 km, and 2.0 km, on the optical collimator to acquire IR images. This is to verify the image resolution according to the distance. The results are shown in Fig. 9 and the target is being well resolved according to the target temperature. We can confirm that the conditions of GRD are sufficient, satisfied at a distance of 1.0 km. It can be confirmed that the 4-bar target can be clearly distinguished at distances of 1.4 km and 2.0 km. Based on the analysis of GRD performance in the previous section, the optical system specifications were determined, manufactured, and assembled. The resolution of the camera measured in the laboratory environment shows that it meets the required GRD.
We conducted an outdoor field test of the developed camera, verifying its basic operational functions, including control of focus, automatic exposure, and resolution capabilities as assessed in the lab. Real scenes (buildings) and 3-bar targets were selected for evaluation. In this session, we present the results obtained for the real scenes (buildings) only. The buildings selected for evaluation were located approximately 1 km away from the camera. The weather conditions during the test included a visibility range of 15 km, a temperature of 10.5 ℃, and a humidity of 55%. Figure 10 presents the images obtained from the telephoto position and wide angle position. These images demonstrate that effective focus adjustment was accomplished. The exposure time was 8 ms and the average image level was 5,543 gray level (33%). It serves as validation for the camera’s effective image resolution capabilities.
The required performance of a LWIR camera is that it should have a 50% probability of recognizing a target of size 3 m × 6 m at an altitude of 1 km and a target range of 1 km. Through GRD (0.7 m) analysis, we derived a design specification for a camera with the LWIR band suitable for drone mounting. Based on the specifications derived through GRD analysis, we designed 3× zoom lens system with a F/# 1.6, focal length range of 75.1 mm – 24.5 mm, and FOV of 9.4° × 7.0° – 28.2°× 21.3°.
We verified whether the optical system met the 3× zoom lens system MTF value of 0.1 (10%), which was predicted through GRD analysis. The measured MTF value in the laboratory was 0.127 (12.7%) and it satisfied the desired performance criterion of 0.1 (10%).
Additionally, the image resolution performance of the camera manufactured based on the analysis results was measured in a laboratory environment. A bar target with a spatial frequency corresponding to the GRD at a target distance of 1.0 km was installed on an optical collimator. To simulate the distance between the target and the camera, the temperature of the target (ΔT : 2.79 K) corresponding to the distance was set, and then the image was acquired. The measurement results confirmed that the GRD target corresponding to 1 km was well resolved.
Finally, we conducted an outdoor field test of the developed camera. The buildings (target) selected for evaluation were located approximately 1 km away from the camera. The field test results revealed that the developed camera exhibited excellent image resolution capability, enabling the clear distinction of window frames on buildings.
In the future, we plan to analyze various photographic images obtained through actual flight tests to get the reliability of the resolution measurement tests performed in the laboratory.
The authors declare no conflict of interest.
Data underlying the results presented in this paper are not publicly available at the time of publication, but may be obtained from the authors upon reasonable request.
We would like to express our thanks to Siyoun Choi and Kyounghoon Baek of LIG Nex1 for their support in manufacturing and measurement.
Agency for Defense Development Grant Funded by the Korean Government (912855101).
TABLE 1 System signal-to-noise ratio (SNR) and ground resolved distance (GRD) analysis conditions of the infared (IR) camera
Item | Value |
---|---|
Wavelength (μm) | 8–12 |
Target Size (m) | 3 × 6 |
Visibility (km) | 15 |
Atmospheric Transmittance (km−1) | 0.9303 |
Temperature Difference (Target & Background) (K) | 3 |
Reference SNR | 5 |
TABLE 2 Design specification of the 3× LWIR zoom system
Parameter | Value | |
---|---|---|
Wavelength (μm) | 8–12 | |
F/# (@ Tele) | 1.6 | |
Zoom Ratio | 3 | |
FOV (@ Tele) (°) | 9.3 × 7.0 (±0.3) | |
MTF (@ Nyquist) | Tele | 20% ↑ (@ 0–filed) |
Tele | ||
DST (Distortion) | Wide | ±5% ↓ |
Tele | ±10% ↓ | |
Diameter (mm) | 70 ↓ | |
OAL (Over All Length) (mm) | 100 ↓ | |
RI (Relative Illumination) (%) | 95 ↑ | |
Weight (g) | 170 ↓ |
TABLE 3 Equivalent condition of flight laboratory environment
Item | Flight Test Environment | Laboratory Environment |
---|---|---|
Temperature Difference (Target & Background) (K) | 3 | 2.79 |
Target Size (GRD) | 0.7 m | 1.06 mm |
Target Source | Target radiation | Black body |