The current development of increasingly sensitive low-light detector technologies in the VNIR/SWIR regions shows many promises for future night vision applications, including digital image fusion. By combining spectral bands from the reflective and the thermally emissive domains, providing complementary band-specific cues and advantages, it is anticipated that a fused representation will increase situational awareness and target discrimination performance. Performance assessment of image fusion still remains an open problem however, as suitable procedures, models and image quality metrics are still largely missing. A night-time data collection was made on a side-aspect two-hand object identification task over several ranges in a rural/woodland area using a common line-of-sight VNIR/LWIR system. Perception experiments based on an 8-alternative forced choice (8AFC) object ID task were performed, on both the two individual bands as well as several common pixel-based fusion algorithms (including maximum, subtraction and averaging). As image fusion is highly task and scene dependent it is difficult to draw any general conclusions from a single experiment, but for the particular task/scene combination investigated most of the fusion algorithms are shown to perform better than the VNIR channel, albeit most of them fail to perform as well as the LWIR. This is thought to be the result of the VNIR channel being contrast-limited for the particular task/scene being studied and the low dynamic range of the low-light EBCMOS camera used in the fusion setup.
This paper investigates optical camera performance in natural atmospheres with low visibility including fog, mist, haze, and precipitation. Cameras that operated in most optical wavelengths windows were tested: VIS, NIR (active GV), SWIR, MWIR, and LWIR. An eye safe lidar was also included to estimate transmission in the SWIR region. The measurements were made against contrast board targets placed at 0.5, 1, and 2 km from the sensors. The targets were heated during parts of the campaign to create a thermal contrast, which was relevant for the infrared images. For each image, the contrast, PSF, and MTF were calculated with an in-house developed analysis tool. The parameters were plotted against visibility for various weather conditions. Long wavelengths have better transmittance through haze and small particle fog. In rain, snow or dense fog, where the particles are large, the transmittance is rather independent of the wavelength as is well documented elsewhere. In haze a SWIR camera can be expected to perform better than a visual (at least as good) but since the performance of the cameras was partly masked by different updates (and quality) this does not appear quite clear in the measurement data. The active GV-camera, operating in NIR gave the best results during low daylight conditions. However, the daylight background dominated over the laser illumination during day operation. In addition to the cameras, LIDAR measurements were made to investigate how the atmospheric attenuation is estimated using a single ended measurement sensor. The LIDAR data was used to calculate the atmospheric backscatter maximum and integrated backscatter and backscatter slope. These parameters in general correlated well with the visibility readings from the weather station.
The ability to detect optics is important for military surveillance, and allows early threat detection. Present-day optics detection systems are exploiting that focusing optics are retro-reflective. Hence, it is possible to discover threats, including riflescopes, electro-optical sensors, and magnifying optical assemblies used for weapon guidance, by illuminating them with a laser. However, the sensors have performance limitations and do not usually provide range information about the target. This work suggests a scanning optics detection system that uses a linear array of avalanche photo diodes (APD) with high sensitivity providing information about the target range and angular location. An experimental system using four pixels of a 16×1 linear APD array was constructed and tested against reference targets outdoors. The receiver assembly consisted of a micro-lens array, focusing optics, bandpass filter, and pre-amplifier circuit. The system also contained a pulsed NIR-laser, motorized pan-tilt stages for the scanning, and a calibrated scene camera to measure the background signal. It was possible to detect reference targets at over kilometre range while distinguishing the background, using dedicated signal analysis and noise reduction. The suggested scheme definitely benefit in long-range performance compared to similar techniques that use CCD/CMOS-sensors. The drawback using an APD array lies in reduced angular resolution and increased complexity of data acquisition electronics. In addition, the experimental results will be discussed in terms of a performance model, influence from turbulence effects and suggestions for future sensor improvements.