Comparison of empirical and predicted ultraviolet aircraft signatures

Abstract

In light of the potential threat to aircraft from missiles using Ultra Violet (UV) wavebands, it is important to understand the signature of an aircraft and how this can be predicted. This study compares empirical UV signature data to modelled data from CAMouflage Electro-Optical SIMulation (CAMEOSIM) to determine how well the contrast between the object and the background can be predicted using local knowledge of the atmosphere. CAMEOSIM uses the standard MODerate resolution atmospheric TRANsmission (MODTRAN) model to estimate the radiative transfer through the atmosphere. Both MODTRAN and CAMEOSIM are well validated in visible and infrared wavebands and MODTRAN can accurately predict UV radiative transfer. Unfortunately, the work so far has concentrated on bulk transfer to describe the sky background in the UV where the aircraft scene is typically a negative contrast 'hole' in a positive sky background. Importantly, path-to-path scattering is a key consideration in this scene since it is this that will tend to blur the edges of an object and reduce the contrast associated with it. A developed understanding of the limitations is required. It was determined that prediction was possible up to ranges of 5km. The local visibility (in km) was required for this prediction.

Dynamic Optical Encryption Fueled via Tunable Mechanical Composite Micrograting Systems

“…Therefore, as shown in Figure 2A and Figure S3 (Supporting Information), a general method for calculating the diffraction point position of oblique incidence is proposed based on the conical diffraction theory. [61] To elucidate the multiple conical grating diffraction model, the first step involves establishing the spatial configuration of the encryption system, which comprises two gratings and a record screen. The direction of incident light perpendicular to two grating planes is defined as Y axis, and the horizontal plane where the record screen is located is XOZ plane.…”

Endmember Learning with K-Means through SCD Model in Hyperspectral Scene Reconstructions

“…This, in turn, will require knowledge of the material property in the scene, such as the types of plants and their optical characteristics, to be known precisely as the input of HSI simulator. In many cases, the detailed material properties of the scene are not known and typically only the broadband RGB, or the multispectral image (MSI) of the scene, is available [4,5]. One approach to solving the problem is to reconstruct the materials of the scene through a learned dictionary to deduce the endmember (EM) characteristics of the scene through an unmixing algorithm [6,7], Equation (1).…”