Hyperspectral Imaging for Real-Time Unmanned Aerial Vehicle Maritime Target Detection

Abstract

This work address hyperspectral imaging systems use for maritime target detection using unmanned aerial vehicles. Specifically, by working in the creation of a hyperspectral real-time data processing system pipeline. We develop a boresight calibration method that allows to calibrate the position of the navigation sensor related to the camera imaging sensor, and improve substantially the accuracy of the target geo-reference. We also develop an unsupervised method for segmenting targets (boats) from their dominant background in real-time. We evaluated the performance of our proposed system for target detection in real-time with UAV flight data and present detection results comparing favorably our approach against other state-ofthe-art method.

Defending Airports from UAS: A Survey on Cyber-Attacks and Counter-Drone Sensing Technologies

“…Gokcce et al [ 83 ] employed vision-based drone detection and distance estimation using traditional features such as histogram of gradients (HOG). Researchers have also achieved detection and object classification by using hyperspectral images [ 84 , 85 ]. The methods can accurately locate and identify drones.…”

Hyperspectral Imaging in Environmental Monitoring: A Review of Recent Developments and Technological Advances in Compact Field Deployable Systems

“…Push broom sensors have been traditionally used for large airborne imaging applications and have recently been successfully miniaturized for use within UAV (unmanned aerial vehicle) systems [10,17,18]. This push broom measurement approach is favored due to its high spatial and spectral resolution [19], however, this image acquisition method, whereby a line of spectral information per exposure is recorded [10,20], can cause difficulties in post-processing [10]. Similarly, whiskbroom sensors, which image a single pixel or spatial location at a time [21,22], using a rotating mirror to sweep a scan line perpendicular to the direction of the sensor platform’s movement [21,22,23], are affected by the same issues [21].…”

“…The literature highlights that although there can be significant variation caused by slit width, lens focal length, and integration time [10], Push Broom sensors, at present, offer a better combination of spatial and spectral resolution. Push Broom sensors are typically more stable than Whiskbroom sensors due to the line-by-line image acquisition process, therefore, confining potential data misalignments to between lines rather than between individual pixels [19]. Furthermore, they often have a significantly greater spectral resolution, for example Jaud et al [26], reports a spectral resolution of 1.85 nm for their Push Broom device.…”

“…Whilst this does not particularly effect traditional remote sensing and ground-based imaging methods, it can become problematic when designing effective UAV integrated payloads [10,21]. Each of these sensor designs has its advantages, depending on the parameters of the proposed application, however, the push broom design has been the most popular, particularly within the field of light-weight UAV image acquisition [19]. Whilst these sensor implementations can involve distortions within the acquired data, they currently outperform full-frame image capturing approaches as the latter systems currently require a compromise between spatial coverage, spatial resolution and spectral resolution [10,26].…”

A Hierarchical Decision-Making Method with a Fuzzy Ant Colony Algorithm for Mission Planning of Multiple UAVs

“…The unmanned aerial vehicle is a new type of combat platform with independent flight capability and independent execution capability, which received huge interest and unprecedented attention worldwide [1]. UAVs exhibit outstanding performance and potential applications, including reconnaissance [2], source seeking [3], target detection, and multiple target attack [4,5]. With the rapid development of UAV technology, more and more UAVs will be used in future battlefields.…”