Get more information on Neptec's ViDAR
Exploiting the benefits of 3D information with Neptec’s ViDAR in ISR applications such as Automatic Target Recognition and Classification, Target Tracking and Automatic Change Detection is advantageous for several reasons:
ViDAR offers actual 3D measurements of the world, providing more tactical and relevant information than traditional 2D data.
ViDAR offers efficient onboard processing and only provides relevant, tactical information, eliminating the need to transmit large volumes of data.
ViDAR’s 3D data is impervious to lighting conditions that often limits the exploitation of 2D data in images, or makes the rendering ambiguous and difficult.
ViDAR’s 3D data is independent of the perspective effect between a camera and objects in the field of view, which also renders the exploitation of 2D images difficult.
ViDAR, combined with Automatic Target Recognition, provides robust information — substantially more than typical 2D data — especially when the target is partially masked.
Neptec’s ViDAR is a state-of-the-art technology that rapidly generates 3D data from a sequence of 2D images. Using standard cameras that are often part of the payload of an aerial platform, ViDAR processes images of a target acquired at two different instants in time to yield a 3D representation of the target. ViDAR’s technique supports a variety of operational scenarios including:
low rate image sequences,
video, and
GPS and/or inertial navigation data for improved speed and quality.
Image sequences can be taken from a fixed-aimed camera flyover from an aerial platform for basic surveillance applications or from a dynamically-aimed camera flyover or circling of selected objects for more robust results.
Comparing acquired 3D data with a knowledge database of 3D target representations, the ViDAR-Automatic Target and Recognition (ATR) and Tracking System presents the operator with the most likely target corresponding to the data at hand. Dubbed 3D ATR, this is a powerful system as it works robustly even when the target is partially hidden. In preliminary field testing, Neptec’s 3D ATR successfully identified a Light Armoured Vehicle (LAVIII) that had been camouflaged by dense foliage.
In addition to target recognition, ViDAR also estimates a target’s pose. This pose provides reference points for the target location and orientation with respect to the sensor platform or a fixed earth reference point enabling the system to track a target in six degrees of freedom (6DOF).
Focused on identifying any type of change to a scene or environment, Automatic Change Detection produces 3D information from 2D imagery. Using 2D data captured by standard cameras found on most aerial platforms at different times from different angles, the system automatically details what has changed between aerial passes. ViDAR-Automatic Change Detection first aligns the current and previous 3D data of a scene using a robust 3D registration algorithm, information compiled from the GPS unit and information from the Inertial Measurement Unit (IMU). 3D information is produced using Neptec’s Shape-from-Motion algorithm. Displayed in a point cloud format, ViDAR-Automatic Change Detection detects clusters of points that exhibit a significant “change” relative to their corresponding points in previous datasets. These changed areas, or areas of interest, are then highlighted for further investigation. In the image above, two different sensor passes of the area have detected changes that have been indicated in red.
VIDAR comes equipped with a unique on-board Hybrid Processor. Typically, Unmanned Aerial Vehicles (UAVs) used for ISR purposes require collected data to be down-linked to a ground station for further processing and analysis by an operator. This is not time efficient, nor is it efficient use of an already limited data down-link bandwidth. ViDAR processes raw sensor data while in-flight and only downlinks pertinent and relevant information, saving the operator time as well as eliminating bandwidth limitations.
ViDAR’s fast, on-board processing capabilities are due to its Hybrid Processor. Designed with both a logic processor and a GPU, the Hybrid Processor reduces high volumes of digital data to smaller, concise pieces of actionable information, facilitating the effective use of sensors with capacities that far exceed the available human or RF bandwidth.
Small, lightweight and low-powered, the Hybrid Processor measures approximately 75 millimeters by 75 millimeters by 20 millimeters, weighs 50 grams and consumes less than 5 Watts of power. With a variety of interfaces, the Hybrid Processor can control cameras, gimbals, input and output video and still imagery, and connect to the host aircraft using any of several interface standards.
