SightLine ApplicationsĀ has announced its latest software release, 2.25, which improves existing functions and adds new features to its onboard video analytics systems.
Drone detection with deep learning classification can identify detected objects as āDroneā or āNot Droneā and can help to reduce false detections, promoting a seamless transition to track initiation. This ability to classify drones can aid counter-Unmanned Aerial Systems (cUAS) applications for defense and government sectors.
In order to be useful, cUAS detection systems must generate low levels of false negatives and false positives, which is difficult achieve. UAS detection elements must be sensitive enough to detect all drones operating within the area of use, but systems that are too sensitive may create an overwhelming number of false positives, rendering the system unusable.
Other new functionality in 2.25 includes:
- Dead Pixel Removal and Non-Uniformity Correction (DPR/NUC)
- Gas plume enhancement as the first step towards autonomous gas leak detection
- IP video decoding
- New OSD fonts
DPR/NUC, similar to encoding and OSD, is often performed on a separate board. According to SightLine, enabling these functions on SightLine hardware reduces system complexity, power consumption, and cost.
This extends SightLineās continual improvements to its existing market-leading capabilities. Improvements have been made to tracking, detection, KLV telemetry, recording, and precision landing.
āSightline Applications is proud to introduce this next software release,ā said Mark Zanmiller, Director of Business Development. āWe will continue to work closely with our customers to provide world-class technical support and improve our powerful suite of onboard software functionality.ā