Manufacturing

ProHawk Vision is highly relevant to manufacturing use cases such as predictive maintenance of equipment and infrastructure, inspection and detection of quality defects, and personnel safety (on manufacturing floors as well as perimeter security). ProHawk AI’s original use case was to expose details regardless of lighting, weather or environmental conditions to detect cracks and other damage at a nuclear powerplant infrastructure in a crisis. 

ProHawk’s patented computer vision algorithms restore video streams for the clearest imagery possible.  This is particularly useful in noisy or less-than-desirable environmental conditions, such as poor lighting (darkness, nighttime, low-light, sun glare, backlight, and headlights), weather (rain, fog, snow), particulates in the air (dust, smoke, smog, dirt, sand, steam), or adverse environments including underwater, outer space, internal medicine and tinted windows.  Additionally, ProHawk brings out details missed from motion blur, hidden in shadows or otherwise lost in low-resolution video.

ProHawk Vision software is typically deployed early in the video pipeline—close to the source or camera—to enhance the effectiveness and precision of downstream components like object detection, tracking, analytics etc.  ProHawk is ideal for use cases involving real-time video due to our patented low-latency design and heavy leverage of NVIDIA CUDA parallelism. Restored video streams are made available via a the NVIDIA DeepStream plugin, as a RTSP proxy or through the ProHawk Vision SDK.