english to korean - open to bidding

withing 3 days Video surveillance systems are currently the most efficient technical tool to ensure public safety by the means of recording facts and controlling the situation at any site. The global increase in crime, terrorist attacks and, most especially, public concern about safety are the factors that promote the development of the global video surveillance market. According to Markets&Markets, the industry’s turnover reached $30.37 billion in 2016. The growth rate is seemingly unabated, as the industry is expected to reach $75.64 billion in turnover by 2022. The problem is something similar - most video surveillance systems are of little use, and therefore inherently inefficient, since it is only possible to record data via videos, and store video archives. They are therefore backward-looking and dumb in that they cannot, and are completely unable to react to a situation when it is actually happening in real time. Computer vision technologies are expensive to develop and require the additional expense of requiring substantial computing resources to run. Only a few very expensive B2B solutions have computer vision and video content analysis implemented as a part of their technology stack. As a consequence, their technologies are still very early-stage compared to the possibilities offered by the vast potential of neural networks. There are currently no products to analyze video streams by means of objects, faces or event recognition that are remotely affordable to consumers.