RNET Technologies has developed a video compression analysis system named Virtual Object Based Compression (VOBC) that includes many attractive options, e.g., real-time compression, AES encryption, streaming video mode, synchronized audio and video, object tracking, and 3-D stereo function capabilities. The compression performance of this new system is much better than that which can be achieved with MPEG-2 DVD technology and is comparable to that achieved by MPEG-4 AVC standard but at a relatively lower cost of implementation, both in hardware and software.
In addition, the range of features allows VOBC to be deployed in diverse application areas. Superior performance is achieved via the introduction of an innovative compression technology using "virtual objects" (real objects are typically included within virtual objects). Virtual objects are separated from the background and compressed using wavelets in both the space and time domains. The background is compressed separately using JPEG2000 (or any other suitable compression technology). The compression framework relies on extracting virtual objects (VOs) from the source video, compressing the VOs and the background separately, and combining them to generate compressed video. The block diagrams below illustrate this innovative concept.
Virtual Objects (VOs) typically include "Real Objects" and are dynamically extracted from the video. For the purpose of illustration, the VOs extracted during video compression of an "IR Video" and a "Visual Video" inputs are marked by "white rectangles" in the picture shown below.
The Virtual Objects (VOs) can be compressed either in a losses manner or in a lossy manner. We use wavelet transforms both in spatial and temporal directions for compression. Typical compression ratios we get are in the 20 to 30 times range, and the compression ratios increase substantially if the number of VOs is much smaller. The great advantage of our method is that it is implementation, both in hardware and software.
The Virtual Object Based Compression (VOBC) system includes many attractive options, e.g., real-time compression, AES encryption, streaming video mode, synchronized audio and video, object tracking, 3-D stereo function capabilities, and image fusion. The "synchronized audio plus video compression" system will be highly beneficial when the pilot has to provide voice data for the targets. The water-marking system will facilitate "information hiding" and could be highly valuable to the DOD.
VOBC is naturally a very computationally efficient compression algorithm. In addition, VOBC is being implemented onto a Xilinx FPGA. This Xilinx implementation will allow VOBC to create high image compression rates in an embedded system, at very low hardware and power consumption costs.