Video Clarification Better Than EverA NEW TECHNIQUE THAT MAKES EXISTING video images clearer and steadier than anything in current use has several uses in the law enforcement, medical and meteorology fields. However, it could be most cost-effective for the consumer home video market. Two scientists at NASA's Marshall Space Flight Center—Dr. David Hathaway of the Space Sciences Laboratory and Paul Meyer of the Global Hydrology and Climate Center (GHCC) in the computer laboratory—have developed the Video Image Stabilization and Registration (VISAR) technique, a computer algorithm that corrects existing images for zoom, tilt and jitter. Reviews by the Los Alamos National Laboratory concluded that VISAR was unsurpassed in its clarification of distorted video images.
"With VISAR," said Hathaway, "a sequence of video images won't move around, zoom in and out, or rotate." VISAR's corrective power depends on how many video frames are available to be blended together, but generally VISAR can correct image jitter to about one-tenth of a pixel, a tiny square of color that makes up an image. It can correct magnification and zoom to 0.1 percent and angles to within 0.03 degrees. When the FBI's Southeast Bomb Task Force asked if anyone at Marshall Space Flight Center could help improve the clarity of poor quality video from the 1996 Atlanta Olympic Games bombing, both Hathaway and Meyer believed they might be able to help. Two years of trial and error resulted in the ability to stabilize, sharpen and brighten the images. The process also took them from using a $30,000 "QuBit" video-capturing device to using devices as low as $200, and a change in Windows version software reduced processing speed from five minutes to 15 seconds.
"Telescopes are always shaky," said Hathaway, a solar physicist who uses video stabilization techniques to enhance pictures of the Sun. Measurements of the Sun's position or features on the solar surface can be affected by this jittery affect. Meyer, a meteorologist and computer scientist working in the GHCC, processes weather satellite images. The key to clarifying a video sequence is to stabilize the image, according to Hathaway. VISAR allows you to combine several video images together, and noise can be averaged out among the frames. The more images you can combine, the greater the corrective power of VISAR. In the past, video stabilization has been limited to registering horizontal and vertical image movements. These methods do not account for rotational or zooming effects in video data sequences, and they are sensitive to the effects of parallax when items in the background and foreground move at different rates and/or different directions. VISAR can correct images when of all of these adverse effects are present. Current techniques do not take into account how the clarity problem occurred in the first place. Today's techniques are unable to combine images and thus only work on a frame of video at a time. The more a camera operator zooms in, an image becomes larger and more spread out. Because today's techniques can only sharpen the edges of an image, noise and distortion in the image increase. As mentioned above, computer and video images are made up of tiny squares of color, called pixels. By registering VISAR on an object in the image, the pixels from several video frames can be lined up together. The result is a steadier video. VISAR surpasses existing image-correction technology, which cannot compensate for the effects of zoom or tilt, but the VISAR algorithm did. By steadying and reducing the noise in the FBI video images, Hathaway and Meyer brought out a wealth of information, revealing new, previously obscured details. In terms of law enforcement use, the police often use video to identify suspects by recognizing faces in a crowd or repeat crime scene visitors, to investigate a crime scene or to spot identifying characteristics. VISAR could be used to steady images of car chases shot from inside a moving police car, enabling the police to focus on a license plate number or an image of the driver's face reflected in the rear-view mirror. In medical imaging, VISAR could help clarify ultrasound images, infamous for their grainy, blurred quality. More importantly, with VISAR, doctors could make better medical diagnoses, and medical students would train better with steadier, clearer images. Applied to meteorology, VISAR could track cloud formations and storms and be used to determine any changes in the images of a hurricane's eye. Determining a tornado's wind speed may be possible using VISAR to steady a home video camera image to track objects whirling on the outside of the tornado. The home consumer could benefit the most. Although many consumer camcorder devices currently have built-in anti-jitter devices, no devices are available to fix zoom and tilt problems that occur during videotaping. "VISAR can be used to correct these mistakes afterward," Hathaway said. VISAR is currently covered under a provisional patent, and it will soon be available for licensing. The pair hopes to develop real-time stabilization in the future to actually correct footage as it is being videotaped. For more information, contact David Harris at Marshall Space Flight
Center. |
|
