A multisensor 180° panoramic camera uses a single casing, but contains multiple sensors/lenses (often three). The images from the sensors overlap slightly and are stitched together by the software to create an undistorted panoramic view of 180° (figure 5).
Since these sensors use standard lenses instead of a wide-angle lens, there is no need to dewarp the image, which means less power consumption and less image distortion. However, the panoramic view from a multisensor camera needs to be manually adjusted in the software when first installed. A major benefit of multisensor cameras is that they achieve higher pixel density and better image quality, thus outperforming wide-angle single-sensor cameras. Also noteworthy is the fact that most multisensor cameras - despite the multiple sensors - only require a single installation point, a single network cable and a single software license, thus making it cheaper than installing three conventional cameras.
What’s wrong with my panoramic view? It looks like three different cameras.
A common misunderstanding regarding the panoramic view of a multisensor camera is that it will be seamlessly interlaced. When placing a multisensor camera in a complex scene, e.g. an indoor parking garage (figure 6), the sensors adapt individually, for the best interpretation of the scene. This will result in individual settings for gain, white balance and exposure time, to get an image with as much detail as possible. As this will result in the panoramic view appearing to come from three different cameras, the user-experience might be less than satisfactory. However, from a forensic point of view, this view is perfect, as it has optimal settings for each sensor/scene and provides great detail. If the multisensor camera is placed in a less complex scene, where the lighting is more uniform, then the panoramic view will also be more uniform (figure 7).
Examples of suitable areas: Airports, train stations, city surveillance/traffic intersections, school yards, loading docks, stadiums, hotel lobbies, parking lots.
- High pixel density
- Several sensors in one casing = one installation
- Improved image settings (WDR)
- Larger casing than single sensor cameras