Fujitsu’s drive-in pictures

Fujitsu Laboratories has developed image compression technology for automotive use to support digital image transmission over an in-vehicle LAN.


The new technology, SmartCODEC, features a compression rate of one third and a compression-decompression time of two to three milliseconds. This enables compression and multiplex transmission of images to inside the vehicle, such as images from multiple automotive cameras located externally on the vehicle. It also reduces the amount of cable required within the vehicle to achieve fast and high-quality image transmission of multiple images.


Fujitsu said there is an increasing need to use multiple video channels in automobiles. This includes video images from multiple surveillance cameras that the driver checks on the monitor, as well as navigation system images and DVDs for passengers watching on rear-seat monitors. This has resulted in demand for a technology capable of multiplex transmission of images on a single in-vehicle LAN.


SmartCODEC is compression technology dedicated for in-vehicle image transmission. Designed specifically for in-vehicle image transmission, it is lightweight and offers minimal delay time and high image quality. In order to lighten the circuitry, the compression ratio was limited to one-third and a simplified differential pulse code modulation (DPCM) scheme was used.


To clearly reproduce camera images and the edges of letters and lines on the navigation system, the modulation level is optimised at the pixel level. By encoding pixels in groups of two, Fujitsu accelerated the processing speed without increasing the basic clock speed, enabling minimal delay time.