How do you think the new GigE standards will influence the machine vision industry?
Respond or ask your question now!
With seafood exports totaling $2 billion in 20061 and accounting for half of Alaska's total export value, the fishing industry is Alaska's largest employer and one of the State's main economic resources. The Alaska Department of Fish and Game (ADF&G) is the state agency primarily responsible for conducting research on and managing these resources.
ADF&G began experimenting with underwater video technology for scallop stock assessment in 1999. A prototype system consisting of a towed sled that held a camcorder in a pressure housing and used battery-powered lights to image the seafloor was deployed for several years beginning in 2000.
This system proved to have a number of limitations, including inefficiency due to batteries that needed frequent recharging and insufficient image resolution from the camcorder, which did not allow consistent detection of small objects, such as juvenile scallops, on the seafloor. The search for a more efficient, higher resolution system led ADF&G to the work of the HabCam project from Wood's Hole, Mass., a collaboration between the fishing industry and Wood's Hole Oceanographic Institution scientists that led to the first use of machine vision cameras for fisheries seafloor imaging in 20042.
HabCam engineers suggested a system3 based on a GigE Vision camera, so ADF&G contacted Prosilica (Burnaby, BC, Canada) in January 2006. Prosilica recommended its GE1380C, a compact 1.4 megapixel color digital camera that incorporates the Sony ICX285 CCD sensor to provide high-sensitivity, low noise, and excellent image quality. The camera runs up to 20 frames per seconds at full resolution and is capable of streaming data at a sustained rate of 125 megabytes per second. Its GigE interface is virtually plug-and-play and allows data transmission distances of up to 100m over conventional Cat5e network cable or several kilometers using fiber-optic cable.
The imaging system developed by ADF&G basically is a Local Area Network (LAN) that connects a bottom-tending camera sled to a research vessel using an armored fiber optic tow cable. The Prosilica GE1380C camera is mounted on the sled in a pressure housing surrounded by four strobe lights in individual housings. The camera is fitted with an 8mm fixed focal length lens and images a 1m by 0.75m area of the bottom from a height of 1.2m. The strobe lighting ensures crisp images while towing at speeds of 5-8.5 km per hour. A telemetry bottle on the sled contains a Gigabit Ethernet switch and a serial device server that allow the camera and other sensors to connect to the network. On the towing vessel, a second GigE switch and personal computers connected to the LAN allow monitoring and control of sensors on the sled as well as viewing and recording of image data.